This is Bugs95-II.gz in view mode; [Up]
From: fpb@uninova.pt (Fernando Pedro Birra) Newsgroups: comp.sys.next.bugs Subject: Re: Parallel printing problems Date: 30 Jun 1995 15:15:32 GMT Organization: UNINOVA Message-ID: <3t14ek$t16@asterix.uninova.pt> References: <3sm9tt$k7u@asterix.uninova.pt> <3sp566$f18@asterix.uninova.pt> <3sv6k9$s7n@larry.rice.edu> In article <3sv6k9$s7n@larry.rice.edu>, steve@xray.rice.edu (Steve Ludtke) wrote: >Fernando Pedro Birra (fpb@uninova.pt) wrote: >: In article <3sm9tt$k7u@asterix.uninova.pt>, >: fpb@uninova.pt (Fernando Pedro Birra) wrote: >: >I have NeXTSTEP 3.3 installed on my Pentium 90. My motherboard is from Intel >: >with on board PCI HD, Serial and Parallel controllers. My parallel port is >: >configured as LPT1, using port 378h and IRQ 7. >: >I have a HP Deskjet 550C and I tried to use ghostscript to print to it. >: >Everything goes well until the stuff goes to /dev/pp0. When I try to write to >: >/dev/pp0 the printer goes busy and it doesn't print at all. >: > >: >I also tried: >: > >: >$ ls > /dev/pp0 >: > >: >which should work on all printers (I even tried with an old Epson 9pin >: >printer). But it also doesn't work. >: > >: >Can someone help me or tell me where should I ask? > >There is a bug in the parallel port driver, but I'm not sure that this is >what's causing your problem. Try using the pp0cat program included with >GhostHPDJ. It was designed as a workaround for the bug. If it still doesn't >work then you've definitely got some sort of hardware problem. > >ftp://ion.rice.edu/pub/next/wordproc/ghostHPDJ0.2.tar.gz >(also @ Peanuts archive) > >------------------------------------------------------------------------- >Steve Ludtke | Physics Dept., Rice Univ. >steve@ion.rice.edu | >stevel@alumni.caltech.edu | "Don't just sit in silence when you >72335,1537 @ compuserve | know what to do." I finally discovered what was the problem: There is some sort of incompatibility when the advanced power management features of the Premiere II motherboard are enabled. This only happens under NeXTSTEP so I think people at NeXT should try to fix this. Thanks for all your help! Fernando Birra fpb@uninova.pt fpb@fct.unl.pt
From: mike@child (Michael Emmel) Newsgroups: comp.sys.next.advocacy,comp.sys.next.bugs,comp.sys.next.sysdmin Subject: Re: "can't open NetInfo parent domain" Date: 30 Jun 1995 15:34:30 GMT Organization: inteleNET Internet Services Message-ID: <3t15i6$dme@vodka.intele.net> References: <Pine.NXT.3.92.950628221918.13194A-100000@Tomobiki-Cho.CAC.Washington.EDU> <3stf1h$5ac@vodka.intele.net> <Pine.NXT.3.92.950629192015.14149A-100000@Tomobiki-Cho.CAC.Washington.EDU> Mark Crispin <mrc@CAC.Washington.EDU> wrote: >On 29 Jun 1995, Michael Emmel wrote: >> Mark Crispin <mrc@CAC.Washington.EDU> wrote: >> >I had been wondering why my /tmp/console.log keeps on getting filled >> >with thousands of these messages, supposedly from syslog. >> >> >It's bootpd. >> <munch> >Would someone translate the above unintelligible gibberish into English? >Near as I can tell, this individual is claiming that the NetInfo database >needs to be initialized from scratch, and that it is somehow my fault. >Funny, I would have thought that any vendor who develops a binary database >accepts the responsibility for that database and of providing proper tools >to administer, maintain, and repair it. But that is the difference >between a real vendor and NeXT. >According to the charming tools provided by NeXT to browse the NetInfo >database, there is nothing wrong with it. Before I spend several hours >getting NetInfo set up again, I'd like some confirmation, preferably from >someone who is literate, that resetting to an initial state and reloading >all the necessary changes will fix the problem instead of wasting my time. >-- Mark -- >DoD #0105, R90/6 pilot, FAX: (206) 685-4045 ICBM: N 47 39'35" W 122 18'39" >Science does not emerge from voting, party politics, or public debate. Well as I said you posted in advocacy and bitched so I bitched back. Now to get out of advocacy mode somewhat : ) Go to comp.sys.next.admin and ask for help!!! Explain you problem exactly and they will help you. My suggestion is to go back to a clean slate I don't think you can "fix" your netifo problems. I ran into the same problem when i got rid of netinfo master machine all the other machines were pinging for the master machine wich wasn't there I think that's your problem therefore you need to reset netinfo. You did not give enough info for me to determine wether I even know the answer or not. Like IS THIS MACHINE A CLIENT OR MASTER???? Yes No Don't know? Like I said I had a similar problem I asked question I fixed it end of story. Heres my answer to your limited description of your problem. If this machine was a net info client and you disconneted the master or took the server off line then thats your problem. If its a client and you took the master off line or buggered the master your machine is looking for the master! Either restore the master or resset the state of the client. If its a master maybe its looking for a client you removed?? Did you have aworking Netinfo setup that you backed up or is your first attempt? Heres how you restore netinfo to a clean state. The only paper manual I have is the 2.1 version its on page 89 of that. 1. Restore Netinfo databses on the master a. Copy /user/template/client/etc/hostconfig to /etc b. Copy the files in /usr/template/cleint/etc/netinfo/local.nidb to /etc/netinfo/local.nidb This restores the local domain c. Remove /etc/netinfo/network.nidb 2. Restore the Netinfo databases on any clone Netinfo servers by loging into each clone sever and following the procedures in step 1 3. Restore the the local information on any Netinfo clients you think are damged, follow substeps (a) and (b) from step 1 Mike
From: Mark Crispin <mrc@CAC.Washington.EDU> Newsgroups: comp.sys.next.advocacy,comp.sys.next.bugs,comp.sys.next.sysdmin Subject: Re: "can't open NetInfo parent domain" Date: Fri, 30 Jun 1995 18:50:42 -0700 Organization: Networks & Distributed Computing Message-ID: <Pine.NXT.3.92.950630183342.15110C-100000@Tomobiki-Cho.CAC.Washington.EDU> References: <Pine.NXT.3.92.950628221918.13194A-100000@Tomobiki-Cho.CAC.Washington.EDU> <3stf1h$5ac@vodka.intele.net> <Pine.NXT.3.92.950629192015.14149A-100000@Tomobiki-Cho.CAC.Washington.EDU> <3t15i6$dme@vodka.intele.net> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII NNTP-Posting-User: dfb In-Reply-To: <3t15i6$dme@vodka.intele.net> On 30 Jun 1995, Michael Emmel wrote: > Well as I said you posted in advocacy and bitched so I > bitched back. NeXT wrote shitty software like NetInfo, and then fucked up the C library so you can't turn it off without losing DNS. I have every right to "bitch". > My suggestion is to go back to a clean slate I don't > think you can "fix" your netifo problems. I ran into the > same problem when i got rid of netinfo master > machine all the other machines were pinging for the > master machine wich wasn't there I think that's your > problem therefore you need to reset netinfo. If you had been paying attention, you would have realized that there is no "NetInfo master". If you had been paying attention, you would have caught the fact that the messages are coming from bootpd. No other daemon has that exact message. > You did not give enough info for me to determine > wether I even know the answer or not. If you don't know the answer, then you should keep your mouth shut and let someone who does speak. > IS THIS MACHINE A CLIENT OR MASTER???? It is neither. It is a single workstation, on a non-NetInfo network. It neither queries nor exports NetInfo information. I would nuke NetInfo altogether and use the flat files, if it weren't for the fact that NeXT buggered the resolver routines in the C library so that DNS resolution doesn't happen unless NetInfo is running. > If this machine was a net info client and you > disconneted the master or took the server off line then > thats your problem. > If its a client and you took the master off line or > buggered the master your machine is looking for the > master! > Either restore the master or resset the state of the > client. If its a master maybe its looking for a client you > removed?? Once again, there is no master or client. Never have been, never will be. Your diagnosis is very unlikely to be correct, since no other daemons have this problem. sendmail would be complaining "can't open parent NetInfo domain"; a message which is similar, but *NOT* exact, to the message which does occur. Most likely, it is a problem in bootpd. > Did you have aworking Netinfo setup that you backed > up or is your first attempt? I've been running NEXTSTEP since 1988 and 0.8. This problem is new, and most likely coincident with 3.2 or 3.3. -- Mark -- DoD #0105, R90/6 pilot, FAX: (206) 685-4045 ICBM: N 47 39'35" W 122 18'39" Science does not emerge from voting, party politics, or public debate.
From: lje0106@sigma.tamu.edu (Louis J. Everett) Newsgroups: comp.sys.next.bugs Subject: Strange (?) Printer problem. Date: 30 Jun 1995 21:50:22 GMT Organization: Texas A&M University, College Station, TX Distribution: world Message-ID: <3t1riu$hjd@news.tamu.edu> I have two next cpus, one (his) is a cube black and white, the other (mine) is a color slab. I have two next printers (his) and (mine) they are identical as far as I can tell. Problem: 1. His printer works fine on his cpu. 2. My printer works fine on his cpu. 3. His printer works fine on my cpu. 4. My printer does not work on my cpu. Any ideas? My printer on my cpu prints a small bit of the page then says "some or all of the pages in your print request could not be printed" I have powered down and back up after changing printers out, but no go. I have repeated the swapping out function multiple times hoping I messed something up but I get the same response each time. I am printing the test page from the printmanager. HELP! -- ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Louis Everett lje0106@sigma.tamu.edu NeXT mail welcome
From: Dino Bagdadi <dino@ex-nihilo.com> Newsgroups: comp.sys.next.bugs Subject: Re: Save Your Computer! Date: 30 Jun 1995 21:11:50 GMT Organization: ex nihilo, inc. Distribution: world Message-ID: <3t1pam$ssu@anshar.shadow.net> References: <3son75$9gj@dub-news-svc-3.compuserve.com> Hey Jerky (aka BriteStar, Inc.), Why don't you post somewhere more appropriate! Hope you have a supernova. -- Dino Bagdadi ex nihilo, inc. dino@ex-nihilo.com (ASCII, NeXTmail and MIME) Public PGP key available via `finger -l dbagdadi@shadow.net'
From: rragner@badger.vm.iastate.edu (Rod Ragner) Newsgroups: comp.sys.next.bugs Subject: Re: Strange (?) Printer problem. Date: 1 Jul 1995 19:36:55 GMT Organization: Iowa State University, Ames, Iowa Distribution: world Message-ID: <3t484n$15p@news.iastate.edu> References: <3t1riu$hjd@news.tamu.edu> In article <3t1riu$hjd@news.tamu.edu> lje0106@sigma.tamu.edu (Louis J. Everett) writes: > I have two next cpus, one (his) is a cube black and white, the other (mine) is > a color slab. > > I have two next printers (his) and (mine) they are identical as far as I can > tell. > > Problem: > 1. His printer works fine on his cpu. > 2. My printer works fine on his cpu. > 3. His printer works fine on my cpu. > 4. My printer does not work on my cpu. > > Any ideas? My printer on my cpu prints a small bit of the page then says "some > or all of the pages in your print request could not be printed" I have powered > down and back up after changing printers out, but no go. I have repeated the > swapping out function multiple times hoping I messed something up but I get the > same response each time. I am printing the test page from the printmanager. > > HELP! > > -- > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^ > Louis Everett > lje0106@sigma.tamu.edu > NeXT mail welcome Louis, You have the same problem which I had about 9 months ago. I bet that you just bought your printer, right? The problem is that there is an incompatibility between old NeXTstation Color motherboards and new (i.e., high serial number) NeXT 400dpi Laser Printers. Fear not, however, this was apparently known in 1992, but not published widely. You can call Bell Atlantic and they will swap you motherboard for a new one which will work for no charge (yes, they actually do a few things for free). Once you have swapped you motherboard, you will be able to use any NeXT 400 dpi Laser Printer, until then you can only use very old ones... -- Rod Ragner Veterinary Diagnostic Laboratory, College of Veterinary Medicine, Iowa State University, Ames, Iowa 50011 Voice: (515) 294-4751, FAX: (515) 294-6961 or 3564, (NeXT Mail accepted) Email: rragner@stallion.vm.iastate.edu or stryder@iastate.edu
From: mike@child (Michael Emmel) Newsgroups: comp.sys.next.advocacy,comp.sys.next.bugs,comp.sys.next.sysdmin Subject: Re: "can't open NetInfo parent domain" Date: 1 Jul 1995 23:33:58 GMT Organization: inteleNET Internet Services Message-ID: <3t4m16$8vt@vodka.intele.net> References: <Pine.NXT.3.92.950628221918.13194A-100000@Tomobiki-Cho.CAC.Washington.EDU> <3stf1h$5ac@vodka.intele.net> <Pine.NXT.3.92.950629192015.14149A-100000@Tomobiki-Cho.CAC.Washington.EDU> <3t15i6$dme@vodka.intele.net> <Pine.NXT.3.92.950630183342.15110C-100000@Tomobiki-Cho.CAC.Washington.EDU> Well this is stupid. I have three Next Station that are at work that I took of a Netinfo network using the procedure I described to you now these machines are at 2.1 so of course if your problem is 3.2 3.3 related I of course had never seen it. Heres your original message: I had been wondering why my /tmp/console.log keeps on getting filled with thousands of these messages, supposedly from syslog. It's bootpd. Of course, there is no hope in getting NeXT, the company that never tests their software, to fix it. NeXT just doesn't care. -- Mark -- DoD #0105, R90/6 pilot, FAX: (206) 685-4045 ICBM: N 47 39'35" W 122 18'39" Science does not emerge from voting, party politics, or public debate. Mark Crispin <mrc@CAC.Washington.EDU> wrote: >On 30 Jun 1995, Michael Emmel wrote: >> Well as I said you posted in advocacy and bitched so I >> bitched back. >NeXT wrote shitty software like NetInfo, and then fucked up the C library >so you can't turn it off without losing DNS. I have every right to >"bitch". >> My suggestion is to go back to a clean slate I don't >> think you can "fix" your netifo problems. I ran into the >> same problem when i got rid of netinfo master >> machine all the other machines were pinging for the >> master machine wich wasn't there I think that's your >> problem therefore you need to reset netinfo. >If you had been paying attention, you would have realized that there is no >"NetInfo master". Why that was so obivous from your original post wasn't it >If you had been paying attention, you would have caught the fact that the >messages are coming from bootpd. No other daemon has that exact message. Nice message what message? >> You did not give enough info for me to determine >> wether I even know the answer or not. >If you don't know the answer, then you should keep your mouth shut and let >someone who does speak. >> IS THIS MACHINE A CLIENT OR MASTER???? >It is neither. It is a single workstation, on a non-NetInfo network. It >neither queries nor exports NetInfo information. Well that nice to know know makes my answer wrong. >Your diagnosis is very unlikely to be correct, since no other daemons have >this problem. sendmail would be complaining "can't open parent NetInfo >domain"; a message which is similar, but *NOT* exact, to the message which >does occur. Which is similar to the message I got looking for a netinfo master so what, I have never the actual error message posted any were in this thread. >Most likely, it is a problem in bootpd. Well to boot off the Net I thought you had to have Netinfo up I certianly don't have a clue how to net boot without Netinfo. If your not netbooting why do you think its bootpd? >> Did you have aworking Netinfo setup that you backed >> up or is your first attempt? >I've been running NEXTSTEP since 1988 and 0.8. This problem is new, and >most likely coincident with 3.2 or 3.3. >-- Mark --< Can't help not that you expected me to anyway. I run 3.3 at home but through PPP so probably can't reproduce your problem anyway but my setup has no or minimal Netinfo and doesn't ping. But agian no ethernet so ? I'm not going to post anymore because : ) 1.) Your right I misunderstood you 2.) Your wrong as far as I'm concerned that it's a bug. Show me some else with the problem. I don't run netinfo and I dont have it. But I don't have a Network card either. 3.) You should have posted to comp.sys.next.sysadmin and asked for help instead of bitching. Thats what pissed me off. And thats the only reason I wrote the first time. Unless the post I followed up on was not your original post. I don't see how you think anybody can help you. I searced up and down all the article that are still on the disk but din't see any others on the thread so I assumed it started there. Now that you say your not running Netinfo I really don't see waht the point in not backing up the files and redoing it is. Shoundn't take you twenty minutes to set up thats what it to me to configure the machines at work as non-netinfo. Its trivail but alas no exported printer do you know how to export a nextprinter without netinfo up? Maybe I missed something seems I'm pretty stupid according to you. Enlighten me. In short your orignal post about bootpd did not make sense to me just made me think your on a Netinfo Network. If your not I sure don't see the relevence because Next does not explain how to Netboot or Export a Printer w/o netinfo (to other Next's) so I assume your a guru and have done some intresting hacking to bypass Netinfo. And can post cryptic bullshit messages and expect somebody whos not a moron (me) to respond good luck. finslly got a sig! and its flame proof. Mike (the moron)
Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.software From: gwolfe@primenet.com Subject: HELP! URGENT! Sender: news@serval.net.wsu.edu (News) Message-ID: <DAz2sH.FBp@serval.net.wsu.edu> Date: Fri, 30 Jun 1995 06:17:04 GMT Organization: Washington State University I am trying to get a SupraFax modem 14.4KB to actually recieve a fax. For lack of a better way of testing I am using the 18002maxtor maxfax document service. I have connected my Supra modem to serial port B on my colorstation. I went into PrintManager to setup my Faxmodem. The only 2 choices in the types list are the HSD and Internafax or somesuch. Which brings up another question ; how do I add types of modems...??? I have it set for 1 ring before it answers and automatic printing disabled. So I call Maxtor and request the catalog and wait. About 1 minute later the phone rings and my modem picks up, connects to thier fax and I am recieving data. The modem is in fax mode because the front panel LED matrix display says FX and at 9600 baud at that. Everything seems to be going fine until the modem gets everything and hangs up (ie. success!) but no my life isn't that easy :-( The console reports this: TIFF Error: : Bad code word at scanline 10 (x 1246). TIFF Error: : Bad code word at scanline 11 (x 126). TIFF Error: : Bad code word at scanline 12 (x 21). TIFF Error: : Bad code word at scanline 14 (x 105). TIFF Error: : Bad code word at scanline 23 (x 65). TIFF Error: : Bad code word at scanline 28 (x 264). TIFF Error: : Bad code word at scanline 29 (x 98). TIFF Error: : Bad code word at scanline 32 (x 589). TIFF Error: : Bad code word at scanline 34 (x 1717). TIFF Error: : Bad code word at scanline 37 (x 99). TIFF Error: : Bad code word at scanline 39 (x 158). TIFF Error: : Bad code word at scanline 40 (x 463). TIFF Error: : Bad code word at scanline 41 (x 115). TIFF Error: : Bad code word at scanline 42 (x 882). TIFF Error: : Bad code word at scanline 44 (x 183). TIFF Error: : Bad code word at scanline 45 (x 1259). TIFF Error: : Bad code word at scanline 48 (x 183). TIFF Error: : Bad code word at scanline 49 (x 1259). TIFF Error: : Bad code word at scanline 52 (x 183). TIFF Error: : Bad code word at scanline 53 (x 1259). TIFF Error: : Bad code word at scanline 56 (x 183). TIFF Error: : Bad code word at scanline 57 (x 1259). TIFF Error: : Bad code word at scanline 60 (x 183). TIFF Error: : Bad code word at scanline 61 (x 1259). TIFF Error: : Bad code word at scanline 64 (x 183). TIFF Error: : Bad code word at scanline 65 (x 1259). TIFF Error: : Bad code word at scanline 68 (x 183). TIFF Error: : Bad code word at scanline 69 (x 1259). TIFF Error: : Bad code word at scanline 72 (x 183). TIFF Error: : Bad code word at scanline 73 (x 1259). TIFF Error: : Bad code word at scanline 76 (x 1230). TIFF Error: : Bad code word at scanline 79 (x 1247). TIFF Error: : Bad code word at scanline 84 (x 156). TIFF Error: : Bad code word at scanline 85 (x 46). TIFF Error: : Bad code word at scanline 86 (x 257). TIFF Error: : Bad code word at scanline 88 (x 664). TIFF Error: : Bad code word at scanline 95 (x 451). TIFF Error: : Bad code word at scanline 97 (x 460). TIFF Error: : Bad code word at scanline 99 (x 315). TIFF Error: : Bad code word at scanline 103 (x 135). TIFF Error: : Bad code word at scanline 104 (x 217). TIFF Error: : Bad code word at scanline 106 (x 223). TIFF Error: : Bad code word at scanline 107 (x 282). TIFF Error: : Bad code word at scanline 114 (x 219). TIFF Error: : Bad code word at scanline 117 (x 48). TIFF Error: : Bad code word at scanline 132 (x 35). TIFF Error: : Bad code word at scanline 135 (x 736). TIFF Error: : Bad code word at scanline 138 (x 582). TIFF Error: : Bad code word at scanline 139 (x 417). TIFF Error: : Bad code word at scanline 142 (x 1599). TIFF Error: : Bad code word at scanline 144 (x 626). TIFF Error: : Bad code word at scanline 147 (x 18). TIFF Error: : Bad code word at scanline 149 (x 1387). TIFF Error: : Bad code word at scanline 160 (x 960). TIFF Error: : Bad code word at scanline 162 (x 74). [a bunch more of the same...shortened to save BW] And when I use the FaxReader app to look at the fax it is just a bunch of lines with no words or any resemblance to anything. How can I get this to work correctly?? PLEASE HELP! What is going on here? Please direct responses to e-mail. Thanks. All help deeply appreciated, Gary Wolfe gwolfe@primenet.com
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 3 Jul 1995 04:15:11 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <3t7qsf$c98@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:Michael_Pizolato@afs.com) Dan Grillo (mailto:dan_grillo@next.com)
Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.software From: rdl@world.std.com (Robert La Ferla) Subject: Re: HELP! URGENT! In-Reply-To: gwolfe@primenet.com's message of Fri, 30 Jun 1995 06:17:04 GMT Message-ID: <RDL.95Jul3064531@world.std.com> Sender: rdl@world.std.com (Robert La Ferla) Organization: The World Public Access UNIX, Brookline, MA References: <DAz2sH.FBp@serval.net.wsu.edu> Date: Mon, 3 Jul 1995 10:45:31 GMT You need to purchase a third party product. Black and White Software sells an excellent one called "NXFax" Send e-mail to info@bandw.com Robert La Ferla Registered NS Consultant Boston, MA + 1 (617) 252-0088 In article <DAz2sH.FBp@serval.net.wsu.edu> gwolfe@primenet.com writes: I am trying to get a SupraFax modem 14.4KB to actually recieve a fax. For lack of a better way of testing I am using the 18002maxtor maxfax document service. I have connected my Supra modem to serial port B on my colorstation. I went into PrintManager to setup my Faxmodem. The only 2 choices in the types list are the HSD and Internafax or somesuch. Which brings up another question ; how do I add types of modems...??? I have it set for 1 ring before it answers and automatic printing disabled. So I call Maxtor and request the catalog and wait. About 1 minute later the phone rings and my modem picks up, connects to thier fax and I am recieving data. The modem is in fax mode because the front panel LED matrix display says FX and at 9600 baud at that. Everything seems to be going fine until the modem gets everything and hangs up (ie. success!) but no my life isn't that easy :-( Gary Wolfe gwolfe@primenet.com
From: fischer@fokus.gmd.de (Robert Fischer) Newsgroups: comp.sys.next.bugs Subject: gdb does'nt know where it is... Date: 3 Jul 1995 15:22:47 GMT Organization: GMD-FOKUS Message-ID: <3t9207$cf7@stern.fokus.gmd.de> Hi, have I missed some bug report concerning the 'gdb'? I have'nt developed since we have upgraded to 3.3 so I ran into the gdb bug just now. In the gdb-Panel of Edit the lines in the first column seem to be upside down. Clicking on the first entry jumps to the code corresponding to the last line. Inspecting objects/variables is sometimes correct, sometimes only at the second click but mostly they are incorrect. I watch this behaviour with NS 3.3 on HP and Moto. Anybody else? Robert. ----------------------------------------------------------------------- Robert Fischer @ GMD-Fokus -------- __o ------- _`\<,_ fischer@fokus.gmd.de ------- (*)/ (*) ## NeXT-Mail welcome ## -----------------------------------------------------------------------
From: mkreis@pws-mk.itr.ch (Matthias Kreis,) Newsgroups: comp.sys.next.bugs Subject: Can't write to the Default Database as root Date: 3 Jul 1995 16:13:19 GMT Organization: ITR (Interkantonales Technikum Rapperswil) Message-ID: <3t94uv$o0d@sky.itr.ch> Mime-Version: 2.6 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Hello I have a problem here with the DefaultDatabase. The tools dwrite and dread sad to me the can't open the database. This happens only, when I logged in as root. NextStep Version 3.3 is running on this computer. Any Ideas to solve this problem? Wrong permissions on the .Next Folder? Thank for the answers. Bye Matthias
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: Can't write to the Default Database as root Date: 3 Jul 1995 21:24:48 GMT Organization: Technical University of Berlin, Germany Message-ID: <3t9n70$3ae@news.cs.tu-berlin.de> References: <3t94uv$o0d@sky.itr.ch> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Matthias Kreis, (mkreis@pws-mk.itr.ch) wrote: : Hello : I have a problem here with the DefaultDatabase. : The tools dwrite and dread sad to me the can't open the database. This : happens only, when I logged in as root. NextStep Version 3.3 is running on : this computer. : Any Ideas to solve this problem? Wrong permissions on the .Next Folder? No, this can't be the reason if you're root. Maybe root's home-directory is invalid, or the database is corrupted. Check if the directory given by 'echo ~root' really exists. Check for ~root/.NeXT/.NeXTdefaults.[LD] Maybe you have to remove .NeXTdefaults.[LD] and reconstruct it from scratch (do a dread -l >rootdefaults first, so you can easily restore them). Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: paul@pth.com (Paul Tognato-Haddad) Newsgroups: comp.sys.next.bugs Subject: Re: Can't write to the Default Database as root Date: 4 Jul 1995 00:09:21 GMT Organization: FishNet Distribution: world Message-ID: <3ta0rh$3nf@wanda.pond.com> References: <3t9n70$3ae@news.cs.tu-berlin.de> Howdy, In article <3t9n70$3ae@news.cs.tu-berlin.de> kiwi@cs.tu-berlin.de (Axel Habermann) writes: > Matthias Kreis, (mkreis@pws-mk.itr.ch) wrote: > : Hello > > : I have a problem here with the DefaultDatabase. > : The tools dwrite and dread sad to me the can't open the database. This > : happens only, when I logged in as root. NextStep Version 3.3 is running on > : this computer. > > : Any Ideas to solve this problem? Wrong permissions on the .Next Folder? Root's home directory gets changed to /root when you modify the account using (L)UserManager. Use NetInfoManager to change it / and everything should be fixed. Or as root try niutil -createprop / /users/root home / niutil -createprop . /users/root home / See ya, -- Paul (NeXTmail preferred) # Paul R. Tognato-Haddad (paul@pth.com) (Home) 610-265-3025 (Fax) 610-265-3026
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs,comp.infosystems.www.misc From: G.C.Th.Wierda@AWT.NL (Gerben Wierda) Subject: CERN httpd 3.0 triggers bug in NEXTSTEP 3.3 Message-ID: <DB7A4p.86@AWT.NL> Sender: news@AWT.NL Organization: Adviesraad voor het Wetenschaps- en Technologiebeleid Mime-Version: 1.0 Date: Tue, 4 Jul 1995 16:36:25 GMT I use CERN httpd 3.0, because it has caching and proxy support, and I need the cache. When I run httpd under NEXTSTEP 3.3 on intel, and I try to restart the system from the login window, the restart process hangs. The culprit is httpd, it seems, because without httpd the syetsm reboots ok. Also, rebooting via shutdown works OK. I guess that rebooting or shutting down via loginwindow.app tries to kill all processes separately, while shutdown kills init with a SIGTERM. The latter seems to work. Why do I think that httpd is to blame? Well, if I leave it out the restart problems are not there. Besides, killing httpd itself with SIGTERM will crash NEXTSTEP. On intel, the big console window appears and the system hangs. Luckily, the system still reacts to ALT-Numlock followed by 'r', so that a clean reboot is possible. Killing httpd with SIGKILL does not trigger the bug. I would like to get in contact with people running CERN httpd 3.0 under NEXTSTEP to see if this can be worked out. Or I'd like to get into contact with people who know how to solve this problem, or what triggers it. Thanks,
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin From: G.C.Th.Wierda@AWT.NL (Gerben Wierda) Subject: Hanging reboot culprit found: CERN httpd 3.0 Message-ID: <DB7A6s.8w@AWT.NL> Sender: news@AWT.NL Organization: Adviesraad voor het Wetenschaps- en Technologiebeleid Mime-Version: 1.0 Date: Tue, 4 Jul 1995 16:37:40 GMT See my separate post on the CERN httpd 3.0 bug, this is just to let intersted people know I found the culprit. --Gerben
From: guichard@parmenide (Eric Guichard) Newsgroups: comp.sys.next.bugs Subject: only root can mount floppy (after installing 3.3) on Elonex PC-4100/1 Date: 4 Jul 1995 18:09:15 GMT Organization: Ecole Normale Superieure Distribution: world Message-ID: <3tc04b$omr@nef.ens.fr> When I select the WorkSpace manager menu "Check disk" I got this message (French localisation): Jul 4 19:51:41 Workspace[205]: Cannot check removable disks: Autorisation refusee I did check the permissions on the dev brw-rw-rw- 1 root 1, 0 Nov 12 1994 fd0a I checked also the suid executables in /usr/filesystems -rwsr-xr-x 1 root 24576 Oct 21 1994 CDROM.util* -rwsr-xr-x 1 root 32768 Oct 21 1994 DOS.util* -rwsr-xr-x 1 root 40960 Oct 21 1994 cdaudio.util* -rwsr-xr-x 1 root 57344 Oct 21 1994 mac.util* Any clues? Eric
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Bug in Librarian Date: 4 Jul 1995 19:58:35 GMT Organization: Technical University of Berlin, Germany Message-ID: <3tc6hb$rga@news.cs.tu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Recently I've found a rather amusing bug in Librarian: If you drag a folder onto Librarian's shelf which contains an icon (.dir.tiff) of a bigger size than icons normally have (it has been scaled down by the Workspace to fit into 48x48), it looks fine first, but if you quit and restart Librarian, the icon gets displayed in full size. I tried this with the icon from the Mux sorce distibution on NS3.3 (Intel & Motorola) Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
Newsgroups: comp.sys.next.bugs From: Fabien_Roy@free.fdn.org Subject: Re: gdb does'nt know where it is... Message-ID: <1995Jul4.215445.10284@free.fdn.org> Sender: news@free.fdn.org Organization: Fabien Roy Consultant, Paris, France References: <3t9207$cf7@stern.fokus.gmd.de> Date: Tue, 4 Jul 1995 21:54:45 GMT fischer@fokus.gmd.de (Robert Fischer) wrote: > Hi, > have I missed some bug report concerning the 'gdb'? I have'nt developed since > we have upgraded to 3.3 so I ran into the gdb bug just now. > In the gdb-Panel of Edit the lines in the first column seem to be upside down. > Clicking on the first entry jumps to the code corresponding to the last line. > Inspecting objects/variables is sometimes correct, sometimes only at the second > click but mostly they are incorrect. > I watch this behaviour with NS 3.3 on HP and Moto. > Anybody else? > Robert. > ----------------------------------------------------------------------- > Robert Fischer @ GMD-Fokus -------- __o > ------- _`\<,_ > fischer@fokus.gmd.de ------- (*)/ (*) ## NeXT-Mail welcome ## > ----------------------------------------------------------------------- Compile with no optimization only -g option (no -O or -O2). On risc machine optimization can change the order of some sequential instructions. Hope that helps. Fabien -- --------------------------------------------------------------------- Fabien_Roy@free.fdn.org (NextMail/MIME accepted) Fabien Roy Consultant NEXTSTEP/OPENSTEP/EOF Consultant, SYBASE DBA 10 rue de la DEFENSE 93100 MONTREUIL, France Tel: 33 1 45 28 32 23 Fax: 33 1 48 55 09 90
Newsgroups: comp.sys.next.bugs Subject: Re: getcwd and posix libs Message-ID: <1995Jul6.103803.45127@yogi.urz.unibas.ch> From: frank@ifi.unibas.ch Date: 6 Jul 95 10:38:03 MET References: <1995Jun26.115903.45046@yogi.urz.unibas.ch> <3sos3n$dm5@papoose.quick.com> jq@phcs.phcs.com (James E. Quick) wrote: > In article <1995Jun26.115903.45046@yogi.urz.unibas.ch>, Robert Frank <frank@marcie> wrote: > >A bug? > > > >I've been trying to compile a posix / standard C program which > >uses getcwd. > > > >Compilation is with -ansi and -posixstrict, linking is with -ansi > >and -posix (tries -posixstrict >too). > > > >Compilation is fine, however, the executable never finds a > >working directory! Getcwd always returns NULL and complains that > >there is no such path, while the old bsd getwd correctly returns > >the working directory. > The most common programming error when using getcwd() is the failure to properly initialize its second argument, length. Is properly set. Tried a constatn number as well as the posix supplied macros. > Make sure that the buffer used by getcwd is at least 1K bytes long, also make sure that this length is passed correctly to getcwd(). It is. > This should solve your problem. Nope. Still get just a null pointer. > If getcwd() is the only reason you are compiling under posix then [...] Definitely not! -Robert -- Institut fuer Informatik tel +41 (0)61 321 99 67 Universitaet Basel fax. +41 (0)61 321 99 15 Robert Frank Mittlere Strasse 142 rfc822: frank@ifi.unibas.ch (NeXT,MIME mail ok) CH-4056 Basel X400: S=frank;OU=ifi;O=unibas;P=switch;A=arcom;C=ch Switzerland
From: fischer@fokus.gmd.de (Robert Fischer) Newsgroups: comp.sys.next.bugs Subject: Re: gdb does'nt know where it is... Date: 6 Jul 1995 12:41:44 GMT Organization: GMD-FOKUS Message-ID: <3tglm8$em7@stern.fokus.gmd.de> References: <1995Jul4.215445.10284@free.fdn.org> Fabien_Roy@free.fdn.org writes > fischer@fokus.gmd.de (Robert Fischer) wrote: > > ... > > have I missed some bug report concerning the 'gdb'? I have'nt developed > > ... > Compile with no optimization only -g option (no -O or -O2). > On risc machine optimization can change the order of some sequential > instructions. > > Hope that helps. > Maybe, but I followed Art Isbells advice and reinstalled from NS 3.2 /usr/lib/GdbClient.bundle and /bin/gdb. That works. Robert. ----------------------------------------------------------------------- Robert Fischer @ GMD-Fokus -------- __o ------- _`\<,_ fischer@fokus.gmd.de ------- (*)/ (*) ## NeXT-Mail welcome ## -----------------------------------------------------------------------
From: fischer@fokus.gmd.de (Robert Fischer) Newsgroups: comp.sys.next.bugs Subject: Re: gdb does'nt know where it is... Date: 6 Jul 1995 12:41:53 GMT Organization: GMD-FOKUS Message-ID: <3tglmh$em8@stern.fokus.gmd.de> References: <1995Jul4.215445.10284@free.fdn.org> Fabien_Roy@free.fdn.org writes > fischer@fokus.gmd.de (Robert Fischer) wrote: > > ... > > have I missed some bug report concerning the 'gdb'? I have'nt developed > > ... > Compile with no optimization only -g option (no -O or -O2). > On risc machine optimization can change the order of some sequential > instructions. > > Hope that helps. > Maybe, but I followed Art Isbells advice and reinstalled from NS 3.2 /usr/lib/GdbClient.bundle and /bin/gdb. That works. Robert. ----------------------------------------------------------------------- Robert Fischer @ GMD-Fokus -------- __o ------- _`\<,_ fischer@fokus.gmd.de ------- (*)/ (*) ## NeXT-Mail welcome ## -----------------------------------------------------------------------
From: fischer@fokus.gmd.de (Robert Fischer) Newsgroups: comp.sys.next.bugs Subject: Re: gdb does'nt know where it is... Date: 6 Jul 1995 12:42:09 GMT Organization: GMD-FOKUS Message-ID: <3tgln1$em9@stern.fokus.gmd.de> References: <1995Jul4.215445.10284@free.fdn.org> Fabien_Roy@free.fdn.org writes > fischer@fokus.gmd.de (Robert Fischer) wrote: > > ... > > have I missed some bug report concerning the 'gdb'? I have'nt developed > > ... > Compile with no optimization only -g option (no -O or -O2). > On risc machine optimization can change the order of some sequential > instructions. > > Hope that helps. > Maybe, but I followed Art Isbells advice and reinstalled from NS 3.2 /usr/lib/GdbClient.bundle and /bin/gdb. That works. Robert. ----------------------------------------------------------------------- Robert Fischer @ GMD-Fokus -------- __o ------- _`\<,_ fischer@fokus.gmd.de ------- (*)/ (*) ## NeXT-Mail welcome ## -----------------------------------------------------------------------
Newsgroups: comp.sys.next.bugs From: gerald@kurt.in-berlin.de (Gerald Erdmann) Subject: Re: Can't write to the Default Database as root Message-ID: <1995Jul6.003051.12754@kurt.in-berlin.de> Sender: news@kurt.in-berlin.de References: <3t94uv$o0d@sky.itr.ch> <3t9n70$3ae@news.cs.tu-berlin.de> Date: Thu, 6 Jul 1995 00:30:51 GMT kiwi@cs.tu-berlin.de (Axel Habermann) wrote: > Matthias Kreis, (mkreis@pws-mk.itr.ch) wrote: > : Hello > : I have a problem here with the DefaultDatabase. > : The tools dwrite and dread sad to me the can't open the database. This > : happens only, when I logged in as root. NextStep Version 3.3 is running on > : this computer. > : Any Ideas to solve this problem? Wrong permissions on the .Next Folder? It's a typical 3.3 UserManager.app problem. Check the home directory of root with NetInfoManager and change it from "/root" to "/". Gerald -- --------------------------------------------------------------------- | GERALD ERDMANN | email: gerald @ kurt.in-berlin.de (NeXTmail welcome) | voice: +49 30 372 43 10 (Germany - Berlin) | crypt: pgp2 public key available |
From: flight@mathi.uni-heidelberg.de (Gregor Hoffleit) Newsgroups: comp.sys.next.bugs Subject: NEXTSTEP and 1.8M disks ;-) Date: 7 Jul 1995 12:04:37 GMT Organization: University of Heidelberg, Germany Message-ID: <3tj7sl$qpj@sun0.urz.uni-heidelberg.de> In the PC world, it gets more and more common to format floppy disks with more than 80 tracks. Even IBM delivers OS/2 Warp on 1.8M formatted disks, and it works. So I formatted a disk with a DOS utility to 1.8M and tried to mount this disk with NEXTSTEP, and - what a surprise - it worked! NEXTSTEP/Intel 3.3 finds 1.66MB free. But now it's not possible to write more than 1.39 MB on the disk. The filebrowser and df say that there are some two hundred KB free, but it's impossible to write any more. Perhaps this could be changed in a next release of the DOS.fs ? Gregor -- | Gregor Hoffleit admin MATHInet / contact HeidelNeXT | | MAIL: Mathematisches Institut PHONE: (49)6221 56-5771 | | INF 288, 69120 Heidelberg / Germany FAX: 56-3812 | | EMAIL: flight@mathi.uni-heidelberg.de (NeXTmail) |
From: ccwf@locke.klab.caltech.edu (Charles Fu) Newsgroups: comp.sys.next.bugs Subject: Re: getcwd and posix libs Date: 7 Jul 1995 18:15:46 GMT Organization: California Institute of Technology, Pasadena Message-ID: <3tjtki$5e9@gap.cco.caltech.edu> References: <1995Jun26.115903.45046@yogi.urz.unibas.ch> In article <1995Jun26.115903.45046@yogi.urz.unibas.ch>, Robert Frank <frank@marcie> wrote: > I've been trying to compile a posix / standard C program which uses getcwd. > ... > Compilation is fine, however, the executable never finds a working > directory! Getcwd always returns NULL and complains that there is no such > path, while the old bsd getwd correctly returns the working directory. Note that getcwd() calls readdir() recursively, so you may be getting bitten by the buggy readdir() function in the POSIX library, about which I have posted previously on multiple occasions. The readdir bug is most likely to affect directories with mount points; so if your getcwd bug happens primarily when the working directory is on a mounted disk and doesn't usually affect working directories on the root disk (and which have no parent directories with mount points), then you are probably experiencing the readdir bug. Again, the simplest workaround for the readdir bug is to grab the sources from somewhere and compile it yourself. If you are using POSIX under NEXTSTEP extensively, you probably want to do this anyways. -ccwf
From: stefanos@Vir.com (Stefanos Kiakis) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs,comp.infosystems.www.misc Subject: Re: CERN httpd 3.0 triggers bug in NEXTSTEP 3.3 Followup-To: comp.sys.next.sysadmin,comp.sys.next.bugs,comp.infosystems.www.misc Date: 7 Jul 1995 20:03:49 -0400 Organization: Communications Vir, Internet Access Montreal. Message-ID: <3tki15$q86@mysite.mydomain> References: <DB7A4p.86@AWT.NL> I'm running the CERN httpd 3.0 under NS3.3 for Intel and have not experienced this problem. My machine is stand alone and is not connected to any network yet. stef Gerben Wierda (G.C.Th.Wierda@AWT.NL) wrote: : I use CERN httpd 3.0, because it has caching and proxy support, and I need the : cache. : When I run httpd under NEXTSTEP 3.3 on intel, and I try to restart the system : from the login window, the restart process hangs. The culprit is httpd, it seems, : because without httpd the syetsm reboots ok. Also, rebooting via shutdown : works OK. I guess that rebooting or shutting down via loginwindow.app tries to : kill all processes separately, while shutdown kills init with a SIGTERM. The : latter seems to work. : Why do I think that httpd is to blame? Well, if I leave it out the restart problems : are not there. : Besides, killing httpd itself with SIGTERM will crash NEXTSTEP. On intel, the big : console window appears and the system hangs. Luckily, the system still reacts : to ALT-Numlock followed by 'r', so that a clean reboot is possible. : Killing httpd with SIGKILL does not trigger the bug. : I would like to get in contact with people running CERN httpd 3.0 under : NEXTSTEP to see if this can be worked out. Or I'd like to get into contact with : people who know how to solve this problem, or what triggers it. : Thanks,
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: NEXTSTEP and 1.8M disks ;-) Date: 7 Jul 1995 23:17:39 GMT Organization: Technical University of Berlin, Germany Message-ID: <3tkfaj$are@news.cs.tu-berlin.de> References: <3tj7sl$qpj@sun0.urz.uni-heidelberg.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Gregor Hoffleit (flight@mathi.uni-heidelberg.de) wrote: : In the PC world, it gets more and more common to format floppy disks : with more than 80 tracks. Even IBM delivers OS/2 Warp on 1.8M : formatted disks, and it works. This is not due to formatting more than 80 tracks. More than 80 tracks are a bad idea 'cause there might be differences in the mechanics of several floppy drives. IBM achives the higher capacity by formatting less but bigger sectors thus saving space for sector headers and inter sector gaps. : So I formatted a disk with a DOS : utility to 1.8M and tried to mount this disk with NEXTSTEP, and - what : a surprise - it worked! NEXTSTEP/Intel 3.3 finds 1.66MB free. This is no surprise to me because the disk start looks the same (if you're using more tracks). Did you try to read one of the IBM OS/2 Floppies. _This_ would be a surprise (if you had success :-) : But now : it's not possible to write more than 1.39 MB on the disk. The : filebrowser and df say that there are some two hundred KB free, but : it's impossible to write any more. : Perhaps this could be changed in a next release of the DOS.fs ? Support of the IBM format: Yes, 'cause this seems sensible to do (no compatibility problems). Support for formats with more than 80 tracks: no, 'cause this is going to do more harm than it does good. Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs,comp.infosystems.www.misc From: gemoe@proximus.north.de (Gerhard Mpller) Subject: Re: CERN httpd 3.0 triggers bug in NEXTSTEP 3.3 References: <DB7A4p.86@awt.nl> Organization: German NeXT User Group, Oldenburg. Date: Sat, 8 Jul 1995 13:14:31 GMT Message-ID: <1995Jul8.131431.12985@proximus.north.de> In article <DB7A4p.86@awt.nl>, Gerben Wierda <G.C.Th.Wierda@AWT.NL> wrote: >I use CERN httpd 3.0, because it has caching and proxy support, and I need the >cache. There is a speed penalty using the Cache and the Server on the same machine. If you have several machines and get a lot of http requests, you should consider installing the NCSA httpd (or Apache or CERN httpd) and use the Harvest Object Cache. However, I still use the CERN httpd 3.0, caching enabled on black h/w w/o any problem so far. Gerhard. -- N < principiis obsta! >------------------< PGP Key available on request > N e Gerhard Moeller, Amselweg 16, 26122 Oldenburg (FRG) [*: 02/21/1968] e X Private: gemoe@proximus.north.de Phone (voice): +49-441-507856 X T Uni: Gerhard.Moeller@Informatik.Uni-Oldenburg.DE NeXTmail & MIME T NoGeNUG - Northern German NeXT User Group: NoGeNUG@proximus.north.DE
From: thwang@atom.ecn.purdue.edu (Tommy Kuei-che Hwang) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs,comp.infosystems.www.misc Subject: Re: CERN httpd 3.0 triggers bug in NEXTSTEP 3.3 Date: 9 Jul 1995 01:21:41 GMT Organization: Purdue University Engineering Computer Network Message-ID: <3tnav5$msq@mozo.cc.purdue.edu> References: <DB7A4p.86@awt.nl> <1995Jul8.131431.12985@proximus.north.de> i I need that CERN httpd. How did you guys get it? for some reason, I can not even decompresse it.. I did use binary transfer mode. help???
From: steve@xray.rice.edu (Steve Ludtke) Newsgroups: comp.sys.next.bugs Subject: Problem with control-c in Teminal! Date: 9 Jul 1995 02:54:42 GMT Organization: Rice University, Houston, Texas Message-ID: <3tngdi$okf@larry.rice.edu> The strangest thing has been happening under 3.3. I belive the problem started occuring when I installed 3.3. Control-c doesn't work in Terminal.app. It seems to return something and a carriage return. I checked the keymapping with Keyboard.app, and checked for the correct response using the keyboard panel in Preferences.app. I also checked stty for any obious strangenesses. I couldn't find anything. It also doesn't seem to depend which shell I use. Any suggestions??? Anyone??? -- ------------------------------------------------------------------------- Steve Ludtke | Physics Dept., Rice Univ. steve@ion.rice.edu | stevel@alumni.caltech.edu | * Those who do ARE * 72335,1537 @ compuserve | The converse also holds
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 10 Jul 1995 04:15:14 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <3tq9gi$lof@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:Michael_Pizolato@afs.com) Dan Grillo (mailto:dan_grillo@next.com)
Newsgroups: comp.sys.next.bugs Subject: Re: getcwd and posix libs Message-ID: <1995Jul10.152150.45152@yogi.urz.unibas.ch> From: frank@ifi.unibas.ch Date: 10 Jul 95 15:21:50 MET References: <1995Jun26.115903.45046@yogi.urz.unibas.ch> <3tjtki$5e9@gap.cco.caltech.edu> ccwf@locke.klab.caltech.edu (Charles Fu) wrote: > In article <1995Jun26.115903.45046@yogi.urz.unibas.ch>, Robert Frank <frank@marcie> wrote: > > I've been trying to compile a posix / standard C program which uses getcwd. > > ... > > Compilation is fine, however, the executable never finds a working > > directory! Getcwd always returns NULL and complains that there is no such > > path, while the old bsd getwd correctly returns the working directory. > Note that getcwd() calls readdir() recursively, so you may be > getting bitten by the buggy readdir() function in the POSIX library, > about which I have posted previously on multiple occasions. The > readdir bug is most likely to affect directories with mount points; > so if your getcwd bug happens primarily when the working directory is > on a mounted disk and doesn't usually affect working directories on > the root disk (and which have no parent directories with mount > points), then you are probably experiencing the readdir bug. > Precicely the case, thanks! > Again, the simplest workaround for the readdir bug is to grab the > sources from somewhere and compile it yourself. If you are using > POSIX under NEXTSTEP extensively, you probably want to do this > anyways. > I believe that this is NeXT's businiss - not mine (after all, I pay for this service...) -Robert -- Institut fuer Informatik tel +41 (0)61 321 99 67 Universitaet Basel fax. +41 (0)61 321 99 15 Robert Frank Mittlere Strasse 142 rfc822: frank@ifi.unibas.ch (NeXT,MIME mail ok) CH-4056 Basel X400: S=frank;OU=ifi;O=unibas;P=switch;A=arcom;C=ch Switzerland
From: lamdong@cory.Berkeley.EDU (LD) Newsgroups: comp.sys.next.bugs Subject: UPS interface with NeXTStep, Help !! Date: 10 Jul 1995 21:36:04 GMT Organization: University of California, at Berkeley Message-ID: <3ts6g4$atg@agate.berkeley.edu> Hi, I have UPS (Back-APC 400), NextStep 3.2, Powerguardian from Benatong(Software monitor UPS on NeXt) , but I don't have cable to connect UPS interface port (D9) to PC Serial Port (COM2-D9pin). If anyone has used and installed UPS , please give me some help. I appriciate for your time. Thanks in advance. - Lam (lamdong@cory.berkeley.edu)
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin From: Gerben_Wierda@RnA.NL Subject: Netinfo problem? Netinfo <-> route? Message-ID: <DBIuF0.2H4@RnA.NL> Sender: gerben@RnA.NL (Gerben Wierda) Organization: G.R.O.S.S. Date: Mon, 10 Jul 1995 22:28:11 GMT MST PPP under NEXTSTEP. Getting dial-on-demand to work. Start situation pppd is up, idle (just started). A default route to the receiving pppd is available. Situation one: using either nslookup from NEXTSTEP or nslookup from BIND 4.9 to trigger a connection. Using a name we know is not available in the cache of named. Part of log at debug 11: 7/10-23:23:31-2982 du0: pppd auto idle 120 passive cufa 38400 debug 11 192.168.1.1~:193.78.249.34 7/10-23:23:31-2982 new_state '[off, du0]' 7/10-23:24:07-2982 Allocating a 2122-byte buffer 7/10-23:24:07-2982 udp 192.168.1.1/domain -> 192.36.148.17/domain 64 bringup 7/10-23:24:07-2982 new_state '[dialing, du0]' 7/10-23:24:07-2982 Reading /etc/ppp/Devices 7/10-23:24:07-2982 Reading /etc/ppp/Dialers 7/10-23:24:07-2982 Lock file /usr/spool/uucp/LCK/LCK..cufa created Situation two: using ftp from NEXTSTEP to trigger a connection. Using a name we know is not available in the cache of named. The main difference is that we know nslookup will never use netinfo, where of ftp, we are not sure. Part of log at debug 11: 7/10-23:28:30-3001 du0: pppd auto idle 120 passive cufa 38400 debug 11 192.168.1.1~:193.78.249.34 7/10-23:28:30-3001 new_state '[off, du0]' 7/10-23:28:59-3001 Allocating a 2122-byte buffer 7/10-23:30:02-3001 udp 192.168.1.1/domain -> 192.52.195.10/domain 65 bringup 7/10-23:30:02-3001 new_state '[dialing, du0]' 7/10-23:30:02-3001 Reading /etc/ppp/Devices 7/10-23:30:02-3001 Reading /etc/ppp/Dialers 7/10-23:30:02-3001 Lock file /usr/spool/uucp/LCK/LCK..cufa created Note the timeing info especially. Between the allocation statement and the bringup statement one minute is lost, which makes it look like a netinfo timeout. I tried no 2 again, now with lookupd logging enabled. And look: getservbyname (ftp) Ncalls: 31 Elapsed: 9 Total time: 521 getpwnam (gerben) Ncalls: 4 Elapsed: 6 Total time: 43 gethostbyname (ftp.omnigroup.com.) Ncalls: 1 Elapsed: 62512 Total time: 62512 getservbyport (53) Ncalls: 1 Elapsed: 28 Total time: 28 the gesthostbyname call takes 1 minute, then it starts for port 53 (the name server). But roughly at the same moment that this happens, ftp reports a timeout. Does lookupd report a failure before named had it's chance? Does lookupd use named at all? Let's see the same order of events when pppd is up: getservbyname (ftp) Ncalls: 62 Elapsed: 9 Total time: 868 getpwnam (gerben) Ncalls: 17 Elapsed: 7 Total time: 137 gethostbyname (ftp.lighthouse.com.) Ncalls: 2 Elapsed: 5714 Total time: 68226 And the connection is established! No getservbyport (53) call! Another test: using nslookup and checking if lookupd sees something. Well, it does not report a gethostbyname() but it does report a getservbyport (53), after which the dialing process starts. This may or may not be a coincidence. I am baffled. gethostbyname() in the first example times out and after the timeout ppp starts to dial. In the second example when the link is up, it does not. What is going on here?!?! I attach a complete MST PPP log below, though I get the distinct feeling that Netinfo might be to blame here. LAST MINUTE THOUGHT: is the fact that there is a default route to the other PPP side, reason for NetInfo to try to find a Netinfo parent maybe? But if that is the case, why isn't pppd reporting this activity on its interface? And besides, hostmanager is set to 'use local domain only' and 'readable by local domain only'. ======================================== LOG =========================== 7/10-23:28:29-3000 Allocating a 2122-byte buffer 7/10-23:28:29-3000 tun_get_flags(du0) = 0x10 7/10-23:28:29-3000 tun_out_open: tun_fd=5 tun_if_name=du0 7/10-23:28:29-3000 tun_set_flags(du0, 0x10) 7/10-23:28:29-3000 tun_set_dst_address(du0, 193.78.249.34) 7/10-23:28:29-3000 tun_set_address(du0, 192.168.1.1) 7/10-23:28:29-3000 tun_set_flags(du0, 0x11) 7/10-23:28:29-3000 tun_set_mtu: Setting mtu to 1500 7/10-23:28:29-3001 Reading /etc/ppp/Filter 7/10-23:28:29-3001 Compiling /etc/ppp/Filter... 7/10-23:28:29-3001 get_host_addr("193.78.249.34") = 193.78.249.34 7/10-23:28:29-3001 get_host_addr("193.78.240.11") = 193.78.240.11 7/10-23:28:30-3001 /etc/ppp/Filter installed 7/10-23:28:30-3001 Reading /etc/ppp/Systems 7/10-23:28:30-3001 get_host_addr("193.78.249.34") = 193.78.249.34 7/10-23:28:30-3001 Morning Star Technologies PPP 7/10-23:28:30-3001 Version 1.4.1.6 [5-Mar-1995 20:35:12 next-3.2] 7/10-23:28:30-3001 This is demonstration software and will expire Thu Jul 27 15:22:54 1995 7/10-23:28:30-3001 new_state '[off, du0]' 7/10-23:28:30-3001 du0: pppd auto idle 120 passive cufa 38400 debug 11 192.168.1.1~:193.78.249.34 7/10-23:28:30-3001 new_state '[off, du0]' 7/10-23:28:59-3001 Allocating a 2122-byte buffer 7/10-23:30:02-3001 udp 192.168.1.1/domain -> 192.52.195.10/domain 65 bringup 7/10-23:30:02-3001 new_state '[dialing, du0]' 7/10-23:30:02-3001 Reading /etc/ppp/Devices 7/10-23:30:02-3001 Reading /etc/ppp/Dialers 7/10-23:30:02-3001 Lock file /usr/spool/uucp/LCK/LCK..cufa created 7/10-23:30:04-3001 Opened /dev/cufa on tty_fd=0 7/10-23:30:04-3001 Dialing 193.78.249.34 (cufa 38400 *,071661551 MT2834ZDX) 7/10-23:30:04-3001 abort on (NO DIALTONE) 7/10-23:30:04-3001 timeout set to 60 seconds 7/10-23:30:04-3001 expect () 7/10-23:30:04-3001 got it 7/10-23:30:04-3001 send (ATM1^M) 7/10-23:30:04-3001 expect (OK) 7/10-23:30:04-3001 ATM1^M^M 7/10-23:30:05-3001 OKgot it 7/10-23:30:05-3001 send (ATDT*,071661551^M) 7/10-23:30:06-3001 expect (CONNECT) 7/10-23:30:06-3001 ^M 7/10-23:30:06-3001 ATDT*,071661551^MSIGTERM -- Gerben_Wierda@RnA.nl (Gerben Wierda) NEXTSTEP RD242 "If you don't know where you're going, any road will take you there" Paraphrased in Alice in Wonderland, originally from the Talmud.
Newsgroups: comp.sys.next.bugs From: Gerben_Wierda@RnA.NL Subject: SOLVED: CERN httpd 3.0 and system death problem Message-ID: <DBLJ9F.FA@RnA.NL> Sender: gerben@RnA.NL (Gerben Wierda) Organization: G.R.O.S.S. Date: Wed, 12 Jul 1995 09:20:03 GMT Why did killing CERN httpd kill my system? A small tour in UNIX behaviour. I have installed CERN httpd 3.0 from the binary distribution, since I wanted to use a caching proxy. I also wanted to test letting httpd itself use another proxy (www.nl.net). Since, for NEXTSTEP, you cannot do that from within the configuration file, I had to create a shell script for starting httpd (note the environment variables are commented out, because the proxy www.nl.net was very slow): The file /usr/local/WWW/bin/rna-httpd looks like this: #!/bin/sh #http_proxy=http://www.nl.net/ #gopher_proxy=http://www.nl.net/ #no_proxy=www.rna.nl #export http_proxy #export gopher_proxy #export no_proxy /usr/local/WWW/bin/httpd -r /usr/local/WWW/config/rna.conf >/dev/console 2>&1 & In /etc/rc.local it reads: /usr/local/WWW/bin/rna-httpd because, why put a description on how you want to start it in two places? The result however is that httpd ends up in the 'process group' on 'controlling terminal' console with some other fairly important system programs. Result: when httpd receives a SIGTERM, it will kill those other important processes, effectively killing your environment. Here is a short description of actions and results. Logged in as root, I fired Stuart and opened a window with 'sh' as shell. (Note 'tcsh of csh won't demonstarte the problem). # ps ax | grep httpd 1124 p2 S 0:00 grep httpd There is no httpd running. Start it directly: # /usr/local/WWW/bin/httpd -r /usr/local/WWW/config/rna.conf >/dev/console 2>&1 Check if it is running: # ps ax | grep httpd 1127 ? SW 0:00 /usr/local/WWW/bin/httpd -r /usr/local/WWW/config/rna.conf 1128 p2 S 0:00 grep httpd It is. Note '?' denotes no controlling terminal. # kill -15 1127 Killed it. Nothing special happens. # bin/rna-httpd Started it from the shell script. # ps ax | grep httpd 1146 p2 SWN 0:00 /usr/local/WWW/bin/httpd -r /usr/local/WWW/config/rna.conf 1147 p2 S 0:00 grep httpd Check: it is running. Note 'p2' denotes a controlling terminal. the 'N' in SWN is 'nice' since it was started by sh in the background. This, btw, is important. # kill -15 1146 And BOOM, there goes the window, it's shell together with httpd. Now figure out when this happens when httpd shares the process group and the controlling terminal with loginwindow.app, the WindowServer, etc and you get the picture what happens. Note, when you remove the '&' from the sh script, it works fine. When you have the '&' in place, and atsrat it from tcsh it also detaches itself. Only when you start the sh script from another sh and you have the '&' in place, httpd does not detach itself from the group, nor is detached by others, so the foundation of catastrophe is in place. BTW, on intel systems, this also prevents yor system from powering down or rebooting properly from loginwindow.app. On black, it reboots anyway, suggesting that there are subtle differences between black and white loginwindow.app and WindowServer, probably something like loginwindow.app hanging on some output under intel, where it happily goes on on m68k. This probably is a bug in httpd, since it should detach itself from the process group and the controlling terminal under all circuimstances, I guess. As found out by: Tom_Hageman@RnA.nl and Gerben_Wierda@RnA.nl One problem down, one to go (MST PPP not dialing on demand properly, probably some lookupd problem?) -- Gerben_Wierda@RnA.nl (Gerben Wierda) NEXTSTEP RD242 "If you don't know where you're going, any road will take you there" Paraphrased in Alice in Wonderland, originally from the Talmud.
Newsgroups: comp.sys.next.bugs From: Gerben_Wierda@RnA.NL Subject: SOLUTION: your NS/i486 hangs when powering down from loginwindow.app Message-ID: <DBLJp1.G3@RnA.NL> Sender: gerben@RnA.NL (Gerben Wierda) Organization: G.R.O.S.S. Date: Wed, 12 Jul 1995 09:29:25 GMT Various people asked me to tell them if I ever found out about my cern-httpd problem with crashing system and hanging reboot. I refer to my separate post in csn.bugs and csn.sysadmin for that particular story, but here is a small tip on locating such a 'bug'. The bug is probably not really a bug of NEXTSTEP. Here is what to look for: if you start things at boot time though startup, and you put them in the background (through '&', also look in shells started from rc.local) and if they don't detach themselves properly from controlling terminal/process group, they might end up in the same group as loginwindow.app and the WindowServer. This means that when loginwindow.app starts rebooting, it sends a SIGTERM to all processes, but the non-detached process will pass that on to loginwindow.app itself. In effect, loginwindow will SIGTERM itself that way. If you reboot through the console, or from a shell with the 'reboot' or 'shutdown' command, the init process will kill everything nicely. Just FYI (man this was a funny problem), -- Gerben_Wierda@RnA.nl (Gerben Wierda) NEXTSTEP RD242 "If you don't know where you're going, any road will take you there" Paraphrased in Alice in Wonderland, originally from the Talmud.
From: guichard@elias.ens.fr (Eric Guichard) Newsgroups: comp.sys.next.bugs Subject: NS 3.3 and sas Date: 12 Jul 1995 11:54:26 GMT Organization: Ecole Normale Superieure Distribution: world Message-ID: <3u0d5i$a19@nef.ens.fr> Keywords: statistics, sas Since I upgraded my old Nexts with NS 3.3, I ve got a big problem with a statistical software called "sas". I was before under NS 3.0. This software looses datas, runs during hours without giving any result (I'm obliged to kill it) and so on. The people of sas do not seem to be able to give a quick solution (I call them everyday for one week). What I do not understand is that NS 3.3 is not so new, and that sas is a rather well developed company in the US. Did you find the same kind of problem, and how did you fix it? Or do you know a sas and NS expert? My opinion is that the problem doesn't come from NS3.3. Thanks Eric Guichard guichard@ens.fr
Newsgroups: comp.sys.next.bugs From: NextStep 3.3 screen hangs Subject: Documentation of Application class looks very strange in 3.3 Message-ID: <1995Jul12.180343.10963@kurt.in-berlin.de> Sender: news@kurt.in-berlin.de Date: Wed, 12 Jul 1995 18:03:43 GMT Hi! Open /NextLibrary/Documentation/NextDev/GeneralRef/02_ApplicationKit/Classes/App lication.rtf and you will see the documentation of the ActionCell class :-( Does anyone have the correct document? Please send me a copy .... Gerald -- --------------------------------------------------------------------- | GERALD ERDMANN | email: gerald @ kurt.in-berlin.de (NeXTmail welcome) | voice: +49 30 372 43 10 (Germany - Berlin) | crypt: pgp2 public key available |
From: mark@bohica.net (Mark Gregory Salyzyn) Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Re: Mux driver and DigiBoard PC/8e trouble Message-ID: <DBy801.HyK@bohica.net> Date: Wed, 19 Jul 1995 05:45:35 GMT References: <3u6c5n$p62@agate.berkeley.edu> Followup-To: comp.sys.next.hardware Organization: UNIX drivers'R'Us lamdong@cory.Berkeley.EDU (LD) writes: > Does anyone install or have some experiences MuxV1.7 driver for Eight port >intelligent multi-channel board (DigiBoard PC/8e) Mux works with `dumb' multiplexed interrupt cards like the DigiBoard PC/8, and AST multiport cards, or the BOCA six port card. The PC/8e is a `smart' serial card and does not work with Mux (sorry). Does anyone have the technical specs on the PC/8e handy? Enjoy -- Mark
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.advocacy,comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Hey NeXT! if you can fix the serial drivers, you can fix lookupd! Date: 18 Jul 1995 15:40:08 GMT Organization: Information Technology Solutions, Inc. Message-ID: <3ugkko$226@www.its.com> References: <DBoJnv.5JM@RnA.NL> <3u5vad$1hm@tybrin4.hsv.tybrin.com> <3u6t8d$dso@gap.cco.caltech.edu> madler@cco.caltech.edu (Mark Adler) wrote: > I hope you guys are telling NeXT and not just posting here. People at NeXT do read the comp.sys.next newsgroups, and they do respond to ideas brought up. However, I'd suspect that the people at NeXT who read Usenet news are only going to do something about an idea if they understand it and agree...so sending feedback to NeXT directly is always a good thing to do. -Chuck -- Charles Swiger -- chuck@its.com | Information Techology Solutions, Inc. --------------------------------+-------------------------------------- CrashCatcher Development, Systems and Networking Administrator
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: fozztexx@nvc.cc.ca.us (Chris Osborn) Subject: Bug in Intel binaries when compiling Quad fat Message-ID: <DBzKx4.9I7@nvc.cc.ca.us> Sender: news@nvc.cc.ca.us Organization: Napa Valley College Date: Wed, 19 Jul 1995 23:22:16 GMT I've discovered a bug when I compile programs Quad fat under 3.3/Dev3.3. It doesn't cause a problem in the m68k binary, but it does in the i386 (don't know about hppa or sparc). The contents of the memory that self points to are getting set to zero after I enter a loop. defaultRun = malloc(sizeof(NXRun)); defaultRun->font = aFont; defaultRun->chars = 0; style = malloc(sizeof(NXTextStyle)); style->indent1st = 0; style->indent2nd = 0; style->lineHt = [defaultRun->font pointSize] + 1; style->descentLine = 2; style->alignment = NX_LEFTALIGNED; style->numTabs = 10; style->tabs = malloc((style->numTabs * 2) * sizeof(NXTabStop)); /* Things are fine here (gdb) print *self $7 = { isa = 0x625a0, theText = 0x0, theRuns = 0x0, runStack = 0x0, defaultRun = 0x470ea0 } */ width = [defaultRun->font getWidthOf:TABWIDTH]; for (i = 0; i < style->numTabs; i++) { style->tabs[i].kind = NX_LEFTTAB; style->tabs[i].x = (int) (width * (i + 1)); /* Self is wiped right here after the first time through (gdb) print *self $8 = { isa = 0x0, theText = 0x0, theRuns = 0x0, runStack = 0x0, defaultRun = 0x0 } */ } I haven't seen this problem when compiling for only i386 and m68k, but when I throw in the hppa and sparc, this problem occurs. -- Chris Osborn, Network Administrator Voice: 707 253 3130 Napa Valley College Fax: 707 253 3063 2277 Napa-Vallejo Hwy., Napa, CA, 94558 <fozztexx@nvc.cc.ca.us> MIME ok, NeXTMail tolerated
From: plongsi@falcon.inetnebr.com (Pohl Longsine) Newsgroups: alt.amateur-comp,comp.os.coherent,comp.sys.next.bugs,fj.sci.psychology Subject: Re: ===>> FREE 1 yr. Magazine Sub sent worldwide- 300+ Popular USA Titles (fwd) Followup-To: alt.amateur-comp,comp.os.coherent,comp.sys.next.bugs,fj.sci.psychology Date: 20 Jul 1995 12:46:33 GMT Organization: Internet Nebraska Message-ID: <3ulj79$714@duck.inetnebr.com> References: <Pine.SUN.3.91.950717164143.13645D-100000@amanda.dorsai.org> Does anybody have a shell script to generate 10,000 anti-usenet-advertising flames randomly and send them all via an anonymous mail server? -- ____/| | Pohl Longsine, OpenStep Software Developer \ o.O| GPF! | "I don't do Windows." =(_)= CTLALTDLT! | plongsi@inetnebr.com (Internet Nebraska) U (Bill Gates, The Cat) | NeXT & MIME mail formats accepted.
Newsgroups: comp.sys.next.bugs From: fred@improve.fdn.org (Frederic STARK) Subject: NEXTSTEP GCC bug (?) Message-ID: <1995Jul20.090102.11641@improve.fdn.org> Sender: news@improve.fdn.org Organization: improve SA - La Defense, Paris, France. Date: Thu, 20 Jul 1995 09:01:02 GMT The folowing program does not compile with gcc. main() { int i=0; i*i; } > cc -ansi gcc-test.c -o gcc-test gcc-test.c:3: undefined type, found `i' I does however compiles with -traditional option. Cheers, ----------------------------------------------------------------- frederic stark -- fred@improve.fdn.org -----------------------------------------------------------------
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: HELP: dd crashed my root partition Date: Fri, 21 Jul 1995 00:48:08 +0200 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.950721003756.10957A-100000@hphalle0.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I'm crossposting this to sysadmin and bugs, because I believe it's interesting to both groups: I recently tried to use dd for backup purposes. Because I'm not used to dd I tried it with the floppy and tried to put a file to the floppy as given in the manpage: dd in=file of=/dev/rfd0a This didn't work, so I thought I did something wrong. Now I was trying to access the whole floppy disk: dd in=file of=/dev/rfd0d and this _seemed_ to work. In deed it did nothing to the floppy but seriously to my rootdisk!!!! From now on, booting resulted in an endless loop. After some trials and errors, I got the following working: Booting from floppy and entering at the prompt: sd()mach_kernel Hups! What is happening here? What data did I destroy, and how reliable is my system now! It works, but how can I be sure it will continue? Any tips on how to go on (now I'm going to restore the bootblock again, but I believe I crashed even more) would be very much appreciated! One additional question: during my testing to get back my system I did the following: /usr/etc/driverLoader d=SCSITape. I always answered yes, but I couldn't access my tape in single user mode. Mode sense went wrong! Why? Many thanks in advance, Boerny. (who might spend 1000$ in backup media now...) -- Bernhard Scholz (IRC: Boerny) Look at http://www.leo.org/archiv/NeXT/ scholz@informatik.tu-muenchen.de, scholz@c86501.rm.op.dlr.de 'X is what it was designed for --- to open multiple terminal windows.' T.W.
Control: cancel <DC10pt.C8o@nvc.cc.ca.us> Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: fozztexx@nvc.cc.ca.us (Chris Osborn) Subject: cancel Message-ID: <DC1GpE.FHn@nvc.cc.ca.us> Sender: news@nvc.cc.ca.us Organization: Napa Valley College Date: Thu, 20 Jul 1995 23:46:26 GMT <DC10pt.C8o@nvc.cc.ca.us> was cancelled from within trn. -- Chris Osborn, Network Administrator Voice: 707 253 3130 Napa Valley College Fax: 707 253 3063 2277 Napa-Vallejo Hwy., Napa, CA, 94558 <fozztexx@nvc.cc.ca.us> MIME ok, NeXTMail tolerated
From: doug@foxtrot.ccmrc.ucsb.edu (Douglas Scott) Newsgroups: comp.sys.next.bugs Subject: Re: NEXTSTEP GCC bug (?) Date: 20 Jul 1995 23:11:34 GMT Organization: Center for Computer Music Research and Composition, U.C.S.B. Message-ID: <3umnr6$eav@yuggoth.ucsb.edu> References: <1995Jul20.090102.11641@improve.fdn.org> In article <1995Jul20.090102.11641@improve.fdn.org> fred@improve.fdn.org writes: ] ] ]The folowing program does not compile with gcc. ] ]main() ]{ int i=0; ] i*i; ]} ] ]> cc -ansi gcc-test.c -o gcc-test ]gcc-test.c:3: undefined type, found `i' ] ]I does however compiles with -traditional option. ] Interesting...The old gcc compiler on the NeXT must be interpreting this like: int i=0; i *i; And thinks 'i' should be a typename. Newer versions of gcc compiled on the NeXT give no warning or error unless -Wall is used, and then you get: iii.cc: In function `int main()': iii.cc:3: warning: statement with no effect -- Douglas Scott | Senior Development Engineer Tel: (805) 893-8352 | Center for Computer Music Research and Composition Internet (NeXTMail ok): | University of California, Santa Barbara <doug@ccmrc.ucsb.edu> | http://www.ccmrc.ucsb.edu/
From: pete@ohm.york.ac.uk (-bat.) Newsgroups: comp.sys.next.bugs Subject: Mounding MS-DOS CD-ROM remotely Date: 21 Jul 1995 09:46:38 GMT Organization: The University of York, UK Message-ID: <3unt1u$576@mailer.york.ac.uk> If I put a CD rom into the drive on my NeXT, export it and mount it anywhere else all the files come out as garbage. Anyone know why ? This is true mounting it on a PC or a NeXT. The machine is running 3.3 -bat.
Newsgroups: comp.sys.next.bugs From: fred@improve.fdn.org (Frederic STARK) Subject: Re: NEXTSTEP GCC bug (?) Message-ID: <1995Jul21.093751.15224@improve.fdn.org> Sender: news@improve.fdn.org Organization: improve SA - La Defense, Paris, France. References: <1995Jul20.090102.11641@improve.fdn.org> Date: Fri, 21 Jul 1995 09:37:51 GMT In article <1995Jul20.090102.11641@improve.fdn.org> fred@improve.fdn.org (Frederic STARK) writes: > > The folowing program does not compile with gcc. > > main() > { int i=0; > i*i; > } > > > cc -ansi gcc-test.c -o gcc-test > gcc-test.c:3: undefined type, found `i' > > I does however compiles with -traditional option. Sorry to folow-up my own post, but I should be more precise (I received several mail :-) Changing i*i to anything else is not an issue: This program is ANSI-compliant (even if main doesn't return anything) and gcc should compile it properly with -ansi option. Otherwise gcc is broken or I missed somthing about ANSI-C. Cheers, ----------------------------------------------------------------- frederic stark -- fred@improve.fdn.org -----------------------------------------------------------------
Newsgroups: comp.sys.next.bugs From: fred@improve.fdn.org (Frederic STARK) Subject: Strange behaviour of ProjectBuilder with Libraries Message-ID: <1995Jul21.100405.15384@improve.fdn.org> Keywords: PB, LIB, BUG Sender: news@improve.fdn.org Organization: improve SA - La Defense, Paris, France. Date: Fri, 21 Jul 1995 10:04:05 GMT Happens on Black/White/SPARC with NS3.3Dev+EOF1.1 How to reproduce: * Create a new project of type Library (name: Lib) * Create a subproject of type Subproject (name: Sub) * Save If you open Lib and look in the attribute part, you'll find 3 TextField If you then go to the Sub part and ask for attributes, you'll find 4 TextField. Close the project, re-open it and directly go to the Sub part. If you ask for the attributes, you'll find 2 TextField and 1 PopUp If you go to the root(Lib) part and ask for the attributes, you'll find 1 TextField and 1 PopUp. (If you play with them, you'll mess your project) Only the first attribute inspector of each case is OK. Workaround: Close your Library project before inspecting attributes of any node. Cheers, ----------------------------------------------------------------- frederic stark -- fred@improve.fdn.org -----------------------------------------------------------------
Newsgroups: comp.sys.next.bugs From: oneill@cs.sfu.ca (Melissa O'Neill) Subject: Re: NEXTSTEP GCC bug (?) Message-ID: <1995Jul21.213405.6705@cs.sfu.ca> Organization: Simon Fraser University, Burnaby, BC, Canada References: <1995Jul20.090102.11641@improve.fdn.org> Date: Fri, 21 Jul 1995 21:34:05 GMT Frederic STARK <fred@improve.fdn.org> wrote: > The folowing program does not compile with gcc. > > main() {int i=0; i*i;} > >> cc -ansi gcc-test.c -o gcc-test > gcc-test.c:3: undefined type, found `i' > > It does however compiles with -traditional option. Yes, it is a bug, but it's not a GCC bug, it's a bug in NeXT's horrible "NeXT DevKit-based CPP 3.1". If you don't compile Objective-C programs, you can just remove (or hide) /lib/cpp-precomp. If you're in love with NeXT's `precompiled headers', then you're probably stuck with it. You can also do `cc -traditional-cpp' to fix the problem, since it'll give you /lib/cpp and not /lib/cpp-precomp, but it won't be truely ANSI, since the preprocessor will be running in `traditional' mode. The failure of NeXT's CPP is almost understandable, in fact. It tries to grok variable declarations so that it can stuff them into it's precompiled header files, and a statement like `foo *bar;' can be read as a declaration of a pointer bar to an object of type foo, and so it wonders why you've not declared the type `foo'. Hope this helps, Melissa. --- I really wanted to stay in bed today. // Melissa O'Neill <oneill@cs.sfu.ca>
From: seagull@borult.edu (seagull) Newsgroups: comp.sys.next.bugs Subject: `+=;>`"W70``_UD Date: Sat, 22 Jul 1995 10:23:21 GMT Message-ID: <rztvePIsPjLCo@borult.edu> Organization: borult Sender: seagull@borult.edu NNTP-Posting-Host: borult.edu ********************* * * * FREE MONEY!!! * * * ********************* CAN BE YOURS! If you are an 80-IQ welfare mother producing illegitimate offspring at 9-month intervals starting at the age of 13. Generous funds are also available if you are an illegal alien in need of medical services while you give birth to a new "American" citizen, now that you are north of the Rio Grande. Grants to "help you get started" are available if you are a member of Jewish organized crime gangs newly arrived as "refugees" from Russia. If you are a homosexual "performance artist" the National Endowment for the Arts will pay you handsomely to literally or figuratively fling dung at your audience. If you are a foreign dictator currently in the good graces of the New World Order elitists, multiple billions can be yours, as long as you "cooperate". And if you burn down your own neighborhood when a court decision doesn't go your way, don't worry: truckloads of money will soon arrive to build you new breeding colonies--er, I mean houses, apartments, community centers, swimming pools, etc. Oh, I forgot to mention one thing: all this free money is not available to you if you are an ordinary straight White American, a descendant of the men and women of Europe who discovered, pioneered, and built America and made her the greatest nation ever known. If you are one the quiet, sober, thrifty, hard-working, decent and almost infinitely tolerant White majority, your job is to work hard to provide all the free money and free goodies that the criminals--uh, I mean politicians who like to give away to buy votes for the minority and special interest voting blocs. IF YOU LIKE THINGS THE WAY THEY ARE-- keep smiling, keep watching TV, keep on voting Democrat and Republican, and, above all, keep on working hard to keep the "free money" flowing! BUT--IF YOU WOULD RATHER give the crooked politicians a fair trial, and the minority parasites packing to fend for themselves, then listen to our weekly radio program, American Dissident Voices, and contact us to find out how we're working to restore America to her rightful owners. NATIONAL ALLIANCE BOX 90 HILLSBORO, WV 24946 U.S.A. Selected transcripts of American Dissident Voices, along with other patriotic texts can be obtained from our ftp site: ftp://ftp.netcom.com/pub/NA/NA/ - - - - - - - - - - - - - - - - - - AMERICAN DISSIDENT VOICES SCHEDULE - - - - - - - - - - - - - - - - - - The only truly uncensored patriotic radio program in America, that is not afraid to talk about the REAL forces destroying western civilization. The issues other so-called "conservatives" don't dare mention, is now heard worldwide on shortwave, on satellite, and on AM and FM radio in most of North America. WORLDWIDE ON SHORTWAVE: Saturday 11:30am Central Time (16:30 UTC)....15420 kHz via WRNO Saturday 8:00pm Central Time (01:00 UTC)....7355 kHz via WRNO VIA SATELLITE TO ALL OF NORTH AMERICA Wednesday 5:00pm Central Time...............C1, channel 15, 7.56 MHz audio PACIFIC AND MOUNTAIN USA from California to Washington to Montana, via KVEG, Las Vegas: Sunday night 9:30pm Pacific Time........840 on your AM dial MIDWEST AND GULF SOUTH USA from the Gulf of Mexico to the Dakotas and Minnesota, plus parts of south central Canada via KAAY, Little Rock, Arkansas: Friday night/Saturday morning 1:00am Central Time.......1090 on your AM dial Saturday night 11:30pm Central Time......1090 on your AM dial UPPER MIDWEST AND NORTHERN MOUNTAIN USA from Minnesota and Iowa west to Colorado and Montana, plus a large part of western Canada via KXEL, Waterloo, Iowa: Mon-Fri 9:30pm Central Time.......1540 on your AM dial LITTLE ROCK, ARKANSAS AREA (these stations can be heard in most of Arkansas) via KMTL, Little Rock: Saturday 10:00am.....................760 on your AM dial via KAAY, Little Rock: Friday night/Saturday morning 1:00am....................1090 on your AM dial Saturday 11:30pm...................1090 on your AM dial WICHITA FALLS, TEXAS AREA via KSEY AM-FM, Seymour: Saturday 8:30am.....................94.3 on your FM dial Saturday 10:00am....................1230 on your AM dial HOUSTON, TEXAS AREA via KGOL, Houston: Saturday 6:00pm.....................1180 on your AM dial HUNTSVILLE / DECATUR, ALABAMA AREAS via WAJF and WYAM, Decatur/Hartselle: Wednesday 9:00am....................1490 on your AM dial Wednesday 9:00am.....................890 on your AM dial TAMPA, FLORIDA AREA via WTIS, St. Petersburg: Saturday 11:30am....................1110 on your AM dial CINCINNATI, OHIO AREA via WTSJ, Cincinnati: Friday 6:00pm.......................1050 on your AM dial RHODE ISLAND, EASTERN MASSACHUSETTS, AND CONNECTICUT via WALE, Providence: Monday 10:00am.......................990 on your AM dial <= Crusader =>
Control: cancel <rztvePIsPjLCo@borult.edu> From: seagull@borult.edu (seagull) Newsgroups: comp.sys.next.bugs Subject: cmsg cancel <rztvePIsPjLCo@borult.edu> Date: 22 Jul 95 23:39:35 GMT Message-ID: <cancel.rztvePIsPjLCo@borult.edu> Organization: borult Sender: seagull@borult.edu Mass spam cancelled by news@uunet.uu.net.
Newsgroups: comp.sys.next.bugs From: tom@basil.icce.rug.nl (Tom Hageman) Subject: Re: NEXTSTEP GCC bug (?) Message-ID: <DC51E2.EI5@basil.icce.rug.nl> Originator: root@obelix.icce.rug.nl Sender: news@basil.icce.rug.nl (NEWS pusher) Organization: Warty Wolfs References: <1995Jul20.090102.11641@improve.fdn.org> <1995Jul21.213405.6705@cs.sfu.ca> Date: Sat, 22 Jul 1995 22:06:02 GMT oneill@cs.sfu.ca (Melissa O'Neill) wrote: > Frederic STARK <fred@improve.fdn.org> wrote: > > The folowing program does not compile with gcc. > > > > main() {int i=0; i*i;} > > > >> cc -ansi gcc-test.c -o gcc-test > > gcc-test.c:3: undefined type, found `i' > > > > It does however compiles with -traditional option. > Yes, it is a bug, but it's not a GCC bug, it's a bug in NeXT's horrible > "NeXT DevKit-based CPP 3.1". If you don't compile Objective-C programs, > you can just remove (or hide) /lib/cpp-precomp. If you're in love with > NeXT's `precompiled headers', then you're probably stuck with it. You > can also do `cc -traditional-cpp' to fix the problem, since it'll give > you /lib/cpp and not /lib/cpp-precomp, but it won't be truely ANSI, > since the preprocessor will be running in `traditional' mode. Not quite. `cc -traditional-cpp' runs GNU cpp in ANSI mode. `cc -traditional' runs GNU cpp in `traditional' pre-ANSI mode (at least since NS3.2dev). An example to illustrate this: tom@basil 77) cat >t.c #define CAT(a,b) a##b int CAT(aap,noot) = 0; tom@basil 78) cc -E -traditional-cpp t.c # 1 "t.c" int aapnoot = 0; tom@basil 79) cc -E -traditional t.c # 1 "t.c" int aap##noot = 0; tom@basil 80) -- __/__/__/__/ Tom Hageman <tom@basil.icce.rug.nl> [NeXTmail/Mime OK] __/ __/_/ IC Group <tom@icgned.nl> (work) __/__/__/ "...to baldly go where no one has gone before." __/ _/_/ -- star trek TNG
From: csa@netcom.com (Ken Nagn) Newsgroups: comp.soft-sys.nextstep,comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.sysadmin Subject: How to get Nextstep 3.2 to use Logitec Serial Mouse Date: 23 Jul 1995 15:52:20 GMT Organization: Paragon Distribution: inet Message-ID: <3utr7k$s79@wally2.hti.net> Mime-Version: 1.0 I am installing nextstep 3.2 and it seems to think I have a bus mouse. Consequently when the summary of devices comes up, I am unable to change my mouse to something else. Is there some key stroke I can use to get it to select the mouse device, or will it even support a serial mouse? Thanks csa@netcom.com
From: Michael Chan <mchan@technet.sg> Newsgroups: comp.sys.next.bugs Subject: Re: NEXTSTEP GCC bug (?) Date: 24 Jul 1995 02:43:19 GMT Organization: Technet, Singapore Message-ID: <3uv1c7$m33@raffles.technet.sg> References: <1995Jul20.090102.11641@improve.fdn.org> <1995Jul21.093751.15224@improve.fdn.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: fred@improve.fdn.org >> > cc -ansi gcc-test.c -o gcc-test >> gcc-test.c:3: undefined type, found `i' Anyway to compile gcc without the Developer's Release? Or that expensive thing is a must in order to compile anything at all? I wish to use gcc, but it requires the cc compiler to work.
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 24 Jul 1995 04:15:11 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <3uv6of$l9a@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: roffel@biw-ag.de (Lee Roffel) Newsgroups: comp.sys.next.bugs Subject: Root Dock reverts to Default Date: 24 Jul 1995 09:00:03 GMT Organization: BIW Systemhaus Message-ID: <3uvnej$idq@next01.biw-ag.de> Hi kids... On a new NeXT 3.3 installation... logged in as root... the changes to the dock are not held between logins. The ME user works ok.. a new user works ok.. but everytime the root user logs out and back in... the dock is on the default settings. The german keyboard settings are also not working for the root user... (ok for others) Any suggestions before we do a new installation??? ciao... .. Lee -- Lee Roffel NeXTMail: roffel@biw-ag.de
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: Root Dock reverts to Default Date: 24 Jul 1995 13:31:10 GMT Organization: Technical University of Berlin, Germany Message-ID: <3v07au$rgp@news.cs.tu-berlin.de> References: <3uvnej$idq@next01.biw-ag.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lee Roffel (roffel@biw-ag.de) wrote: : Hi kids... : On a new NeXT 3.3 installation... logged in as root... the changes to the : dock are not held between logins. The ME user works ok.. a new user works : ok.. but everytime the root user logs out and back in... the dock is on : the default settings. The german keyboard settings are also not working : for the root user... (ok for others) : Any suggestions before we do a new installation??? Maybe you're trapped by UserManager which likes to make /root root's homedirectory. If this does not exist, Workspace doesn't know where to put the defaults. Suggestion: reset root's home to / using NetInfo-Manager or create a new home for root at /root Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: root@arch-ws2 (Operator) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin,de.comp.sys.next, Subject: Parallel Port on Sparc10 does not work Date: 24 Jul 1995 13:51:40 GMT Organization: University of Kassel - Germany Message-ID: <3v08hc$m31@hrz-ws11.hrz.uni-kassel.de> Hardware Configuration: Processor: SPARC (Generic) Primary memory: 32.00 MB Hostname: arch-ws2 software: Software Version 3.3risc (Rose2L) We have not managed to use the parellel port driver on any of our SPARC10 -s. The console says : Cannot open output device '/dev/pp0': Device busy However we haven't had any Problems on the SPARC20-s ....... The problem is very serious, Please help.... Please mail to gabor@architektur.uni-kassel.de or to knobi@architektur.uni-kassel.de
Newsgroups: comp.sys.next.bugs From: oneill@cs.sfu.ca (Melissa O'Neill) Subject: Re: NEXTSTEP GCC bug (?) Message-ID: <1995Jul24.202649.9044@cs.sfu.ca> Summary: Aliens invate Manhattan. (Well, not really, but hey...) Keywords: cc, cpp, next, nextstep, braindamage, bugs, -traditional-cpp, aliens Organization: Simon Fraser University, Burnaby, BC, Canada References: <1995Jul20.090102.11641@improve.fdn.org> <1995Jul21.213405.6705@cs.sfu.ca> <DC51E2.EI5@basil.icce.rug.nl> Date: Mon, 24 Jul 1995 20:26:49 GMT Tom Hageman <tom@basil.icce.rug.nl> quoted (almost) my whole posting, and then added at the bottom: > Not quite. `cc -traditional-cpp' runs GNU cpp in ANSI mode. > `cc -traditional' runs GNU cpp in `traditional' pre-ANSI mode (at least > since NS3.2dev). Yeah, I realized that -traditional-cpp doesn't invoke -traditional in cpp on my NEXTSTEP system soon after I posted. I was fooled by this deviation from standard GCC behaviour, since the regular GCC manpage reads: -traditional-cpp Attempt to support some aspects of traditional C preprocessors. This includes the items that specifi- cally mention the preprocessor above, but none of the other effects of `-traditional'. I debated posting a correction, but decided to let it slide, since I was fairly sure that earlier NEXTSTEP releases did have this behaviour. In fact, it seems reasonable to say that NeXT's cc has a bug in that it should have honoured the normal GNU -traditional-cpp option and introduced a new option -gnu-cpp, or perhaps -fno-braindamage, or whatever. Ho hum, Melissa. P.S. You can easily see what your compiler does by using the -v option to cc and seeing what options it gives to /lib/cpp. --- Isn't it time you went to the bathroom? // Melissa O'Neill <oneill@cs.sfu.ca>
From: czarny@luna (Mischa Czarny) Newsgroups: comp.soft-sys.nextstep,comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.sysadmin Subject: cmsg cancel <3utr7k$s79@wally2.hti.net> Control: cancel <3utr7k$s79@wally2.hti.net> Date: 26 Jul 1995 14:15:27 GMT Organization: Institut fuer Informatik der Universitaet Muenchen Distribution: inet Message-ID: <3v5ilv$n2h@arcadia.informatik.uni-muenchen.de> cancel
From: czarny@luna (Mischa Czarny) Newsgroups: comp.soft-sys.nextstep,comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.sysadmin Subject: cmsg cancel <3utr7k$s79@wally2.hti.net> Control: cancel <3utr7k$s79@wally2.hti.net> Date: 26 Jul 1995 14:15:32 GMT Organization: Institut fuer Informatik der Universitaet Muenchen Distribution: inet Message-ID: <3v5im4$n2h@arcadia.informatik.uni-muenchen.de> cancel
From: czarny@luna (Mischa Czarny) Newsgroups: comp.soft-sys.nextstep,comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.sysadmin Subject: cmsg cancel <3utr7k$s79@wally2.hti.net> Control: cancel <3utr7k$s79@wally2.hti.net> Date: 26 Jul 1995 14:15:34 GMT Organization: Institut fuer Informatik der Universitaet Muenchen Distribution: inet Message-ID: <3v5im6$n2h@arcadia.informatik.uni-muenchen.de> cancel
From: roffel@biw-ag.de (Lee Roffel) Newsgroups: comp.sys.next.bugs Subject: Re: Root Dock reverts to Default Date: 27 Jul 1995 11:59:00 GMT Organization: BIW Systemhaus Message-ID: <3v7v24$p4u@next01.biw-ag.de> References: <3v07au$rgp@news.cs.tu-berlin.de> In article <3v07au$rgp@news.cs.tu-berlin.de> kiwi@cs.tu-berlin.de (Axel Habermann) writes: > Lee Roffel (roffel@biw-ag.de) wrote: > : Hi kids... > > : On a new NeXT 3.3 installation... logged in as root... the changes to the > : dock are not held between logins. > Maybe you're trapped by UserManager which likes to make /root > root's homedirectory. at /root > > Axel > -- This was the problem... (I just got the computer in my hands a couple of days ago... and that was what I found) Thanks for the help... ciao... .. Lee -- Lee Roffel NeXTMail: roffel@biw-ag.de
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: NSI 3.3 NFS Server crashing Date: 27 Jul 1995 13:27:10 GMT Organization: Technical University of Berlin, Germany Message-ID: <3v847e$7qn@news.cs.tu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi everyone, we got an Intel Box running 3.3 crashing regularily with the following output on console repeating several times. After this the machine simply locks up. Jul 27 14:47:03 buran mach: svckudp_send: xdr_replymsg failed Jul 27 14:47:03 buran mach: nfs_server: bad sendreply from <ip of solaris2.4 host> The interesting fact is, that these error messages always contain the ip number of one of our solaris machines running 2.4 I know there has been talk about a fix to another problem with 2.4 nfs client to NS 3.3, but nobody had mentioned that the NS3.3 server would crash on those occasions. Has anyone made any experience with this? The Server is configured as follows: NSI3.3, Pentium 90, 32 MB, Etherexpress 16, Adaptec 2940PCI Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: stephens@il.us.swissbank.com (Kurt Stephens) Subject: HELP: compiling IRC 2.8.2 on NeXTSTEP, -posix bug? Message-ID: <1995Jul27.185138.14482@il.us.swissbank.com> Keywords: posix., irc, compiling Sender: root@il.us.swissbank.com (Operator) Organization: Swiss Bank Corporation CM&T Division Date: Thu, 27 Jul 1995 18:51:38 GMT Has any one tried to build IRC II v2.8.2 (Internet Relay Chat client) on NeXTSTEP using autoconfig? The link line in the Makefile does this: cc -arch i386 -arch m68k -g -O -posix -DHAVE_CONFIG_H -o irc alias.o crypt.o ctcp.o dcc.o debug.o edit.o exec.o flood.o funny.o help.o history.o hold.o hook.o if.o ignore.o input.o irc.o ircaux.o ircsig.o keys.o lastlog.o list.o log.o mail.o menu.o names.o newio.o notice.o notify.o numbers.o output.o parse.o queue.o reg.o scandir.o screen.o server.o stack.o status.o term.o translat.o vars.o whois.o window.o -ltermcap -lsys_s -lposix ld: for architecture i386 ld: Undefined symbols: __NXArgc __NXArgv __environ Why whould these be undefined? Any ideas? Kurt A. Stephens Development Environment Engineering Swiss Bank Corp. Chicago
From: jimk@anchor.eng2.uconn.edu (JIM KOCH) Newsgroups: comp.sys.next.bugs Subject: NewsGrazer Shutdown Date: 27 Jul 1995 19:49:18 GMT Organization: The University of Connecticut Message-ID: <3v8qju$cji@orson.eng2.uconn.edu> My NewsGrazer is shutting down more often, I assume this happens either because it is seeing something it can not handle or because it is overloaded on volume. I am running 3.0-7. Is there a fix using the newsrc file? Is there a patch out there? Thanks Jim -- If your only tool is a hammer, you view all problems as nails!
From: darren@omni.cyberstore.com (Darren Reely) Newsgroups: comp.soft-sys.nextstep,comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.sysadmin Subject: Re: How to get Nextstep 3.2 to use Logitec Serial Mouse Date: 28 Jul 1995 08:04:50 GMT Organization: Cyberstore Systems, Inc. Distribution: inet Message-ID: <3va5n2$onh@sulla.cyberstore.ca> References: <3utr7k$s79@wally2.hti.net> I have some notes from the compatibility guide that may help you. Bus mouse - non serial (Logitech Mouseman), (not serial Logitech type C) Logitech Compatible Bus Mouse Setup and Installation; Set the jumper JMP1 to IRQ 5 before installing the Logitech Bus Mouse interface card. Then NEXTSTEP can recognize the presence of the device automatically. Logitec bus mice with model number M-SR14 are known not to work on some systems. M-SF14 mice (among others) are known to work. NEXTSTEP does not support the Microsoft Bus Mouse because the specifications to the mouse are proprietary. Darren @ dreely@cyberstore.com
From: esterbro@vt.edu Newsgroups: comp.sys.next.sysadmin,comp.sys.next.programmer,comp.sys.next.bugs,comp.sys.next.software,comp.sys.next.misc Subject: Can NEXTSTEP disks be repaired? (need help desparately) Date: 28 Jul 1995 14:35:00 GMT Organization: Virginia Tech, Blacksburg, Virginia Message-ID: <3vasik$o5l@server.cs.vt.edu> Introduction: A rude installation package formatted the first 3% of a SCSI NEXTSTEP disk for DOS. Of course, the disk had important files. Of course it was not completely backed up because I wasn't expecting this. Can I patch the disk somehow? Here is the detailed account for those of you who are interested: I have a Pentium machine with 1 GB EIDE drive and 1 GB SCSI drive. The EIDE has a 150 MB partition with a minimal installation of NEXTSTEP 3.3. The other 874 MB was unallocated. The SCSI had a 1 GB partition with a complete installation of NEXTSTEP 3.3 and my important files. My goal was to install DOS 6.2 to the EIDE 874 MB partition. In my computer's setup I set the EIDE to be enabled. When I do this the EIDE becomes the C: drive and the SCSI becomes D: (as the SCSI ROM BIOS reports at start up). I inserted the DOS setup disk in drive A: which successfully booted. It kindly asked me if I wanted to use the unallocated portion of the disk for DOS. I said "YES" afterwhich it formatted it. No problem there. >>> Then without any prompt or even a warning it immediately <<< >>> began formatting drive D: for DOS! I stopped it after <<< >>> three percent. <<< Now when I boot the NEXTSTEP on EIDE it tries to mount the SCSI drive. I get the panel, "Disk is unreadable - [Ignore] [Initialize]" I choose ignore and the disk is mounted as a DOS disk containing no files. I know my data is still there on the disk. I need it. Can I repair the SCSI by copying the first 30 MB of my NEXTSTEP-EIDE over the SCSI disk and set the partition ID? At work I have a 512 MB SCSI disk, if this might make a better source image. How do i-nodes factor into this? I'm a programmer with NEXTSTEP and some PC-ROM experience. But I am a disk repair novice. Any and all help is greatly appreciated. Please reply directly to esterbro@vt.edu. -Chuck Esterbrook Orca Computer, Inc.
From: Robert Worne <rworne@primenet.com> Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: Re: HELP: compiling IRC 2.8.2 on NeXTSTEP, -posix bug? Date: 28 Jul 1995 15:21:13 GMT Organization: A Big, Black Box. Message-ID: <3vav99$qs@nnrp3.primenet.com> References: <1995Jul27.185138.14482@il.us.swissbank.com> stephens@il.us.swissbank.com (Kurt Stephens) wrote: > Has any one tried to build IRC II v2.8.2 (Internet Relay Chat client) on NeXTSTEP using autoconfig? > The link line in the Makefile does this: > cc -arch i386 -arch m68k -g -O -posix -DHAVE_CONFIG_H -o irc alias.o crypt.o ctcp.o dcc.o debug.o edit.o exec.o flood.o funny.o help.o history.o hold.o hook.o if.o ignore.o input.o irc.o ircaux.o ircsig.o keys.o lastlog.o list.o log.o mail.o menu.o names.o newio.o notice.o notify.o numbers.o output.o parse.o queue.o reg.o scandir.o screen.o server.o stack.o status.o term.o translat.o vars.o whois.o window.o > -ltermcap -lsys_s -lposix > ld: for architecture i386 > ld: Undefined symbols: > __NXArgc > __NXArgv > __environ > Why whould these be undefined? Any ideas? I get a very similar error compiling GNU Pascal... nothing I could do would get rid of it. Especially when the very things it claims are undefined actually *are* defined... -- //--------------------------------------------------------------------- // Robert Worne rworne@primenet.com OS/2-NeXT -=Starving CS Undergrad=- //--------------------------------------------------------------------- // Microsoft Network is prohibited from redistributing this work in // any form, in whole or in part. Copyright, Robert Worne, 1995. // License to distribute this post is available to Microsoft for $1000. // Posting without permission constitutes an agreement to these terms. // Please send notices of violation to rworne@primenet.com and // postmaster@microsoft.com
From: tiggr@es.ele.tue.nl (Pieter Schoenmakers) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: Re: HELP: compiling IRC 2.8.2 on NeXTSTEP, -posix bug? Date: 28 Jul 1995 16:13:10 GMT Organization: Eindhoven University of Technology, The Netherlands Message-ID: <TIGGR.95Jul28181310@cobra.es.ele.tue.nl> References: <1995Jul27.185138.14482@il.us.swissbank.com> <3vav99$qs@nnrp3.primenet.com> In-reply-to: Robert Worne's message of 28 Jul 1995 15:21:13 GMT In article <3vav99$qs@nnrp3.primenet.com> Robert Worne <rworne@primenet.com> writes: > Why whould these be undefined? Any ideas? > ld: for architecture i386 > ld: Undefined symbols: > __NXArgc > __NXArgv > __environ These are exactly those symbols defined globally in the data segement of /lib/crt0.o. Strange. What does `cc -v' tell you about the arguments passed to ld? --Tiggr
From: Paul Lynch <Paul_Lynch@plsys.com> Newsgroups: comp.sys.next.bugs Subject: Re: HELP: compiling IRC 2.8.2 on NeXTSTEP, -posix bug? Date: Fri, 28 Jul 1995 21:08:46 GMT Organization: P & L Systems Sender: news@seer.demon.co.uk Message-ID: <1995Jul28.210846.21866@seer.demon.co.uk> References: <TIGGR.95Jul28181310@cobra.es.ele.tue.nl> In article <TIGGR.95Jul28181310@cobra.es.ele.tue.nl> tiggr@es.ele.tue.nl (Pieter Schoenmakers) writes: > In article <3vav99$qs@nnrp3.primenet.com> Robert Worne <rworne@primenet.com> writes: > > > Why whould these be undefined? Any ideas? > > > ld: for architecture i386 > > ld: Undefined symbols: > > __NXArgc > > __NXArgv > > __environ > > These are exactly those symbols defined globally in the data segement of > /lib/crt0.o. Strange. What does `cc -v' tell you about the arguments > passed to ld? --Tiggr Read the INSTALL notes that come with ircii. They specifically mention these errors in the context of NeXT, and tell you how to correct them. I believe that they are due to a bug in the linker when used with the posix option. Paul -- Paul Lynch (NeXTmail) http://www.plsys.co.uk/~paul (under construction)
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 31 Jul 1995 04:15:36 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <3vhld8$nae@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: im_wchaa@uxmail.ust.hk (Wong Cheuk Hang Calvin) Newsgroups: comp.sys.next.bugs Subject: Can't install NSI 3.3 Date: 30 Jul 1995 16:02:58 GMT Organization: The Hong Kong University of Science and Technology Message-ID: <3vgafi$dkm@news.ust.hk> Thanks to : -- Jerry S. Weiss j-weiss@nwu.edu Dept. Medicine, Northwestern Univ. Medical School, Chicago, Illinois %SYSTEM-S-PHALOKTARG, Phasers Locked on Target, Ready to Fire ... and ... Philip McDunnough LakeHaven , {Where sheep may safely graze...} philip@utstat.toronto.edu, {NeXT Mail Preferred} -- The problem is due to the NeXT EIDE/ATAPI 3.31 driver with my 516M harddisk, after using a SCSI cdrom and using the old " IDE driver > 504M 3.30 ", everything ( almost, because i can only access 472M ) is okay. -- Calvin
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs,comp.sys.next.hardware From: rdl@world.std.com (Robert La Ferla) Subject: More problems with the new NeXT ISASerialPort/PortServer driver Message-ID: <RDL.95Aug2211031@world.std.com> Followup-To: comp.sys.next.sysadmin Sender: rdl@world.std.com (Robert La Ferla) Organization: The World Public Access UNIX, Brookline, MA Date: Thu, 3 Aug 1995 01:10:31 GMT I am still experiencing problems with DCD/DTR using the latest NeXT serial driver. Anyone else care to post their experiences? There seems to be more and more incentive to run Mux over the NeXT driver. Why doesn't NeXT just ship Mux and pay Mark some $$$!!! Robert La Ferla Registered NEXTSTEP / OPENSTEP Consultant Boston, MA + 1 (617) 252-0088
Newsgroups: comp.sys.next.bugs Subject: help! rlogind: controlling tty missing. Message-ID: <1995Aug3.144939.1414@ittpub> From: jurgen@ittpub.nl (Jurgen Hildebrand) Date: 3 Aug 95 14:49:39 WET Distribution: world Keywords: rlogind,controlling tty For some reason rlogind (in my situation) does not provide a controlling tty due to what i have heard a wrong compilation define (DEBUG) when rlogind was compiled at NEXT. This is a known bug... Has somebody a way to get that controlling tty? Thanks +--------------------------------------------------------------+ | email: jurgen@ittpub.nl (Jurgen Hildebrand) | | AT&T PersonaLink: A5053644680@attpls.net | | work phone: +31(0)20-5676697 @ ITT Publitec, Amsterdam | +--------------------------------------------------------------+
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <e8703602@student.tuwien.ac.at> Message-ID: <9508030816.AA00312@zaphod> Content-Type: text/plain Mime-Version: 1.0 (NeXT Mail 3.3 v118.2) From: Christian Starkjohann <e8703602@student.tuwien.ac.at> Date: Thu, 3 Aug 95 10:16:26 +0200 Subject: Re: More problems with the new NeXT ISASerialPort/PortServer driver Robert La Ferla wrote: > I am still experiencing problems with DCD/DTR using the latest NeXT > serial driver. Anyone else care to post their experiences? There > seems to be more and more incentive to run Mux over the NeXT driver. > Why doesn't NeXT just ship Mux and pay Mark some $$$!!! I experience reproducable system crashes with NeXT's new driver: when an external device sends lots of data into the line during an attempt to open /dev/tty?, the system crashes. BTW: I found one incompatibility with the old driver and with MUX: the parameters of the line are reset when the line is closed. With the old driver I could do stty 19200 >/dev/ttya cat file >/dev/ttya to send out some data. With the new one I have to write (stty 19200; cat file) >/dev/ttya. And one more thing: every open needs some seconds. What does the driver do during this time? NeXT's driver was a nice experiment, but I will switch back to MUX. -- Christian Starkjohann <cs@ds1.kph.tuwien.ac.at>
From: stephen@genesis1.physics.yale.edu (Stephen B. Selipsky) Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.sysadmin Subject: HP LaserJet III Printer connection help Followup-To: comp.sys.next.sysadmin Date: 3 Aug 1995 18:09:37 GMT Organization: Yale University, Department of Computer Science, New Haven, CT Message-ID: <3vr3d2$j9c@babyblue.cs.yale.edu> Has anyone managed to connect an HP LaserJet III Postscript Cartridge printer to an Intel-based NEXTSTEP 3.2 system? The printer works fine when connected to an Apple, but I've had no success with either a 486-based Intel GX Pro, or a Pentium-based Eclipse. The printer shows no sign of receiving any print files; under one config (Parallel port ?) it did give a spurious "out of paper" message, but no more. I've tried: serial port (A or B, baud rate 9600, with or without hardware handshaking); parallel port (Parallel 1 Port); replacing the ppd file with the new version 4.1 from Adobe (the "DOS" version, as recommended by NextAnswers; renamed /NextLibrary/PrinterTypes/HP_LaserJet_III_PostScript_Cartridge_v4.1.ppd it showed up fine on the PrintManager menu). Thanks for any advice!
From: mitroo@magnus.acs.ohio-state.edu (Varun Mitroo) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.programmer,comp.sys.next.bugs,comp.sys.next.software,comp.sys.next.misc Subject: Re: Can NEXTSTEP disks be repaired? (need help desparately) Date: 3 Aug 1995 19:40:50 GMT Organization: The Ohio State University Message-ID: <3vr8o2$de7@charm.magnus.acs.ohio-state.edu> References: <3vasik$o5l@server.cs.vt.edu> In article <3vasik$o5l@server.cs.vt.edu>, <esterbro@vt.edu> wrote: >Introduction: A rude installation package formatted the first 3% of a SCSI >NEXTSTEP disk for DOS. Of course, the disk had important files. Of course >it was not completely backed up because I wasn't expecting this. > >Can I patch the disk somehow? > >Here is the detailed account for those of you who are interested: > >I have a Pentium machine with 1 GB EIDE drive and 1 GB SCSI drive. > >The EIDE has a 150 MB partition with a minimal installation of NEXTSTEP >3.3. The other 874 MB was unallocated. The SCSI had a 1 GB partition with a >complete installation of NEXTSTEP 3.3 and my important files. > >My goal was to install DOS 6.2 to the EIDE 874 MB partition. > >In my computer's setup I set the EIDE to be enabled. When I do this the >EIDE becomes the C: drive and the SCSI becomes D: (as the SCSI ROM BIOS >reports at start up). I inserted the DOS setup disk in drive A: which >successfully booted. It kindly asked me if I wanted to use the unallocated >portion of the disk for DOS. I said "YES" afterwhich it formatted it. No >problem there. > >>>> Then without any prompt or even a warning it immediately <<< >>>> began formatting drive D: for DOS! I stopped it after <<< >>>> three percent. <<< > >Now when I boot the NEXTSTEP on EIDE it tries to mount the SCSI drive. I >get the panel, "Disk is unreadable - [Ignore] [Initialize]" I choose ignore >and the disk is mounted as a DOS disk containing no files. > >I know my data is still there on the disk. I need it. > >Can I repair the SCSI by copying the first 30 MB of my NEXTSTEP-EIDE over >the SCSI disk and set the partition ID? At work I have a 512 MB SCSI disk, >if this might make a better source image. > >How do i-nodes factor into this? > >I'm a programmer with NEXTSTEP and some PC-ROM experience. But I am a disk >repair novice. > >Any and all help is greatly appreciated. Please reply directly to >esterbro@vt.edu. > >-Chuck Esterbrook > Orca Computer, Inc. ---------------------------- I have had exactly the same experience as this! I was building a DOS partition on my boot drive about one month ago. I created the DOS partition and then rebooted the computer with the DOS floppy disk. The setup program started and offered to format my disk and install dos. MS-DOS 6.22 formatted my 100 meg dos partition, and then begain immediately formatting my 1 gig NEXTSTEP partition. I saw this and stopped it after about 2-3%. I was livid, and made some phone calls to Microsoft demanding that they pay to recover my lost data. I finally got connected to someone at tech support at Microsoft who made the brilliant suggestion "can't you just run unformat on the partition?". I had to pay $25.00 to hear this and other similar calibre suggestions. It turns out that Microsoft says that my dos 6.22 disks were an OEM product with the company that I bought my computer from (Packard Bell), and Microsoft said that Packard Bell had modified the boot floppy for dos. They also said that THEIR dos disk would not automatically format a second hard drive. I didn't believe it, and when I talked to Packard Bell, their tech support said that they really don't make any big changes to dos setup - just change some parameters. They said it must be a problem with Microsoft. At that point, their only suggestion was to write a letter to Packard Bell explaining what happened. Packard Bell said that for most people, this would not be a problem since they only have one drive, and they use the entire drive for DOS. I tried many things to recover my data. I tried building another disk with NEXTSTEP, and then manually mounting my damaged disk. It could not mount the disk. I then deleted the DOS partition and remade the NEXTSTEP partition using fdisk. I still could not mount the disk. I tried running the "disk" utility with the raw SCSI device and searching for backup boot blocks and running fsck with the alternate boot blocks. Nothing worked, and eventually I had to reformat the disk. I did lose some important data from the drive, and I'm being much more careful since then with my data. I now have my main NEXTSTEP files on a disk with 1024 byte sectors, so it isn't even recognised by DOS! The setup program is a very poorly handled install program by Microsoft, with total ignorance of any other operating system. It's disgraceful that a company with such an inferior product has become such a dominating position in the computer world. Varun
From: lamdong@cory.Berkeley.EDU (LD) Newsgroups: comp.sys.next.bugs Subject: DPS client library error Date: 3 Aug 1995 21:49:13 GMT Organization: University of California, at Berkeley Message-ID: <3vrg8p$7de@agate.berkeley.edu> Hi, How to make connection to WorkSpace manager from rc.local file ? I can run my NeXT app in my workspace (either click on the icon, or run from a terminal, but when I put my program in rc.local file, it can't run and there's error "DPS client library error: Could not form connection, host local host " I appreciate your help. - Lam
From: msb2@doc.ic.ac.uk (M S Bennett) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: -posix bug in crt0 (was: Compiling ircII) Date: 4 Aug 1995 11:36:52 +0100 Organization: Department of Computing, Imperial College, University of London, UK. Message-ID: <3vst84$n1r@fuji.doc.ic.ac.uk> References: <1995Jul27.185138.14482@il.us.swissbank.com> <3vav99$qs@nnrp3.primenet.com> <TIGGR.95Jul28181310@cobra.es.ele.tue.nl> Ok I have built ssh (secure shell) fine - but when I try and run it - I get 'Floating Exception' and gdb says that the program is dying in 'NXCreateZone' in crt0 - called from malloc_init - from malloc etc.. I'm using gcc 2.6.3 and NS 3.3 and Developer 3.2 Any ideas? Sean Bennett PS - E-mail with a response as well - I should get this News feed, but I might be distracted ;-) -- Sean Bennett |sean@world.std.com/msb2@doc.ic.ac.uk Earth Resources|+44(0)181 959 2702 <- Parents Engineering |+44(0)181 789 8102 <-Flat +44(0)171 594 7404 <- Work
Newsgroups: comp.sys.next.bugs From: franc@xenicos.fdn.fr (Francois BIENTZ) Subject: Compaq XE , DPS errors Message-ID: <1995Aug4.221744.1470@xenicos.fdn.fr> Sender: franc@xenicos.fdn.fr Organization: Individual. Date: Fri, 4 Aug 1995 22:17:44 GMT Upgraded a Compaq Deskpro XE33S with a DX4/100 overdrive. RAM :24Mb, 2Mb VRAM, no cache card, 525 Mb Hitashi IDE HD. display 800X600 60hz. the Window server exits offten : on console can read : dps_err_write last message read 20 times exiting due to excessive DPS errors. WM[201]: DPS client library error:Error while writing to connection, DPSContext aef54, data -102 use someone the XE s ???
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: csaldanh@mae.carleton.ca (Chris Saldanha) Subject: Re: -posix bug in crt0 (was: Compiling ircII) Message-ID: <DCuMDo.L26@cunews.carleton.ca> Followup-To: comp.sys.next.bugs,comp.sys.next.programmer Sender: news@cunews.carleton.ca (News Administrator) Organization: computerActive Inc References: <1995Jul27.185138.14482@il.us.swissbank.com> <3vav99$qs@nnrp3.primenet.com> <TIGGR.95Jul28181310@cobra.es.ele.tue.nl> <3vst84$n1r@fuji.doc.ic.ac.uk> Date: Sat, 5 Aug 1995 17:39:24 GMT M S Bennett (msb2@doc.ic.ac.uk) wrote: : Ok I have built ssh (secure shell) fine - but when I try and run it - I get : 'Floating Exception' and gdb says that the program is dying in : 'NXCreateZone' in crt0 - called from malloc_init - from malloc etc.. : I'm using gcc 2.6.3 and NS 3.3 and Developer 3.2 What flags did you give to the compiler (or did the Makefile give) when building? If it's POSIX source-code, then you want to give -posix always, and -lposix when linking. We just went through all this stupidity of POSIX when building SLiRP. My first builds of SLiRP, which used -lposix, but NOT -posix gave EXACTLY the same errors that you cite. The intuitive "Floating Exception", and this NXCreateZone stuff when debugged. The autoconf-generated "./configure" script in SLiRP ran POSIX source through cc without -posix, so it couldn't get proper values for sizeof(), couldn't tell if certain header files were POSIX-compliant, didn't know if the machine was big-endian or not, etc, etc, etc... I don't know anything about ssh, but if it uses a GNU autoconf-generated configure script, it's probably filled the configure output with lies. --Chris Chris Saldanha, Software Analyst -------------------------------------- computerActive, Inc |"The telephone was not invented by | chris@computerActive.on.ca (NeXTMail) | Alexander Graham Unitel" -Bell Ad | http://www.mae.carleton.ca/~csaldanh --------------------------------------
From: kay@kauri.vuw.ac.nz (Cameron Kay) Newsgroups: comp.sys.next.bugs Subject: problems with the icon dock when logged in as root Date: 06 Aug 1995 21:41:15 GMT Organization: Victoria Uni. of Wellington, NZ. Distribution: world Message-ID: <KAY.95Aug7094115@kauri.vuw.ac.nz> I seem to be having problems with the icon dock when logged as root. For some strange reason adding applications to the dock or setting existing ones to auto boot at login time are not being saved. Every time I login a root the dock reverts back to its default configuration. Anyone have any idea how to fix this short of reinstalling the system? Thanks in advance - Cameron -- Email Cameron.Kay@kauri.vuw.ac.nz Post Computer Science Department Phone + 64 4 472 1000 x7032 (Work) Victoria University + 64 4 237 5895 (Home) P.O.Box 600 Fax + 64 4 495 5232 Wellington, New Zealand
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: problems with the icon dock when logged in as root Date: 7 Aug 1995 10:31:17 GMT Organization: Technical University of Berlin, Germany Distribution: world Message-ID: <404q1l$nlo@news.cs.tu-berlin.de> References: <KAY.95Aug7094115@kauri.vuw.ac.nz> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Cameron Kay (kay@kauri.vuw.ac.nz) wrote: : I seem to be having problems with the icon dock when logged as : root. For some strange reason adding applications to the dock or : setting existing ones to auto boot at login time are not being : saved. Every time I login a root the dock reverts back to its default : configuration. Anyone have any idea how to fix this short of : reinstalling the system? This should be put in the FAQ. UserManager likes to revert root's homedirectory to /root, were it doesn't exist in the standard installation. So Workspace doesn't know where to write the defaults to. Use NetInfoManager to set root's home to / or create a new home for root at /root. Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: tmunson@cwis.unomaha.edu (todd s. munson) Newsgroups: comp.sys.next.hardware,comp.sys.next.bugs Subject: DPT 2024 SCSI Card Problems Date: 6 Aug 1995 14:56:55 GMT Organization: University of Nebraska Omaha Message-ID: <402l7n$84q@s-cwis.unomaha.edu> I have been trying to install NEXTSTEP on a Dell Dimension XPS P90 system that has a DPT2024 PCI SCSI card. I have gone to Next Answers and obtained the DPT SCSI driver (Version 3.32). I have a NEC Quad Speed CD-ROM drive connected to the SCSI card. I am trying to install NEXTSTEP to a 618 MB partition of an IDE drive. (I am using the IDE driver for large systems). Everything seems to work for a while. We are able to start the installation and files are starting to transfer. Something stops the process at the same place every time I try to install. (About a quarter of the way into the install.) I have checked the termination on the CD-ROM and SCSI Card. Both are terminated. I have tried numerous configurations for the SCSI Card, all to no avail. I have also changed the SCSI cable hoping that I had a cheesy SCSI Cable. The specific error is an EATA Timeout. It then tries to reset the SCSI bus and call attention to the CD-ROM. It does this twice and then decides to halt the machine. Are there any known problems with the DPT 2x24 Driver that would cause this type of problem? Is there anything else that could be causing the error? Thanks for your help, Todd S. Munson -- tmunson@cwis.unomaha.edu Todd S. Munson, UNO ACM President
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.hardware,comp.sys.next.bugs Subject: Re: DPT 2024 SCSI Card Problems Date: 7 Aug 1995 16:27:57 GMT Organization: Express Access Online Communications, USA Message-ID: <405eud$4bj@news4.digex.net> References: <402l7n$84q@s-cwis.unomaha.edu> tmunson@cwis.unomaha.edu (todd s. munson) wrote: > I have been trying to install NEXTSTEP on a Dell Dimension XPS > P90 system that has a DPT2024 PCI SCSI card. I have gone to Next > Answers and obtained the DPT SCSI driver (Version 3.32). I have > a NEC Quad Speed CD-ROM drive connected to the SCSI card. I don't exactly know what the problem is, but my DPT works just fine. However, here are some tips. Set the card to Level (Auto) under your AMI bios (or whatever system bios you have) and to PCI IRQ 9 or 10 (it seems to like those best. The vanilla DPT bios settings seem to work just fine, so just restore the card to the factory defaults. The problem of which you speak of seems to be a command queue problem, or a reset problem. I think the DPT bios has settings for both of these. If the problem persits, try turning them off. Also, I would try giving DPT a call. They actually know about NS and know how to fix associated problems. Good luck, -- Thanks, Later, John monoChrome, Inc. | New York Law School | NEXTSTEP Developer John Kheit )^> %^) | Opinions expressed | MIME & NeXTmail OK--3P jkheit@cnj.digex.net | represent me only... | Telepathy...It's coming...
From: bestor@spam.cs.wisc.edu (Gareth Bestor) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: bug in NS3.3 tar? Date: 7 Aug 1995 20:28:48 GMT Organization: U of Wisconsin CS Dept Message-ID: <405t20$879@spool.cs.wisc.edu> Perhaps this is a FAQ bug, but is there something screwed up with the tar shipped with NEXTSTEP 3.3 (either Intel or Moto)? I have a NeXTstation with a Hewlett-Packard SCSI DAT drive and it cannot correctly write tar files to the tapes. I can create a tar tape file with "tar -cvf /dev/rst0" and write it to the /dev/rst0 device no problem. This can be read with "tar -xv" and also listed with "tar -tv" no problem too. However, any attempt to _append_ to this tape file with "tar -rv" gives the error: > tar -rvf /dev/rst0 foo* tar: tape backspace error: I/O error This occurs with three different DAT drives and also happens on an Intel NEXTSTEP machine. I even tried using different physical tapes with no effect. The killer is that I tried this with the gnutar (which NEXT ships with 3.3) and it works GREAT!!! So it seems pretty obvious to me that something is screwed up with NeXT's tar. I'm surprised because something as major as not being able to append to tar tape backup files should have been spotted and fixed long ago. Is this a bug in tar or is it just me? - Gareth bestor@cs.wisc.edu
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin From: rdl@world.std.com (Robert La Ferla) Subject: Re: bug in NS3.3 tar? In-Reply-To: bestor@spam.cs.wisc.edu's message of 7 Aug 1995 20:28:48 GMT Message-ID: <RDL.95Aug7215418@world.std.com> Sender: rdl@world.std.com (Robert La Ferla) Organization: The World Public Access UNIX, Brookline, MA References: <405t20$879@spool.cs.wisc.edu> Date: Tue, 8 Aug 1995 01:54:18 GMT Try /dev/nrst0 instead. Robert La Ferla Registered NEXTSTEP / OPENSTEP Consultant Boston, MA + 1 (617) 252-0088 In article <405t20$879@spool.cs.wisc.edu> bestor@spam.cs.wisc.edu (Gareth Bestor) writes: Perhaps this is a FAQ bug, but is there something screwed up with the tar shipped with NEXTSTEP 3.3 (either Intel or Moto)? I have a NeXTstation with a Hewlett-Packard SCSI DAT drive and it cannot correctly write tar files to the tapes. I can create a tar tape file with "tar -cvf /dev/rst0" and write it to the /dev/rst0 device no problem. This can be read with "tar -xv" and also listed with "tar -tv" no problem too. However, any attempt to _append_ to this tape file with "tar -rv" gives the error: > tar -rvf /dev/rst0 foo* tar: tape backspace error: I/O error This occurs with three different DAT drives and also happens on an Intel NEXTSTEP machine. I even tried using different physical tapes with no effect. The killer is that I tried this with the gnutar (which NEXT ships with 3.3) and it works GREAT!!! So it seems pretty obvious to me that something is screwed up with NeXT's tar. I'm surprised because something as major as not being able to append to tar tape backup files should have been spotted and fixed long ago. Is this a bug in tar or is it just me? - Gareth bestor@cs.wisc.edu
From: Rex Dieter <rdieter@math.unl.edu> Newsgroups: comp.sys.next.bugs Subject: SMCUltraNetworkDriver Problems Date: 8 Aug 1995 14:57:42 GMT Organization: University of Nebraska--Lincoln Message-ID: <407u16$h2r@crcnis3.unl.edu> We have been experiencing problems trying to use the relatively new ethernet driver (the SMCUltraNetworkDriver, v3.31) from NEXTANSWERS with an SMC Elite Ultra ISA Ethernet Adapter. The card was configured as: I/O base: 280 IRQ: 10 RAM Base D0000 Wait States: YES (I tried NO wait states, but the card doesn't recognize any connections in or out ) The card and system (a 486DX2-66 VLB) appear to work fine, but only after a small amount of time, odd file system corruption occurs. I've tried the card in two machines, with two different hard drives and SCSI controllers (each AHA 1542CF's); and the file system errors appear in as little as 30 minutes of light use (and as much as 4 days). The system usually panics with 'freeing i-node' errors, and fsck's on reboot clear out a lot of unlinked files, etc.... It seems fine for awhile, and then... the cycle repeats. I've reported this problem via BugNeXT, and it has been assigned a log number of 72136. Is anyone else using this adapter successfully? Any suggestions? --- Rex A. Dieter rdieter@math.unl.edu (NeXT/MIME) Research Associate Voice: (402)472-9747 Department of Mathematics and Statistics FAX: (402)472-8466 University of Nebraska - Lincoln http://www.math.unl.edu/~rdieter/
From: root@dpls.dacc.wisc.edu (Operator) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: Re: bug in NS3.3 tar? Date: 8 Aug 1995 15:06:26 GMT Organization: University of Wisconsin, Madison Distribution: world Message-ID: <407uhi$go0@news.doit.wisc.edu> References: <RDL.95Aug7215418@world.std.com> In article <RDL.95Aug7215418@world.std.com> rdl@world.std.com (Robert La Ferla) writes: > Try /dev/nrst0 instead. > Tried that and it doesn't help. Here's what happens with tar: dpls2> tar cvf /dev/nrst0 bestor <--- Create new tape... a bestor/.NeXT/.NeXTdefaults.D 1 block <--- OK .. a bestor/Changes 11 blocks a bestor/.forward 1 block dpls2> tar rvf /dev/nrst0 rice <--- Try to append to it... tar: tape read error: I/O error <--- fails (not rewound?) dpls2> tar rvf /dev/rst0 rice <--- Try again... tar: tape read error: I/O error <--- fails (but rewinds) dpls2> tar rvf /dev/nrst0 rice <--- Try again after rewound... tar: tape backspace error: I/O error <--- still fails dpls2> tar rvf /dev/rst0 rice <--- Try again with rst0... tar: tape backspace error: I/O error <--- still fails Whereas if I try EXACTLY the same thing with gnutar on the same computer, same drive and same tape it works like a charm... dpls2> gnutar cvf /dev/rst0 bestor <--- Create new tape... bestor/ <--- OK bestor/.NeXT/ bestor/.NeXT/.NeXTdefaults.D .. bestor/Changes bestor/.forward bestor/.elm/ bestor/Mail/ dpls2> gnutar rvf /dev/rst0 haley <--- Append to it haley/ <--- OK haley/%backup%~ haley/.NeXT/ .. haley/docs/ haley/docs/pums80a.cbk Methinks tar is screwy. - Gareth bestor@cs.wisc.edu
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 7 Aug 1995 04:15:13 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <40440h$480@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: helge@haddock.cd.chalmers.se (Markus H|rnkvist) Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Symbios53c810 driver failure Date: 09 Aug 1995 00:19:30 GMT Organization: Chalmers University of Technology Distribution: world Message-ID: <HELGE.95Aug9021930@haddock.cd.chalmers.se> Recently I got an old harddrive, a Maxtor 7120S, to use as a boot and swap device. However, the otherwise excellent driver from Symbios Logic fails with this drive. When searching for devices on the SCSI chain it reports "INQUIERY FAILED REQUEST SENSE ERROR 2" instead of reporting what kind of harddrive it finds. After that there is no way to access the harddrive. However, the drive works with the driver from Talus, but that driver is awful in comparison. For example the SCSI bus locks up while the tape drive rewinds, when a CD is inserted, when one drive is up to something, etc. I guess some feature of the SCSI protocol is not implemented. Anyhow, does anyone know of a workaround for this? I have been trying to get in touch with Symbios, but to no avail. /Markus
From: tfs@gravity.science.gmu.edu ( Tim) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.programmer,comp.sys.next.bugs,comp.sys.next.software,comp.sys.next.misc Subject: Re: Can NEXTSTEP disks be repaired? (need help desparately) Followup-To: comp.sys.next.bugs Date: 6 Aug 1995 10:13:37 GMT Organization: George Mason University, Fairfax Va. Sender: tfs@gravity.science.gmu.edu Message-ID: <4024kh$654@portal.gmu.edu> References: <3vasik$o5l@server.cs.vt.edu> <OTTO.95Jul31155259@tukki.jyu.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Summary: use disk (eww) The "disk" program has a non-destructive option to search for backup, or alternate blocks. It's the best option to start the recovery process... I will refrain from ranting about "disk"'s other grevious faults, NeXT hasn't fixed them, even thoug they've known for years what they are and I doubt they ever will... (and if you're at NeXT going "I donut know, no one told me" mail me and I'll elaborate in a nice way.) Tim -- ________________________________________________________________ tfs@vampire.science.gmu.edu (NeXTmail) Tim Scanlon George mason University Public Affairs I speak for myself, but often claim demonic posession
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <flexus!rfschtkt@maze.ruca.ua.ac.be> Date: Wed, 9 Aug 95 20:36:06 +0200 From: Raf Schietekat <flexus!rfschtkt@maze.ruca.ua.ac.be> Message-ID: <9508091836.AA03652@flexus> Subject: New KBNS (``Known Bugs in NextStep'') edition (33.0). LS, ftp://ftp.cs.orst.edu/software/NeXT/documents/KBNS.33.0.README.rtf.(gz/Z) ftp://ftp.cs.orst.edu/software/NeXT/documents/KBNS.33.0.rtf.(gz/Z) (Some verification code in same location, KBNS* something.) This is to let you know about the new edition 33.0 of KBNS (``Known Bugs in NextStep''), the *only* public and aspiring-to-be comprehensive database of problems and bugs (with workarounds, where available!) in NeXT's products (the file's name is only historical), to be used together with NeXT's NeXTanswers. It's the first edition mentioning NEXTSTEP 3.3 at all, and there are a few things about OpenStep and EOF as well. So fetch the document, use it, and send me neat&concise reports about whatever is still missing (a new problem, a good workaround), about changes to applicability ranges, about a verification you did, if you volunteer to referee items in certain areas, about any mistakes, or a note about what you think of KBNS (if you think it is no good, I won't know unless you tell me, but please be specific; if you've been able to benefit from it, I'll be very happy to learn of it!). If an existing KBNS entry describes a problem you are having, do send a copy to Bug_NeXT! Perhaps NeXT will get bored and cure some things. KBNS is totally dependent on your cooperation. I see my role as an administrator and moderator for compiling the collective knowledge of the NEXTSTEP community (about bugs, of course). It's up to YOU! (The KBNS file itself is rather hefty (do an ls and/or fetch the README first). If anyone knows a free program to distribute patches to RTF files (GNU diff/patch or any other line-based algorithm won't do), please tell me.) Raf Schietekat, RfSchtkt@maze.ruca.ua.ac.be (NeXTmail), Flanders, Belgium If I don't answer: my mail relay can't handle !, % or .uucp, I think *** The year 2000 will be the last year of the 20th century. ***
From: root@terra.crystalengine.com(Felipe A. Rodriguez) Newsgroups: comp.sys.next.bugs Subject: lookupd Date: 11 Aug 1995 05:30:58 GMT Organization: Network Intensive Message-ID: <40epui$n8s@ni1.ni.net> Its is well known that lookupd can hang NS3.3 until it timesout if it does not recieve a proper reply from a DNS server. When is a fix due? If NeXT can't or won't fix this would someone send me the source so that I can patch it? -- Felipe A. Rodriguez # ...it cannot be called ingenuity to kill Agoura Hills, CA # one's fellow citizens, to betray # friends, to be without faith, without root@terra.crystalengine.com # mercy, without religion; by these means (NeXTmail preferred) # one can acquire power but not glory. (MIMEmail welcome) # --Nicolo Machiavelli
From: Gerald Wildgruber Newsgroups: comp.sys.next.bugs Subject: Stubborn Assignement of Icons to Document Types ?? Date: 11 Aug 1995 14:21:53 GMT Organization: Leibniz-Rechenzentrum, Muenchen (Germany) Distribution: world Message-ID: <40fp21$b5e@sparcserver.lrz-muenchen.de> Hello everybody, running NeXTStep 3.3 on Intel Architecture I met the following problem: what is it for an obscure instance to decide which Application Icon is assigned to a particular document type in the Workspace Manager ? All my html Documents appear under the Icon of an application that left my computer already long ago; the html documents are opened correctly by OmniWeb, but they always appear with the Icon of eText, I installed some months before. The problem is not to link document types and applications, but only the stubborness of some former icons in the WM. Does anyone know how to fix this ? Help would be greatly appreciated. Gerald.
From: peter@bert.psyc.upei.ca Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Aaarrrggh! Root's WorkSpace keeps crashing! Date: 11 Aug 1995 15:35:01 GMT Organization: University of Prince Edward Island, Charlottetown, PEI Canada Message-ID: <40ftb5$jg9@atlas.cs.upei.ca> Yesterday, I was installing a package off a CD-ROM and the installation failed. Ever since, WorkSpace has started crashing whenever I'm logged in as root. The WorkSpace only crashes when I'm looking in the LocalApps directory, in Icon or Listing view mode, and when I scroll to a particular file. I have tried resetting all of the configuration info by renaming /.NeXT from another account, but it still happens. What's wrong here? Peter -- Peter 'Beaker' Burka / GCS d--- h---- s+ g+ p? au a- w+ v++ C++ UL++++/X+++/ Prince Edward Island \ O++ P+ L+>++ 3 N++ K++ W++/--- M- V-\ po-- Y+ t+ 5- pburka@upei.ca / v b+++ D++ b- e+(*) u--- h* f- r- n- y-@ j++ r-- "If only we were weiner dogs our problems would be all solved"
From: doyle@mmm.lanl.gov (Mark Doyle) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Aaarrrggh! Root's WorkSpace keeps crashing! Date: 11 Aug 1995 17:12:26 GMT Organization: Los Alamos National Laboratory Message-ID: <40g31q$g00@newshost.lanl.gov> References: <40ftb5$jg9@atlas.cs.upei.ca> peter@bert.psyc.upei.ca wrote: > The WorkSpace only crashes when I'm looking in the > LocalApps directory, in Icon or Listing view mode, and when I scroll to > a particular file. > > I have tried resetting all of the configuration info by renaming /.NeXT > from another account, but it still happens. > > What's wrong here? I had a similar problem once. The WS crashed for root when I selected a directory in the File Viewer in Browser mode. It was because the executable permissions weren't correct on a directory (in this case, app wrapper I suppose) that was on an NFS mounted filesystem. See if the directory is world executable (I don't remember if read/write permissions were involved as well). Cheers, Mark
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Aaarrrggh! Root's WorkSpace keeps crashing! Date: 11 Aug 1995 19:01:23 GMT Organization: Express Access Online Communications, USA Message-ID: <40g9e3$6v0@news4.digex.net> References: <40ftb5$jg9@atlas.cs.upei.ca> peter@bert.psyc.upei.ca wrote: > Yesterday, I was installing a package off a CD-ROM and the > installation failed. Ever since, WorkSpace has started crashing > whenever I'm logged in as root. The WorkSpace only crashes when > I'm looking in the LocalApps directory, in Icon or Listing view > mode, and when I scroll to a particular file. > I have tried resetting all of the configuration info by renaming > /.NeXT from another account, but it still happens. When you heart-transplanted the .NeXT directory, did you also chown it to root? If root cannot write to the .NeXT directory, it will get angry )-> -- Thanks, Later, John monoChrome, Inc. | New York Law School | NEXTSTEP Developer John Kheit )^> %^) | Opinions expressed | MIME & NeXTmail OK--3P jkheit@cnj.digex.net | represent me only... | Telepathy...It's coming...
From: yoda@cis.uni-muenchen.de (Marc Guenther) Newsgroups: comp.sys.next.bugs Subject: Re: lookupd Date: 12 Aug 1995 19:33:33 GMT Organization: Institut fuer Informatik der Universitaet Muenchen Message-ID: <40ivmd$b9c@arcadia.informatik.uni-muenchen.de> References: <40epui$n8s@ni1.ni.net> root@terra.crystalengine.com(Felipe A. Rodriguez) wrote: > Its is well known that lookupd can hang NS3.3 until it timesout if it does > not recieve a proper reply from a DNS server. It is well known that lookupd hangs your machine, whenever it wants to. > When is a fix due? Fix ? What fix ? NeXT doesn't consider this a problem ! > If NeXT can't or won't fix this would someone send me the source so that I > can patch it? Haha ! Hehe ! Hihi ! Ho !!! *ROTFL* That's a good one... Marc
From: tm8025a@newssrv.soc.american.edu (Torrey McMahon) Newsgroups: comp.sys.next.bugs Subject: Application.rtf Date: 13 Aug 1995 00:48:01 GMT Organization: University of Wisconsin, Madison Message-ID: <40ji41$k5c@news.doit.wisc.edu> I suppose someone has already noticed this but in the on-line developer documentation NeXT messed up. It appears that before distribution the Application.rtf file in the General Reference was copied over with the ActionCell.rtf in the same section. (/NextLibrary/Documentation/NextDev/GeneralRef/02_ApplicationKit/Classes) Does anyone have this file new or slightly new? I am interested in it greatly since it details some important things. NeXTMail to the following address would be greatful. -- Torrey McMahon tm8025a@american.edu
From: root@terra.crystalengine.com(Felipe A. Rodriguez) Newsgroups: comp.sys.next.bugs Subject: Re: lookupd Date: 13 Aug 1995 01:02:22 GMT Organization: Network Intensive Message-ID: <40jiuu$9v3@ni1.ni.net> References: <40ivmd$b9c@arcadia.informatik.uni-muenchen.de> In article <40ivmd$b9c@arcadia.informatik.uni-muenchen.de> yoda@cis.uni-muenchen.de (Marc Guenther) writes: >root@terra.crystalengine.com(Felipe A. Rodriguez) wrote: > >> Its is well known that lookupd can hang NS3.3 until it timesout if it does >> not recieve a proper reply from a DNS server. > >It is well known that lookupd hangs your machine, whenever it wants to. > >> When is a fix due? > >Fix ? What fix ? NeXT doesn't consider this a problem ! > >> If NeXT can't or won't fix this would someone send me the source so that I >> can patch it? > >Haha ! Hehe ! Hihi ! Ho !!! > >*ROTFL* > >That's a good one... > >Marc Please don't dispel my fantasies that someone in Redwood City is listening to my plea's and will soon help. ;-) -- Felipe A. Rodriguez # ...it cannot be called ingenuity to kill Agoura Hills, CA # one's fellow citizens, to betray # friends, to be without faith, without root@terra.crystalengine.com # mercy, without religion; by these means (NeXTmail preferred) # one can acquire power but not glory. (MIMEmail welcome) # --Nicolo Machiavelli
From: dazevedo@telepac.pt (DAzevedo) Newsgroups: comp.sys.next.bugs Subject: Dutch NeXT Company Date: 13 Aug 1995 18:23:52 GMT Organization: Your Organization Message-ID: <40lfvo$4rn@vivaldi.telepac.pt> Mime-Version: 1.0 From dazevedo@telepac.pt Sun 13 Aug 95 18:52:24 Path: news.telepac.pt!usenet From: dazevedo@telepac.pt (DAzevedo) Newsgroups: comp.sys.next.programmer Subject: Dutch NeXT Company Date: 13 Aug 1995 18:04:12 GMT Organization: Your Organization Lines: 3 Message-ID: <40leqs$4rn@vivaldi.telepac.pt> NNTP-Posting-Host: lis3_p12.telepac.pt Mime-Version: 1.0 X-Newsreader: WinVN 0.93.10 Does any one know the name and fax number of the Dutch Company, that are saling Next equipment and components?
From: benb@aitken.velocity.com (Ben Black) Newsgroups: comp.sys.next.bugs Subject: Re: lookupd Date: 11 Aug 1995 18:45:25 GMT Organization: Minnesota Regional Network (MRNet) Message-ID: <40g8g5$85g@news.mr.net> References: <40epui$n8s@ni1.ni.net> >Its is well known that lookupd can hang NS3.3 until it timesout if it does >not recieve a proper reply from a DNS server. > >When is a fix due? > i guess it will show up in 4.0. premium support customers have already gotten about 10MB (compressed) of patched programs for 3.3 user. raise your hand if you think it is ridiculous that you have to pay an extra $50k+ just to get decent support for an $800 OS. >If NeXT can't or won't fix this would someone send me the source so that I >can patch it? hahahahahahahahaha....source? from next? hahahahahahahahah ben
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 14 Aug 1995 04:15:12 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <40mikg$ecj@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
Newsgroups: comp.sys.next.bugs From: wiemer@hrz.uni-bielefeld.de ( Sebastian Wiemer) Subject: Edit.app bug Sender: news@hermes.hrz.uni-bielefeld.de (News Administrator) Message-ID: <DDAvtu.9DM@hermes.hrz.uni-bielefeld.de> Date: Mon, 14 Aug 1995 12:25:06 GMT Organization: Universitaet Bielefeld, Rechenzentrum
From: mueller@sonja.imsd.uni-mainz.DE (Robert Mueller) Newsgroups: comp.sys.next.bugs Subject: 3.2/3.3 NextStep-cluster problems Date: 15 Aug 1995 14:37:33 GMT Organization: Johannes Gutenberg-Universität Mainz, Germany Message-ID: <40qbfd$uu2@kralle.zdv.Uni-Mainz.DE> Hallo, we have a NeXT-Cluster with 6 maschines running on NeXTStep 3.2 (only InterfaceBuilder is 3.3): four m68k and two Intels. We then got a new Intel machine with NeXTStep 3.3. (with Tricom 3C 509 Combo-Isa Adapter Network Card), the other Intel maschines have PC Intel Ether Express Network Cards), and integrated it into the cluster. Most of the network requests from this new machine to other machines of the Cluster fail (up to system chrash) or take a very long time. So the question is: Is it a known problem that 3.3/3.2 maschines mixed in a cluster have network problems. Or do we have the wrong network cards? Any pointers are welcome. Thank you very much in advance.... Robert -- ___________________________________________________________ Robert Mueller Tel. (+49) (0) 6131 17 20 22 Medizinische Informatik Fax. (+49) (0) 6131 17 29 68 (Institute Fax) Projektgruppe "TheMPO" Institut fuer Medizinische Statistik und Dokumentation Universitaetskliniken Mainz
Newsgroups: comp.sys.next.bugs From: gelato@netcom.com (Steve James) Subject: Do this program fail on 3.3/Intel??? Message-ID: <gelatoDDDsM2.5EL@netcom.com> Organization: NETCOM On-line Communication Services (408 261-4700 guest) Date: Wed, 16 Aug 1995 02:08:26 GMT Sender: gelato@netcom20.netcom.com The following program goes into an infinite loop on 3.2/Intel (but not on other 3.2 platforms). Can someone with a 3.3/Intel system compile it and let me know if it fails under 3.3 as well? Thanks. #include <math.h> void main() { printf("%g\n",fmod(1283719823791.0, 9.0e-28)); }
From: bresink@infko.uni-koblenz.de (Marcel Bresink) Newsgroups: comp.sys.next.bugs Subject: Re: 3.2/3.3 NextStep-cluster problems Date: 16 Aug 1995 07:36:22 GMT Organization: University Koblenz / Germany Message-ID: <40s75m$h3g@newshost.uni-koblenz.de> References: <40qbfd$uu2@kralle.zdv.Uni-Mainz.DE> mueller@sonja.imsd.uni-mainz.DE (Robert Mueller) wrote: [...] > Or do we have the wrong network cards? Yes, kind of. The 3Com 3C509 is known to have severe problems when running NEXTSTEP. NEXTSTEP is internally optimized for using a network packet size of 8 KBytes. The 3C509 is optimized for DOS/NetWare environments and has a hardware buffer of only 1 KByte size. The result is that the network board is dropping incoming NFS packets sended by a NEXTSTEP computer. Quote from NeXT Answers Document #1482: "This adapter is not recommended for use in NFS intensive environments." In reality it is even worse: I tested the card myself and had exactly the same problem of extremely slow or crashing ethernet communication when running a NEXTSTEP to Novell NetWare connection with only 3 (!) computers in the ethernet segment. I would not call that an "NFS intensive environment"... There is some kind of work-around: You can try to set the read and write buffer size to 1024 bytes. (Use the expert options in NFSManager to do that.) The best solution is of course to replace the ethernet adapter. Viele Gruesse Marcel --- Marcel Bresink, University of Koblenz, Computer Graphics Lab, Institute for Computer Science Rheinau 1, D-56075 Koblenz, Germany
From: Karsten.Heinze@informatik.tu-chemnitz.de (Karsten Heinze) Newsgroups: comp.sys.next.bugs Subject: Re: Do this program fail on 3.3/Intel??? Date: 16 Aug 1995 09:55:00 GMT Organization: University of Technology Chemnitz, FRG Message-ID: <40sf9k$3ls@pyrrhus-f.hrz.tu-chemnitz.de> References: <gelatoDDDsM2.5EL@netcom.com> gelato@netcom.com (Steve James) wrote: > The following program goes into an infinite loop on 3.2/Intel (but > not on other 3.2 platforms). Can someone with a 3.3/Intel system > compile it and let me know if it fails under 3.3 as well? Thanks. > #include <math.h> > void main() > { > printf("%g\n",fmod(1283719823791.0, 9.0e-28)); > } Same here, with Developer 3.3 Intel. - Karsten --- ***************************************************************** Karsten Heinze Tel.: +49/172/3763092 Fax: +49/37296/15056 e-mail: Karsten.Heinze@informatik.tu-chemnitz.de (NeXT-Mail/MIME) *****************************************************************
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: gnutar and "longLink" message Date: Wed, 16 Aug 1995 12:41:14 -0400 Organization: Princeton University Message-ID: <Pine.NXT.3.91.950816123656.12721B-100000@capitalist.princeton.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII um... I am not sure if this "bug" is on the part of the program or the user, but here is the deal: I am trying to back-up my /LocalApps folder, using gnutar. I sometimes get a line (while watching the verbose mode) that says something like: @LongLink can anyone tell me what is going wrong? Is it me or the gnutar? I am using gnutar and then piping it to gzip, if that matters. thanks TjL -- Timothy J. Luoma luomat@capitalist.princeton.edu NeXT 3.2 m68k Just another convert to the Z-Shell "If I understood the man page, I wouldn't have asked the question." *+*+* MIME or ASCII Mail only !!!! *+*+*+*+
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: gnutar and "longLink" message Date: 17 Aug 1995 13:06:21 GMT Organization: Technical University of Berlin, Germany Message-ID: <40vesd$h7u@news.cs.tu-berlin.de> References: <Pine.NXT.3.91.950816123656.12721B-100000@capitalist.princeton.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Timothy J. Luoma (luomat@capitalist.princeton.edu) wrote: : um... I am not sure if this "bug" is on the part of the program or the : user, but here is the deal: : I am trying to back-up my /LocalApps folder, using gnutar. : I sometimes get a line (while watching the verbose mode) that says : something like: : @LongLink Because gnutar tries to be somewhat compatible to the old tar format, it can't store pathnames longer than 100 chars. In order to store files with longer names, it generates a special file entry containing just the longer filename. These are the long links you see. Nothing to worry about. Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: picard@silcom.com (Donald Burr) Newsgroups: comp.sys.next,comp.sys.next.bugs,comp.sys.next.misc,comp.sys.next.programmer Subject: Bug in NextStep 'scanf()' function - can it be fixed? Followup-To: comp.sys.next.misc Date: 18 Aug 1995 00:06:01 GMT Organization: Starfleet Command Message-ID: <410lh9$fml@ocean.silcom.com> The place where I work has a NeXTstation (yes, a real 680x0 NeXTstation -- but not color) running NeXTstep OS. We've found a bug in the scanf() function in the standard C library, that improperly reads and stores floating point numbers. The company writes software that's heavily-dependent on this functionality, and this bug is, to say the least, irritating. We've consulted with NeXT, and found out that: (a) their tech support, and perhaps the whole company itself, is a LOT worse than it used to be; and (b) NeXT wants an insanely huge amount of money to fix this little bugger. I had an idea last night, though. It seems that NeXTstep is based havily on GNU stuff -- the compiler reports "GNU cc version 1.something" as its version, and (though I have no way of checking) I bet the libraries are based on GNU libc as well. Would it be possible to grab the latest source for gcc, g++, and glibc off the Net, and install them? Would that fix the scanf() problem? Can this be done? Or, if I do it, will it erase some NeXT-specific customizations, custom functions, etc. that the system is dependent upon? Please respond to me at >picard@silcom.com< Thanks! -- Donald Burr [picard@silcom.com], P.O. Box 91212, Santa Barbara, CA 93190-1212 TEL: (805)564-1871 // FAX: 564-2315 // WWW: http://www.silcom.com/~picard PGP Public Key and more are available through the Auto-Responder -- send e-mail to me with the words "SEND HELP" in the Subject: line for info! ** Uphold your right to privacy - Use PGP. **
From: viggo@diku.dk (Allan Kim Schougaard) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Aaarrrggh! Root's WorkSpace keeps crashing! Date: 18 Aug 1995 14:03:55 GMT Organization: Department of Computer Science, U of Copenhagen Sender: viggo@njord.diku.dk Message-ID: <4126kb$2mh@odin.diku.dk> References: <40ftb5$jg9@atlas.cs.upei.ca> <40g9e3$6v0@news4.digex.net> >peter@bert.psyc.upei.ca wrote: >> Yesterday, I was installing a package off a CD-ROM and the >> installation failed. Ever since, WorkSpace has started crashing >> whenever I'm logged in as root. The WorkSpace only crashes when >> I'm looking in the LocalApps directory, in Icon or Listing view >> mode, and when I scroll to a particular file. Could be an error on the disk. I've tried that in several versions, some times WS just finishes off, other times the thing hangs. --Allan --viggo@diku.dk
Newsgroups: comp.sys.next.bugs From: andreas@esag.ch (Andreas Lutsch) Subject: NFS automounter problem Message-ID: <DDI647.34G@esag.ch> Sender: news@esag.ch Organization: Trinex AG, Sissach, CH Date: Fri, 18 Aug 1995 10:50:31 GMT There was a posting in csn.sysadmin that reminded me of an old bug in the NFS automounter. It always mounts directories (from the same computer) in exactly the reverse order you generated the entries in the NFSManager's 'Imported Directories' list. That means, if you choose to import some-host:/ and later add some-host:/usr (both with mount point /Net), the automounter will always mount /private/Net/some-host/usr *before* /private/Net/some-host. This way some-host:/usr is effectively invisible, though mounted. One can work around this bug by adding some-host:/ after some-host:/usr (so it will be mounted before), but only if both imported directories are defined in the same NetInfo domain. Otherwise there's no possibility to influence the mount order. In my opinion, the NFS automounter should sort the directories before mounting them, that shouldn't be too complicated. This bug already existed in 3.2, and it is still there in 3.3. Does anybody know if NeXt is working on it or if one should report it with BugNext? Andreas Lutsch
Newsgroups: comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.bugs From: fozztexx@nvc.cc.ca.us (Chris Osborn) Subject: Server keeps crashing every five days Message-ID: <DDIMv3.AK5@nvc.cc.ca.us> Sender: news@nvc.cc.ca.us Organization: Napa Valley College Date: Fri, 18 Aug 1995 16:52:14 GMT I've been having a problem lately where my server will crash every five days. I'm running NeXTStep 3.3, and I suspect either sendmail, or more likely, the CERN httpd to be a big part of the problem. I checked /usr/adm/messages, and I have a *ton* of this in it: Aug 18 07:25:00 nvcc last message repeated 43 times Aug 18 07:25:00 nvcc mach: proc: table is full Aug 18 07:25:00 nvcc mach: proc: table is full Aug 18 07:25:00 nvcc mach: enrx: no network buffers Aug 18 07:25:22 nvcc last message repeated 27 times Aug 18 07:25:28 nvcc mach: enrx: no network buffers Aug 18 07:26:00 nvcc last message repeated 14 times Aug 18 07:26:00 nvcc mach: proc: table is full Aug 18 07:26:00 nvcc mach: proc: table is full Aug 18 07:26:00 nvcc mach: enrx: no network buffers Aug 18 07:27:00 nvcc last message repeated 54 times Aug 18 07:27:00 nvcc mach: proc: table is full Aug 18 07:27:00 nvcc mach: proc: table is full Aug 18 07:27:01 nvcc mach: enrx: no network buffers Aug 18 07:27:28 nvcc last message repeated 12 times Aug 18 07:27:28 nvcc mach: enrx: no network buffers Aug 18 07:28:00 nvcc last message repeated 41 times Aug 18 07:28:00 nvcc mach: proc: table is full Aug 18 07:28:00 nvcc mach: proc: table is full Aug 18 07:28:02 nvcc mach: enrx: no network buffers Aug 18 07:29:00 nvcc last message repeated 42 times Is there a fix for this, short of stopping the computer from being a web or mail sever? -- Chris Osborn, Network Administrator Voice: 707 253 3130 Napa Valley College Fax: 707 253 3063 2277 Napa-Vallejo Hwy., Napa, CA, 94558 <fozztexx@nvc.cc.ca.us> MIME ok, NeXTMail tolerated
Newsgroups: comp.sys.next.bugs From: rdl@world.std.com (Robert La Ferla) Subject: Re: Is the intel Zappa Mother Board w/triton chipset affectes by the PCI Bug??? In-Reply-To: aprwsa@ix.netcom.com's message of 19 Aug 1995 08:57:49 GMT Message-ID: <RDL.95Aug19115221@world.std.com> Sender: rdl@world.std.com (Robert La Ferla) Organization: The World Public Access UNIX, Brookline, MA References: <41492d$j8s@ixnews4.ix.netcom.com> Date: Sat, 19 Aug 1995 15:52:21 GMT Intel MBs with the Triton chipset are NOT affected by either of the two major PCI bugs (RZ1000 IDE controller and write-post.) Robert La Ferla Registered NEXTSTEP / OPENSTEP Consultant Boston, MA + 1 (617) 252-0088 In article <41492d$j8s@ixnews4.ix.netcom.com> aprwsa@ix.netcom.com (Julie Sulston) writes: Subject says it all......... Please let me know more about this bug. Julie
From: root@rtcc_server.globalx.net (RADium Technology Centre (Canada)) Newsgroups: comp.sys.next.bugs Subject: Re: Do this program fail on 3.3/Intel??? Date: 19 Aug 1995 16:35:23 GMT Organization: Global-X-Change Message-ID: <4153sb$i6u@news.globalx.net> References: <gelatoDDDsM2.5EL@netcom.com> gelato@netcom.com (Steve James) wrote: > The following program goes into an infinite loop on 3.2/Intel (but > not on other 3.2 platforms). Can someone with a 3.3/Intel system > compile it and let me know if it fails under 3.3 as well? Thanks. > #include <math.h> > void main() > { > printf("%g\n",fmod(1283719823791.0, 9.0e-28)); > } I tried this on my system, and it *DIDN'T* loop infinitly. (It didn't printf, either, but that's a different issue). Stephen MacDougall -- ============================================================================ | | | | RADium Technology Centre (Canada) | #include <std_disclaimer.h> | | radium@globalx.net | | | | (NeXTMail and MIME capable) | ============================================================================
From: johnbach@net (Restrict) Newsgroups: comp.sys.next.bugs Subject: Government Restricts Internet!! Date: Sat, 19 Aug 1995 21:58:08 GMT Organization: Netcom Distribution: world Message-ID: <415mmn$7a2@ixnews4.ix.netcom.com> SPEAK UP AMERICA -- MAKE YOUR VOTE COUNT Question: Should the United States Government interfere and put restrictions on the use of the Internet?? CALL: 1-900-945-5600 ext 163 and cast your vote. Cost: $1.98 per call (NOT per minute) Call Today Must be 18+/Touch Tones Only InfoService/Studio City, CA/213-993-3366 Results of this survey will be compiled and sent to members of the House and Senate. Thank you for casting your vote and for making your voice heard.
Date: 19 Aug 1995 22:08:06 GMT From: johnbach@net (Restrict) Message-ID: <cancel.415mmn$7a2@ixnews4.ix.netcom.com> Newsgroups: comp.sys.next.bugs Subject: cmsg cancel <415mmn$7a2@ixnews4.ix.netcom.com> Control: cancel <415mmn$7a2@ixnews4.ix.netcom.com> Spam cancelled by clewis@ferret.ocunix.on.ca
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 21 Aug 1995 04:15:11 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <41918f$dot@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: aprwsa@ix.netcom.com (Julie Sulston) Newsgroups: comp.sys.next,comp.sys.next.bugs,comp.sys.next.advocacy,comp.sys.next.bugs Subject: Which Video Card to Get........HELP!!!! Date: 21 Aug 1995 08:36:12 GMT Organization: Netcom Message-ID: <419ghs$n7v@ixnews5.ix.netcom.com> I need to get a video card for my Intel system. I have PCI slots available, but I can only use the short type cards (not full length). Looking for a 64BIT w/2meg vram Which brand/Model should I get?? Which has the LEAST BUGS w/nextstep? PLEASE TELL ME QUICKLY, I WILL BE BUYING IT AT 5:00pm TODAY REPLY TO E-MAIL PLEASE sorry for shouting.
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: gnutar and "longLink" message Date: 21 Aug 1995 15:44:07 GMT Organization: Information Technology Solutions, Inc. Message-ID: <41a9k7$729@news.its.com> References: <Pine.NXT.3.91.950816123656.12721B-100000@capitalist.princeton.edu> "Timothy J. Luoma" <luomat@capitalist.princeton.edu> wrote: > um... I am not sure if this "bug" is on the part of the program or the > user, but here is the deal: The user. :-) > I am trying to back-up my /LocalApps folder, using gnutar. > I sometimes get a line (while watching the verbose mode) that says > something like: > @LongLink > > can anyone tell me what is going wrong? Is it me or the gnutar? Nothing is going wrong. Tar has a limitation that pathnames can only be 128 characters long, or else the file/directory names get truncated. Gnutar gets around that limitation by renaming files with the @LongLinkXXXX name (so that normal tar will at least be able to extract everything, albeit with the wrong name), and then it will rename the files to the correct names based off of a "hidden" file at the end of the tar archive. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: kulickja@saturn.aitc.rest.tasc.com. (Judy A. Kulick) Newsgroups: comp.sys.next.bugs Subject: AHA Errors Date: 21 Aug 1995 17:50:39 GMT Organization: TASC Message-ID: <41ah1f$fso@viper.is.rest.tasc.com> We are experiencing SCSI errors (apparrently) with our new DEC Celebris XL590 server which is running NeXTStep 3.3. We have put an ADAPTEC SCSI card in it and the errors are AHA interrupt 13 and 14. We swapped out the card, but that did not fix the problem. The problem is intermittent and will crash the server every couple of days. Accessing the tape drive (also on the scsi chain) seems to cause problems. Wondering if anyone else has experienced similar problems on the new Celebris XL590's running NS 3.3. -- NewsGrazer, a NeXTstep(tm) news reader, posting -- M>UQR=&8P7&%N<VE[7&9O;G1T8FQ<9C!<9FUO9&5R;B!#;W5R:65R.WT*7&UA M<F=L,3(P"EQM87)G<C$R,`I<<&%R9%QT>#$Q-3)<='@R,S`T7'1X,S0U-EQT M>#0V,#A<='@U-S8P7'1X-CDQ,EQT>#@P-C1<='@Y,C$V7'1X,3`S-CA<='@Q M,34R,%QF,%QB,%QI,%QU;&YO;F5<9G,R-%QF8S!<8V8P(%=E(&%R92!E>'!E M<FEE;F-I;F<@4T-322!E<G)O<G,@*&%P<&%R<F5N=&QY*2!W:71H(&]U<B!N M97<@1$5#($-E;&5B<FES(%A,-3DP('-E<G9E<B!W:&EC:"!I<R!R=6YN:6YG M($YE6%13=&5P(#,N,RX@(%=E(&AA=F4@<'5T(&%N($%$05!414,@4T-322!C M87)D(&EN(&ET(&%N9"!T:&4@97)R;W)S(&%R92!!2$$@:6YT97)R=7!T(#$S M(&%N9"`Q-"X@(%=E('-W87!P960@;W5T('1H92!C87)D+"!B=70@=&AA="!D M:60@;F]T(&9I>"!T:&4@<')O8FQE;2X@(%1H92!P<F]B;&5M(&ES(&EN=&5R M;6ET=&5N="!A;F0@=VEL;"!C<F%S:"!T:&4@<V5R=F5R(&5V97)Y(&-O=7!L M92!O9B!D87ES+B`@06-C97-S:6YG('1H92!T87!E(&1R:79E("AA;'-O(&]N M('1H92!S8W-I(&-H86EN*2!S965M<R!T;R!C875S92!P<F]B;&5M<RX@(%P* M7`I7;VYD97)I;F<@:68@86YY;VYE(&5L<V4@:&%S(&5X<&5R:65N8V5D('-I M;6EL87(@<')O8FQE;7,@;VX@=&AE(&YE=R!#96QE8G)I<R!83#4Y,"=S(')U 1;FYI;F<@3E,@,RXS+B`*?0H@ `
From: aprwsa@ix.netcom.com (Julie Sulston) Newsgroups: comp.sys.next.bugs Subject: Is the intel Zappa Mother Board w/triton chipset affectes by the PCI Bug??? Date: 19 Aug 1995 08:57:49 GMT Organization: Netcom Distribution: world Message-ID: <41492d$j8s@ixnews4.ix.netcom.com> Subject says it all......... Please let me know more about this bug. Julie
From: tracker@coho.halcyon.com (Russell Wilcoxon) Newsgroups: comp.sys.next.bugs Subject: Re: AHA Errors Date: 22 Aug 1995 17:56:59 GMT Organization: Northwest Nexus Inc. Message-ID: <41d5pb$1e6@news1.halcyon.com> References: <41ah1f$fso@viper.is.rest.tasc.com> kulickja@saturn.aitc.rest.tasc.com. (Judy A. Kulick) writes: >We are experiencing SCSI errors (apparrently) with our new DEC >Celebris XL590 server which is running NeXTStep 3.3. We have put an >ADAPTEC SCSI card in it and the errors are AHA interrupt 13 and 14. >We swapped out the card, but that did not fix the problem. The >problem is intermittent and will crash the server every couple of >days. Accessing the tape drive (also on the scsi chain) seems to >cause problems. >Wondering if anyone else has experienced similar problems on the new >Celebris XL590's running NS 3.3. Adaptec cards have a setup utility in the BIOS. (cntl-a at boot) If I remember correctly you need to disable some "features" to get rid of the AHA. (dos-multiple drives is one.) Step down the bus speed also. If you still have problems, I'll look at my setup for you. Good luck, -Russ
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: POSIX gethostbyname() broken, Bug_NeXT log #72854 Date: 22 Aug 1995 18:59:24 GMT Organization: Information Technology Solutions, Inc. Message-ID: <41d9ec$hev@news.its.com> The POSIX version of gethostbyname() is broken under NEXTSTEP 3.3, as the following test program indicates: % cat test.c #include <ansi/ansi.h> #include <netdb.h> int main(argc, argv) int argc; char *argv[]; { struct hostent *h = NULL; if (argc > 1) h = gethostbyname(argv[1]); if (h) { printf("\nhostname = %s\n", h->h_name); return 0; } else { puts("\nHostname not found!\n"); return -1; } } % cc -o test test.c % ./test www.next.com hostname = ftp.NeXT.COM % cc -posix -o test test.c test.c:1: warning: could not use precompiled header '/NextDeveloper/Headers/ansi/ansi.p', because: test.c:1: warning: #ifdef '_POSIX_SOURCE' not defined when precompiled, but now defined: test.c:1: warning: on the command line % ./test www.next.com Hostname not found! ---------------------------- This test code works correctly under NEXTSTEP 3.2. I have a workaround, which is to build a modern version of bind (the DNS server), and link from the libresolv.a resolver library that get created with that software package. If there is interest, I suppose I could package up quad-FAT versions of the 4.9.3-Beta24 named, named-xfer, ndc, and libresolv.a. Hmm, perhaps falling back on a non-beta release (ie, v4.9.2) would be wiser? -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy From: Gerben_Wierda@RnA.NL Subject: Autodial Morningstar PPP/lookupd problem fixed by 3.3 patch! Message-ID: <DDrFz2.AJ@RnA.NL> Sender: gerben@RnA.NL (Gerben Wierda) Organization: G.R.O.S.S. Date: Wed, 23 Aug 1995 11:01:50 GMT The soon to be released 3.3 patch for NEXTSTEP 3.3 fixes the problem with lookupd (single threadedness) that made auto-dial with Morningstar PPP fail. A while ago I posted this in csn.advocacy. >Subject: Hey NeXT! if you can fix the serial drivers, you can fix lookupd!> >Hello NeXT, > >Lookupd is driving many people crazy. Take for instance that some app uses >gethostbyname(), which starts up some other daemon (say mst ppp), which then >does anything which involves lookupd (say getservbyport()), and BANG: deadlock >until the original call times out. A real Catch-22, the first can't finish >because it waits for the second, and the second doesn't get attention before >the first has finished! > >Hey NeXT, it was a very good move to release better serial drivers. Please, >please, please, PLEASE release a functional lookupd, it's a bottleneck of >bottlenecks! > >Yours, Soon after I got a mail from NeXT asking if I wanted to test this problem with the upcoming patch. I did of course (and I can tell about it this time since there was no NDA involved as with alpha-beta releases of the OS/Dev) and the patch arrived today. I installed it and it fixes the problem! Yes!! No I have auto dial firewall-like PPP connection! There are many people more able than I to comment on what this means v.a.v. ongoing discussions on NeXT as customer- or net-aware, so I will leave it at this. Happy hacking, -- Gerben_Wierda@RnA.nl (Gerben Wierda) NEXTSTEP RD242 "If you don't know where you're going, any road will take you there" Paraphrased in Alice in Wonderland, originally from the Talmud.
From: rdieter@mathlab41.unl.edu (Rex Dieter) Newsgroups: comp.sys.next.bugs Subject: Re: POSIX gethostbyname() broken, Bug_NeXT log #72854 Date: 23 Aug 1995 16:09:52 GMT Organization: University of Nebraska--Lincoln Distribution: world Message-ID: <41fjsg$n4v@crcnis3.unl.edu> References: <41d9ec$hev@news.its.com> In article <41d9ec$hev@news.its.com> chuck@its.com (Chuck Swiger) writes: > The POSIX version of gethostbyname() is broken under NEXTSTEP 3.3, as the > following test program indicates: Wierd... It works for me... (NEXTSTEP 3.3, i486) % cat testme.c #include <ansi/ansi.h> #include <netdb.h> int main(argc, argv) int argc; char *argv[]; { struct hostent *h = NULL; if (argc > 1) h = gethostbyname(argv[1]); if (h) { printf("\nhostname = %s\n", h->h_name); return 0; } else { puts("\nHostname not found!\n"); return -1; } } % cc -o testme testme.c % ./testme www.next.com hostname = ftp.NeXT.COM % cc -o testme -posix testme.c testme.c:1: warning: could not use precompiled header '/NextDeveloper/Headers/ansi/ansi.p', because: testme.c:1: warning: #ifdef '_POSIX_SOURCE' not defined when precompiled, but now defined: testme.c:1: warning: on the command line % ./testme www.next.com hostname = ftp.NeXT.COM %
From: rdieter@mathlab41.unl.edu (Rex Dieter) Newsgroups: comp.sys.next.bugs Subject: Re: POSIX gethostbyname() broken, Bug_NeXT log #72854 Date: 23 Aug 1995 16:12:06 GMT Organization: University of Nebraska--Lincoln Distribution: world Message-ID: <41fk0m$n53@crcnis3.unl.edu> References: <41fjsg$n4v@crcnis3.unl.edu> In article <41fjsg$n4v@crcnis3.unl.edu> rdieter@mathlab41.unl.edu (Rex Dieter) writes: > In article <41d9ec$hev@news.its.com> chuck@its.com (Chuck Swiger) writes: > > The POSIX version of gethostbyname() is broken under NEXTSTEP 3.3, as the > > following test program indicates: > > Wierd... It works for me... (NEXTSTEP 3.3, i486) Sorry... I goofed... I have NS-3.3 User + NS-3.2 Developer. So, I just confirmed that it works for 3.2. -- Rex Dieter University of Nebraska-Lincoln, Math/Stat Department rdieter@math.unl.edu
From: reichman@scf.usc.edu (Matthew N. Reichman) Newsgroups: comp.sys.next.bugs Subject: Re: lookupd Date: 23 Aug 1995 18:16:48 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-d-13.usc.edu Message-ID: <41frag$9r6@usc.edu> benb@aitken.velocity.com (Ben Black) writes, > i guess it will show up in 4.0. premium support customers have already > gotten about 10MB (compressed) of patched programs for 3.3 user. Have any of these patched programs been made available to the general (albeit small) general public? Matthew -- Matthew ====================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV | PGP key upon request "Show disorder to make them take a chance..." - Sun Tzu
From: jbryans@csulb.edu (Jack Bryans) Newsgroups: comp.sys.next.bugs Subject: Re: POSIX gethostbyname() broken, Bug_NeXT log #72854 Date: 23 Aug 1995 21:16:57 GMT Organization: Cal State Long Beach Message-ID: <JBRYANS.95Aug23141657@wren.csulb.edu> References: <41d9ec$hev@news.its.com> In article <41d9ec$hev@news.its.com> chuck@its.com (Chuck Swiger) writes: > The POSIX version of gethostbyname() is broken under NEXTSTEP 3.3, as the > following test program indicates: It worked here on a white 3.3. robin:/tmp/test 68$ make CFLAGS='-posix' test cc -posix test.c -o test test.c:1: warning: could not use precompiled header '/NextDeveloper/Headers/ansi/ansi.p', because: test.c:1: warning: #ifdef '_POSIX_SOURCE' not defined when precompiled, but now defined: test.c:1: warning: on the command line robin:/tmp/test 69$ ./test www.next.com hostname = ftp.NeXT.COM So, let's try it on a black 3.3: wren:/tmp/test 74$ make CFLAGS='-posix' test cc -posix test.c -o test test.c:1: warning: could not use precompiled header '/NextDeveloper/Headers/ansi/ansi.p', because: test.c:1: warning: #ifdef '_POSIX_SOURCE' not defined when precompiled, but now defined: test.c:1: warning: on the command line wren:/tmp/test 75$ ./test www.next.com hostname = ftp.NeXT.COM No problem. Jack
Newsgroups: comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.bugs From: rdl@world.std.com (Robert La Ferla) Subject: Re: Server keeps crashing every five days In-Reply-To: fozztexx@nvc.cc.ca.us's message of Fri, 18 Aug 1995 16:52:14 GMT Message-ID: <RDL.95Aug23223216@world.std.com> Sender: rdl@world.std.com (Robert La Ferla) Organization: The World Public Access UNIX, Brookline, MA References: <DDIMv3.AK5@nvc.cc.ca.us> Date: Thu, 24 Aug 1995 02:32:16 GMT I suspect httpd. What version of sendmail are you running? Robert La Ferla Registered NEXTSTEP / OPENSTEP Consultant Boston, MA + 1 (617) 252-0088 In article <DDIMv3.AK5@nvc.cc.ca.us> fozztexx@nvc.cc.ca.us (Chris Osborn) writes: I've been having a problem lately where my server will crash every five days. I'm running NeXTStep 3.3, and I suspect either sendmail, or more likely, the CERN httpd to be a big part of the problem. I checked /usr/adm/messages, and I have a *ton* of this in it: Aug 18 07:25:00 nvcc last message repeated 43 times Aug 18 07:25:00 nvcc mach: proc: table is full Aug 18 07:25:00 nvcc mach: proc: table is full Aug 18 07:25:00 nvcc mach: enrx: no network buffers Aug 18 07:25:22 nvcc last message repeated 27 times Aug 18 07:25:28 nvcc mach: enrx: no network buffers Aug 18 07:26:00 nvcc last message repeated 14 times Aug 18 07:26:00 nvcc mach: proc: table is full Aug 18 07:26:00 nvcc mach: proc: table is full Aug 18 07:26:00 nvcc mach: enrx: no network buffers Aug 18 07:27:00 nvcc last message repeated 54 times Aug 18 07:27:00 nvcc mach: proc: table is full Aug 18 07:27:00 nvcc mach: proc: table is full Aug 18 07:27:01 nvcc mach: enrx: no network buffers Aug 18 07:27:28 nvcc last message repeated 12 times Aug 18 07:27:28 nvcc mach: enrx: no network buffers Aug 18 07:28:00 nvcc last message repeated 41 times Aug 18 07:28:00 nvcc mach: proc: table is full Aug 18 07:28:00 nvcc mach: proc: table is full Aug 18 07:28:02 nvcc mach: enrx: no network buffers Aug 18 07:29:00 nvcc last message repeated 42 times Is there a fix for this, short of stopping the computer from being a web or mail sever? -- Chris Osborn, Network Administrator Voice: 707 253 3130 Napa Valley College Fax: 707 253 3063 2277 Napa-Vallejo Hwy., Napa, CA, 94558 <fozztexx@nvc.cc.ca.us> MIME ok, NeXTMail tolerated
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: POSIX gethostbyname() broken, Bug_NeXT log #72854 Date: 24 Aug 1995 14:44:53 GMT Organization: Information Technology Solutions, Inc. Message-ID: <41i395$o1@news.its.com> References: <41d9ec$hev@news.its.com> <JBRYANS.95Aug23141657@wren.csulb.edu> jbryans@csulb.edu (Jack Bryans) wrote: > In article <41d9ec$hev@news.its.com> chuck@its.com (Chuck Swiger) writes: > It worked here on a white 3.3. [ ... ] > So, let's try it on a black 3.3: [ ... ] > No problem. Very inetresting. Do you have the 3.3 patch installed? All of our 3.3 systems have the patch installed, and the problem does not appear on 3.2. I'd be interested in hearing from other people who've done this test, especially if you have the 3.3 patch installed. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: cpayne@optical (Carl Payne) Newsgroups: comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Server keeps crashing every five days Followup-To: comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.bugs Date: 24 Aug 1995 22:02:16 GMT Organization: Fibernet! Satisfy the need for speed Message-ID: <41ist8$moe@optical.fiber.net> References: <DDIMv3.AK5@nvc.cc.ca.us> <RDL.95Aug23223216@world.std.com> Robert La Ferla (rdl@world.std.com) wrote: : I suspect httpd. What version of sendmail are you running? : In article <DDIMv3.AK5@nvc.cc.ca.us> fozztexx@nvc.cc.ca.us (Chris Osborn) writes: : I checked /usr/adm/messages, and I have a *ton* of this in it: : Aug 18 07:25:00 nvcc last message repeated 43 times : Aug 18 07:25:00 nvcc mach: proc: table is full Iseriously doubt it's the CERN server. I used to get the same thing, but only when radiusd was running (bad int call). Either machine, black or white, and the x86 doesn't run CERN. SO, I strongly suggest it's none of the above. IME, it's radiusd, which I can help you fix. If you're not running radiusd, maybe it's TACACS, or something similar. If I can help, mail me a fresh (post-boot) ps -ax and a netstat -a -- Carl Payne cpayne@fiber.net <--NeXTMAIL, MIME, ASCII all OK
From: sears@uh.edu (Paul S. Sears) Newsgroups: comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Server keeps crashing every five days Date: 25 Aug 1995 14:32:55 GMT Organization: University of Houston Distribution: world Message-ID: <41kmun$78l@masala.cc.uh.edu> References: <41ist8$moe@optical.fiber.net> In article <41ist8$moe@optical.fiber.net> cpayne@optical (Carl Payne) writes: > Robert La Ferla (rdl@world.std.com) wrote: > : I suspect httpd. What version of sendmail are you running? > > : In article <DDIMv3.AK5@nvc.cc.ca.us> fozztexx@nvc.cc.ca.us (Chris Osborn) writes: > : I checked /usr/adm/messages, and I have a *ton* of this in it: > : Aug 18 07:25:00 nvcc last message repeated 43 times > : Aug 18 07:25:00 nvcc mach: proc: table is full > > Iseriously doubt it's the CERN server. I used to get the same thing, but only > when radiusd was running (bad int call). Either machine, black or white, and > the x86 doesn't run CERN. SO, I strongly suggest it's none of the above. IME, > it's radiusd, which I can help you fix. If you're not running radiusd, maybe > it's TACACS, or something similar. > > If I can help, mail me a fresh (post-boot) ps -ax and a netstat -a > Yep, it probably is not httpd. Our web server is currently CERN 3.0 httpd (soon to be switching to apache!) and has been up for: 9:31am up 28 days, 16:10, 0 user, load average: 0.42, 0.18, 0.18 No problems with the daemon. Now, this is the _only_ thing this particular NEXTSTEP box does. No logins, etc... > -- > Carl Payne > cpayne@fiber.net <--NeXTMAIL, MIME, ASCII all OK -- Paul S. Sears * sears@uh.edu (NeXT Mail OK) The University of Houston * <URL:http://www.egr.uh.edu/~sears/> Engineering Computing Center * SSI Diving Certification #755020059 NeXT System Administration * DoD#1967 '83 NightHawk 650SC "Programming is like sex: One mistake and you support it a lifetime." ******* I do not represent the University of Houston *********
Newsgroups: comp.sys.next.bugs From: pjoe@charon.muc.de (Peter Eybert) Subject: Re: POSIX gethostbyname() broken, Bug_NeXT log #72854 Message-ID: <1995Aug25.141745.1929@charon.muc.de> Sender: pjoe@charon.muc.de (Peter Joe Eybert) Organization: None References: <41fk0m$n53@crcnis3.unl.edu> Date: Fri, 25 Aug 1995 14:17:45 GMT Rex Dieter writes > In article <41fjsg$n4v@crcnis3.unl.edu> rdieter@mathlab41.unl.edu (Rex > Dieter) writes: > > In article <41d9ec$hev@news.its.com> chuck@its.com (Chuck Swiger) writes: > > > The POSIX version of gethostbyname() is broken under NEXTSTEP 3.3, as the > > > following test program indicates: > > > > Wierd... It works for me... (NEXTSTEP 3.3, i486) > > Sorry... I goofed... I have NS-3.3 User + NS-3.2 Developer. So, I just > confirmed that it works for 3.2. > But it works for me (NEXTSTEP 3.3 User+Dev, m68k, without EOF1.1): ~/Progs/c-pure 3% cc -Wall -o gethbn gethbn.c ~/Progs/c-pure 4% ~/Progs/c-pure 10% cc -Wall -posix -o gethbnp gethbn.c gethbn.c:1: warning: could not use precompiled header '/NextDeveloper/Headers/ansi/ansi.p', because: gethbn.c:1: warning: #ifdef '_POSIX_SOURCE' not defined when precompiled, but now defined: gethbn.c:1: warning: on the command line ~/Progs/c-pure 11% ~/Progs/c-pure 23% gethbn www.next.com hostname = ftp.NeXT.COM ~/Progs/c-pure 24% gethbnp www.next.com hostname = ftp.NeXT.COM -- ____________________________________________________________ Peter Eybert pjoe@charon.muc.de Appenzellerstr. 123 Tel: +49-89-7593734 81475 Muenchen (NeXTMail welcome) --
From: giddings@menominee.chem.wisc.edu Newsgroups: comp.sys.next.bugs Subject: Bug in NS 3.3 find (intel, not motorola) Date: 25 Aug 1995 16:57:44 GMT Organization: University of Wisconsin, Madison Distribution: world Message-ID: <41kve8$r4i@news.doit.wisc.edu> Summary: Different output based on computer system Keywords: find intel nextstep version The find command does not appear to work correctly on NS/Intel 3.3. Here is an example (I got the same result from a makefile or running this from csh): find ./* -type d -prune -exec sh -c 'cd {} ; ls' \; On an intel machine, the output is: sh: {}: bad directory sh: {}: bad directory sh: {}: bad directory sh: {}: bad directory sh: {}: bad directory In the _same_ directory accessed via NFS from a motorola box, the _same_ command results in the directory listings for the subdirectories, as it should. Anybody know why there is this difference? Thanks, Michael Giddings giddings@whitewater.chem.wisc.edu
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: Bug in NS 3.3 find (intel, not motorola) Date: 25 Aug 1995 22:49:55 GMT Organization: Technical University of Berlin, Germany Distribution: world Message-ID: <41lk2j$bub@news.cs.tu-berlin.de> References: <41kve8$r4i@news.doit.wisc.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit giddings@menominee.chem.wisc.edu wrote: : The find command does not appear to work correctly on NS/Intel 3.3. Here is an example (I got the same result from a makefile or running this from csh): : find ./* -type d -prune -exec sh -c 'cd {} ; ls' \; : On an intel machine, the output is: : sh: {}: bad directory [...] : In the _same_ directory accessed via NFS from a motorola box, the _same_ command results in the directory listings for the subdirectories, as it should. : Anybody know why there is this difference? Just tried it on both Intel and Moto. Works as it should. Maybe something broken with your sh? Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: fischer-michael@cs.yale.edu (Michael Fischer) Newsgroups: comp.sys.next.bugs Subject: Re: Bug in NS 3.3 find (intel, not motorola) Date: 26 Aug 1995 21:40:24 GMT Organization: Yale University Computer Science Dept., New Haven, CT 06520-8285 Distribution: world Message-ID: <41o4c8$eim@babyblue.cs.yale.edu> References: <41kve8$r4i@news.doit.wisc.edu> giddings@menominee.chem.wisc.edu wrote: : The find command does not appear to work correctly on NS/Intel 3.3. Here is an example (I got the same result from a makefile or running this from csh): : find ./* -type d -prune -exec sh -c 'cd {} ; ls' \; : On an intel machine, the output is: : sh: {}: bad directory Same thing happens on my black (m68k) NeXt under 3.3. However, the behavior you observe seems to be correct, for it is what one would expect from the man page for "find": -exec command True if the executed command returns a zero value as exit status. The end of command must be punc- tuated by an escaped semicolon. A command argu- ment {} is replaced by the current pathname. A command is a command name followed by argument strings. In your example, the command name is "sh", which is followed by two arguments, "-c" and the entire string "cd {} ; ls". Since neither argument is (equal to) {}, no substitution takes place. Hence, "sh" subsequently attempts to execute "cd {}", and since "{}" is not a valid directory name, sh print out the cryptic "bad directory" error message, which is talking about the directory you are trying to cd to, not your working directory. Your example would work if substitions took place within argument strings, and perhaps some versions of find work that way, but that is not the way the one described here works, nor the ones on SunOS, Solaris, or AIX. ================================================== | Michael Fischer <fischer-michael@cs.yale.edu> | ==================================================
From: pierre@nova.org (Thomas Piergallini) Newsgroups: comp.sys.next.bugs Subject: Re: Bug in NS 3.3 find (intel, not motorola) {find codes DB not working} Date: 27 Aug 1995 03:12:55 GMT Organization: Megaweb Distribution: world Message-ID: <41onrn$p2a@news-e1a.megaweb.com> References: <41o4c8$eim@babyblue.cs.yale.edu> In article <41o4c8$eim@babyblue.cs.yale.edu> fischer-michael@cs.yale.edu (Michael Fischer) writes: > giddings@menominee.chem.wisc.edu wrote: > : The find command does not appear to work correctly on NS/Intel 3.3. Here is an example (I got the same result from a makefile or running this from csh): > > : find ./* -type d -prune -exec sh -c 'cd {} ; ls' \; > > : On an intel machine, the output is: > : sh: {}: bad directory The find codes database is non-functional for me (NSFIP). I run updatedb out of /usr/lib/find no problem, it creates the find.codes database in /etc. But when you execute the find command, no output is returned. Any ideas? On black boxes this is not a problem, and it works just fine. It did not work in 3.2 on intel either. pierre -- Thomas Piergallini EMail: pierre@3com.com, pierre@nova.org
From: helpme95@ix.netcom.com Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software Subject: Does Windows 95 really work? Date: Sun, 27 Aug 1995 15:44:28 GMT Organization: Netcom Message-ID: <41q099$m1t@ixnews2.ix.netcom.com> It's been working great for me (in test versions) since May. Because WINDOWS 95 seems to be of almost universal interest, I'm posting to a wide variety of computer users. Take the advice of someone at a lonely computer terminal who's been around the block a few times. [E-mail helpme-95@zoom.com for even more info.] >>>>>Into scifi? Take a look four paragraphs down! I'm tired. I wanted it easier. WINDOWS 95 has made it that way. No, the four horsemen didn't make an appearance when I started using WINDOWS 95, and it wasn't as important as a Beatles reunion or a Pearl Jam up-close-and-personal visit or winning a marathon (or the lottery). I haven't yet felt the aura of Bill Gates floating about my monitor. (I didn't notice a halo around his head, anyway.) WINDOWS 95 does have lots of advantages, though. And it is important that you make the best possible use of WINDOWS 95 should you decide to fork over your hard- earned bucks to get the second coming of WINDOWS. I have an informative and fun newsletter about the new stuff--and thought you might like to see it. Free, of course. I'm cheap, too! Forgive me, I mean FRUGAL! E-mail me at helpme-95@zoom.com for the newsletter, and I will be most happy to rush it back to you. From one computer user to another--we all try to do this better and faster whether it's our livelihood or our hobby. Try me! C'mon, people, lighten up. It's WINDOWS 95... Have fun!!! %#%#%%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#% AND NOW, FOR THE SCI-FI DREAM SEQUENCE (Just couldn't resist getting this thing published somewhere!) %#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#% "The truth is out there." Moldier sighed as he looked up to the stars in the dark night sky. "Yeah, sure," replied Scuzzy, "you keep sayin' that but we never go anywhere except those dark buildings where we get to use our flashlights." "Trust me." "Yeah, sure--like I trusted you with my car and you got it all banged up and then left me in the clutches of that alien? I don't trust nobody no more." Moldier begged, "Ah, come on, please let me drive just once more? I promise no more aliens for this week." Scuzzy mumbled under her breath something about red shoes and diaries and handed him the keys. At that moment, the sky above them came alive with lights flashing and weird soundss. "What the @#%?><&???", yelled Moldier as he dove to the ground, grabbing Scuzzy. It was huge and white and was disc shaped with two side rockets. It tried to strafe them with bolts of energy, but luckily missed them. The car, however, got a toasted paint job and the tires melted. Meanwhile, above in the spacecraft, a strange being with pointed ears and his friend were arguing. "Look at that--you MISSED them! I can't believe this!" The friend was having trouble. "I don't know what it is-- I know I programmed this scenario okay." "The pointy- eared fellow laughed and said, "You don't have that new program from Microsoft, do you? How many times did I tell you that it'll make your calculations easier and more accurate? But would you listen to me? Noooooooo!" Church turned around to take a swipe at Speck, but Speck was too fast for him and did a tuck-and- roll out of they way. Church turned again to his computer screen, but Speck stopped him. "Let's try this on my computer," he said. Church gave in grudgingly and Speck set up the run. Down below, Scuzzy and Moldier were zig-zagging through the field, trying to evade the alien ship. I'll get you for this, Moldier," panted Scuzzy, as she dodged another energy bolt. Moldier was too busy to reply, since he was sprinting faster than an Olympic- class runner. Seconds later, after a particularly nasty surge of energy, all that lay in the field were two piles of ash and some melted rubber. "Got em!", yelled Speck. "I told you that program was easier to set up!" Church stood by and glumly replied, "Yeah, but I bet I can beat you the next system we're in. Let's go someplace we've never been before." Speck turned to Church and asked, "Boldly, sir?" "But of course", laughed Church. As the ship turned and soared away, the two friends looked out the window up at the stars in the dark night sky. %#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%# For a good time, e-mail me for my newsletter at: helpme-95@zoom.com. See ya there!! %#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#%#
Newsgroups: comp.sys.next.bugs From: phn2n@poe.acc.Virginia.EDU (Paul R. Starks) Subject: Re: lookupd Message-ID: <DDzBpz.xs@murdoch.acc.Virginia.EDU> Sender: usenet@murdoch.acc.Virginia.EDU Organization: University of Virginia References: <41frag$9r6@usc.edu> Date: Sun, 27 Aug 1995 17:10:47 GMT In article <41frag$9r6@usc.edu>, Matthew N. Reichman <reichman@scf.usc.edu> wrote: >benb@aitken.velocity.com (Ben Black) writes, > >> i guess it will show up in 4.0. premium support customers have >already >> gotten about 10MB (compressed) of patched programs for 3.3 user. > >Have any of these patched programs been made available to the general >(albeit small) general public? > >Matthew >-- >Matthew > >====================================================================== >Matthew Reichman | NeXTStep v.3.3 m68k >reichman@scf.usc.edu | NeXTMAIL & MIME welcome >USC-CNTV | PGP key upon request > > "Show disorder to make them take a chance..." - Sun Tzu Yes, I would also like to know if next will release these patches to non-premium support customers like most of us! -Paul
From: ted@pdgm.com (Edward E. Murphy) Newsgroups: comp.sys.next.bugs Subject: NewsGrazer Crashes Date: 27 Aug 1995 20:03:49 GMT Organization: Paradigm Investment Services, Inc. Message-ID: <41qj35$2qg@pdgm.com> Hi all, NewsGrazer75 seems to crash after pressing the "Catch up" cell a few times. Anyone have a fix for this? Thanks, Ted.
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 28 Aug 1995 04:15:11 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <41rfsf$sh9@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: Paul Lynch <Paul_Lynch@plsys.com> Newsgroups: comp.sys.next.bugs Subject: Re: NewsGrazer Crashes Date: Mon, 28 Aug 1995 06:20:12 GMT Organization: P & L Systems Sender: news@seer.demon.co.uk Message-ID: <1995Aug28.062012.23265@seer.demon.co.uk> References: <41qj35$2qg@pdgm.com> In article <41qj35$2qg@pdgm.com> ted@pdgm.com (Edward E. Murphy) writes: > Hi all, > > NewsGrazer75 seems to crash after pressing the "Catch up" cell a few > times. Anyone have a fix for this? Install version 77. Paul -- Paul Lynch (NeXTmail) http://www.plsys.co.uk/~paul (under construction)
From: gfin@chrome.psych.ualberta.ca (Gary Finley) Newsgroups: comp.sys.next.bugs Subject: Re: NewsGrazer Crashes Date: 28 Aug 1995 15:27:07 GMT Organization: (null pointer) Message-ID: <41sn8b$rpq@rover.ucs.ualberta.ca> In article <41qj35$2qg@pdgm.com> Edward E. Murphy wrote: Hi all, > >NewsGrazer75 seems to crash after pressing the "Catch up" cell a few >times. Anyone have a fix for this? > I've found the best possible solution for the many quirks of the good old NewsGrazer. I bought NewsFlash. I've been using this new product for a couple or three weeks now, and it has run like a charm. I have not seen a single hiccup due to bad item headers. NewsFlash is much more stable than NewsGrazer ever was. They have a demo copy on the archives, and the price is great: Licenses are available for these introductory prices: Individual Single User License ..................... $25 5 User Site License .......................................... $50 10 User Site License ........................................ $95 25 User Site License ........................................ $225 Email Chris Wolf (cwolf@best.com) for more details. (The usual disclaimer applies, they don't pay me, I just like their product.) Cheers. -- ---------------------------------------------- Gary Finley, Psychology Dept. Univ. of Alberta Network manager, Web manager, and postmaster. gfin@psych.ualberta.ca (NeXTmail welcome) http://web.psych.ualberta.ca/staff_bios/finleyg.htmld/index.html
Newsgroups: comp.sys.next.bugs From: wiemer@hrz.uni-bielefeld.de ( Sebastian Wiemer) Subject: Edit.app bug Sender: news@hermes.hrz.uni-bielefeld.de (News Administrator) Message-ID: <DE0svu.7o@hermes.hrz.uni-bielefeld.de> Date: Mon, 28 Aug 1995 12:19:06 GMT Organization: Universitaet Bielefeld, Rechenzentrum Edit.app: The following is a (harmless) bug: Type any character in a new (clear line). Press enter twice. Move the cursor to the right side of the typed charcter and then move the cursor down one line (with the DOWN KEY not with the RIGHT KEY!). The cursor stands now one character right from the correct position. It looks as if there were a space, but it isn't. This is annoying some times. Sebastian Wiemer
From: cwolf@best.com (Christopher A. Wolf) Newsgroups: comp.sys.next.bugs Subject: Re: NewsGrazer Crashes Date: 28 Aug 1995 22:33:14 GMT Organization: WolfWare Distribution: world Message-ID: <41tg7a$esu@shellx.best.com> In article <41sn8b$rpq@rover.ucs.ualberta.ca> Gary Finley wrote:oi >I've found the best possible solution for the many quirks of the good >old NewsGrazer. I bought NewsFlash. I've been using this new product >for a couple or three weeks now, and it has run like a charm. I have >not seen a single hiccup due to bad item headers. NewsFlash is much >more stable than NewsGrazer ever was. They have a demo copy on the >archives, and the price is great: > >Licenses are available for these introductory prices: > >Individual Single User License ..................... $25 >5 User Site License .......................................... $50 >10 User Site License ........................................ $95 >25 User Site License ........................................ $225 > >Email Chris Wolf (cwolf@best.com) for more details. >(The usual disclaimer applies, they don't pay me, I just like their >product.) Cheers. Thank you for the testimonial. Our new preferred new internet addresses are: e-mail: support@wolfware.com ftp: ftp.wolfware.com web: www.wolfware.com Product information and the latest demonstration version (1.66) are available on our web and ftp sites. Sincerely, Christopher Wolf WolfWare -- Christopher Wolf / WolfWare cwolf@wolfware.com (NeXTmail & MIME accepted, finger for PGP key) For information about the NewsFlash newsreader for NeXTSTEP check out WolfWare's home page: http://www.wolfware.com
Newsgroups: comp.sys.next.bugs From: tom@basil.icce.rug.nl (Tom Hageman) Subject: Re: Bug in NS 3.3 find (intel, not motorola) Message-ID: <DDxLq3.M9@basil.icce.rug.nl> Originator: uubasil@obelix.icce.rug.nl Sender: news@basil.icce.rug.nl (NEWS pusher) Organization: Warty Wolfs References: <41kve8$r4i@news.doit.wisc.edu> <41lk2j$bub@news.cs.tu-berlin.de> Date: Sat, 26 Aug 1995 18:51:39 GMT kiwi@cs.tu-berlin.de (Axel Habermann) wrote: > giddings@menominee.chem.wisc.edu wrote: > : The find command does not appear to work correctly on NS/Intel 3.3. Here is an example (I got the same result from a makefile or running this from csh): > : find ./* -type d -prune -exec sh -c 'cd {} ; ls' \; > : On an intel machine, the output is: > : sh: {}: bad directory > [...] > : In the _same_ directory accessed via NFS from a motorola box, the _same_ command results in the directory listings for the subdirectories, as it should. > : Anybody know why there is this difference? > Just tried it on both Intel and Moto. Works as it should. > Maybe something broken with your sh? NeXT-supplied /usr/bin/find fails with this command line. GNU find works. -- __/__/__/__/ Tom Hageman <tom@basil.icce.rug.nl> [NeXTmail/Mime OK] __/ __/_/ IC Group <tom@icgned.nl> (work) __/__/__/ "...to baldly go where no one has gone before." __/ _/_/ -- star trek TNG -- __/__/__/__/ Tom Hageman <tom@basil.icce.rug.nl> [NeXTmail/Mime OK] __/ __/_/ IC Group <tom@icgned.nl> (work) __/__/__/ "...to baldly go where no one has gone before." __/ _/_/ -- star trek TNG
Newsgroups: comp.sys.next.bugs From: sb6fb@rivm.nl (Francois Bourgeois) Subject: Problem with jotting service of Librarian Message-ID: <DE2Ho8.MsK@rivm.nl> Keywords: jotting Librarian Sender: news@rivm.nl Organization: Rijksinstituut voor Volksgezondheid en Milieuhygiene, Bilthoven, NL Date: Tue, 29 Aug 1995 10:12:08 GMT I have a problem with the jotting service of Librarian. After I have opened a bookshelf with Librarian, only the jotting services of targets in the most recently opened booshelf are visible under the Services menu in other applications. I have found only one way to provide the jotting service of _all_ relevant targets to other applications: 1. Start Librarian 2. Open all bookshelves with targets that have the jotting service activated Librarian can than be left activated, hided or you can quit. That makes no difference. Anyone that have a solution for this problem? BTW: Is there a list of "known NeXTSTEP bugs" ? -- RIVM - National Institute of Public Health and Environmental Protection Francois Bourgeois, postbak 15 | e-mail: sb6fb@rivm.nl Risk Assessment Division | e-mail: F.Bourgeois@rivm.nl P.O.Box 1, 3720 BA Bilthoven | voice : +31 30 742962 The Netherlands | fax : +31 30 291492
From: otto@tukki.jyu.fi (Otto J. Makela) Newsgroups: comp.sys.next.bugs Subject: Re: NewsGrazer Crashes Date: 29 Aug 1995 17:03:43 +0300 Organization: University of Jyvaskyla, Finland Sender: otto@tukki.cc.jyu.fi Message-ID: <OTTO.95Aug29170341@tukki.jyu.fi> References: <41qj35$2qg@pdgm.com> In-reply-to: ted@pdgm.com's message of 27 Aug 1995 20:03:49 GMT In article <41qj35$2qg@pdgm.com> ted@pdgm.com (Edward E. Murphy) writes: > NewsGrazer75 seems to crash after pressing the "Catch up" cell a few > times. Anyone have a fix for this? I've given up on NewsGrazer and switched over to Alexandra, which is copyleft and supports ISO 8859.1 character sets. -- /* * * Otto J. Makela <otto@jyu.fi> * * * * * * * * * * * * * * * * * * */ /* Phone: +358 41 613 847, BBS: +358 41 211 562 (V.32bis/USR-HST,24h/d) */ /* Mail: Cygn.k.7 E 46/FIN-40100 Jyvaskyla/Finland, ICBM: 62.14N 25.44E */ /* * * Computers Rule 01001111 01001011 * * * * * * * * * * * * * * * * */
From: celesteh@mills.edu Newsgroups: comp.sys.next.bugs Subject: Thinking of buying NS? --DON'T! Date: 30 Aug 1995 23:16:10 GMT Organization: University of California, Berkeley Message-ID: <422rfq$jeo@agate.berkeley.edu> A while ago, i found a listing for an older version of NS, cheap. I bought it immediately, but not wanting to void the return policy, I called NeXT tech support before I openned it to make sure my SCSI card was supported. I got a call back from a marketing type who incorrectly told me that my card was ok. As soon as I discovered that it wasn't going to work, I left voice mail messages that were ignored. Finally, I wised up and left my school NeXTmail account and someone got back to me. Here is the email I recieved. My messages are quoted at the bottom of the other guy's messages. this is our actual correspondence: From: Rick Collison <Rick_Collison@NeXT.COM> Date: Mon, 24 Apr 95 12:04:35 -0700 To: celesteh@gershwin.mills.edu Subject: Support You best bet is to call OTI (our academic reseller) to help you out with your driver issues. They are at 1-800-452-7608 Also, who told you (name) that your specific card was supported? tx Rick -------- From: Rick Collison <Rick_Collison@NeXT.COM> Date: Mon, 24 Apr 95 14:15:33 -0700 To: Celeste Marie Hutchins <celesteh@gershwin.mills.edu> Subject: Re: Support Cc: jim_swails@objectech.com References: <9504242207.AA00956@gershwin.mills.edu> Celeste, Let's see what Object Technology can do, and then go from there. Rick Begin forwarded message: From: Celeste Marie Hutchins <celesteh@gershwin.mills.edu> Date: Mon, 24 Apr 95 06:07:22 -0800 To: Rick Collison <Rick_Collison@NeXT.COM> Subject: Re: Support Thank you for getting back to me. I can't remeber the name of the person who misinformed me, but I think it was just a misunderstanding. Tech support told her that it would work as a sound card and she (being a sales type) probably didn't really understand scsi and all that. No offence to sales types. However, if I am able to get the issue about the drivers solved, I'd like to ask for an aditional 30 days tech support for help during installation. And if I can't get the driver issue solved, I want to get a refund. Thank you for taking the time to reply. Celeste --------- From: Rick Collison <Rick_Collison@NeXT.COM> Date: Mon, 24 Apr 95 16:09:02 -0700 To: celesteh@gershwin.mills.edu Subject: Re: Support References: <9504242207.AA00956@gershwin.mills.edu> Celeste, I just tried to call you and get an update..... Can you tell me if Object Technology has contacted you? Rick --------- From: Rick Collison <Rick_Collison@NeXT.COM> Date: Wed, 26 Apr 95 18:10:31 -0700 To: Celeste Marie Hutchins <celesteh@gershwin.mills.edu> Subject: Re: Support References: <9504262203.AA02634@gershwin.mills.edu> a couple of questions... 1. What version of NeXTSTEP are your trying to get up and running? 2. 3.1 is "years" old, why is this coming up now? 3. Do you have receipts from your original order? Rick Begin forwarded message: From: Celeste Marie Hutchins <celesteh@gershwin.mills.edu> Date: Wed, 26 Apr 95 06:02:51 -0800 To: Rick Collison <Rick_Collison@NeXT.COM> Subject: Re: Support I talked to OTI last night. There is no support available for any scsi drivers i have. The person I talked to was very helpful, but was trying to sell me an adaptech scsi card. I don't have any money for hardware upgrades right now and would just like to recieve a refund for the $100 I spent on NS 3.1. Celeste ------------ From: Rick Collison <Rick_Collison@NeXT.COM> Date: Fri, 28 Apr 95 14:18:35 -0700 To: Celeste Marie Hutchins <celesteh@gershwin.mills.edu> Subject: Re: Support References: <9504282208.AA03955@gershwin.mills.edu> It sounds like you purchased a copy from someone privately - did you get the CD and all the documentation? It doesn't seem fair that NeXT should be responsible for refunding you your money if you didn't purchase it from us. Rick Begin forwarded message: From: Celeste Marie Hutchins <celesteh@gershwin.mills.edu> Date: Fri, 28 Apr 95 06:07:45 -0800 To: Rick Collison <Rick_Collison@NeXT.COM> Subject: Re: Support >1. What version of NeXTSTEP are your trying to get up and running? >2. 3.1 is "years" old, why is this coming up now? What do you mean by your first question? It is NeXTSTEP 3.1 for white (intel) hardware. This is coming up now because I liked NS (i work as a lab assistant for NeXTs) and wanted to support it, but didn't have a lot of money to spend on software. Somebody emailed me an add for an older, cheap version. The computers I work with are only running 3.0, so i figured being a couple of upgrades behind wasn't that big of a deal. My father is looking for the reciept. He says he can find it. -------------- From: Rick Collison <Rick_Collison@NeXT.COM> Date: Tue, 2 May 95 10:18:53 -0700 To: Celeste Marie Hutchins <celesteh@gershwin.mills.edu> Subject: Re: Support References: <9505012225.AA05680@gershwin.mills.edu> Just wondering.... if you are unable to get NeXT up and running our your system, how are you able to send/receive NeXT mail? Rick Begin forwarded message: From: Celeste Marie Hutchins <celesteh@gershwin.mills.edu> Date: Mon, 1 May 95 06:25:21 -0800 To: Rick Collison <Rick_Collison@NeXT.COM> Subject: Re: Support NeXTmail doesn't want to deal with included stuff right now, so heres another try ay this message: I got it in an unopened shrink wrapped box. According to the list of included stuff, i got everything. It wasn't a private purchase. this is the add i recieved: (the add is cut and pasted to the bottom instead to get this to send) >It doesn't seem fair that NeXT should be responsible for refunding you your money >if you didn't purchase it from us. When I recieved the included thing above. I sent my dad out to get NS 3.1 from the Palo Alto store. As soon as i came home to see my folks again, i was going to install it. In the mean time, I went to look at the hardware compatibility guide at www.next.com. That page is very complicated and difficult for someone who is not a hardware pro to read. I made an effort to research whether or not my scsi card would work BEFORE i voided the return policy by removing the shrink wrap. So I called 1-800-try-NeXT and the sales person i talked to gave me tech support's number and told me to call them. i left voice mail with tech support and the NeXT day i recieved voicemail from someone in the sales department who told me that my Pro Audio 16 *was* supported as long as it was a Pro Audio Spectrum or better. I then, just to make sure, looked at the side of the NeXT box to check for scsi compatibility but saw nothing very helpful. So I tried to install it and it went into system panic. I thought that was weird and so i posted to comp.sys.next.bugs and asked if anybody knew what was wrong. I got mail saying my scsi card wasn't actually supported. So I called tech support and explained all this and they said that c.s.n.b was right and transferred me someplace else where i left voice mail. Nobody got back to me so i tried leaving more voicemail there and in other places on your telephone sytem and still nobody got back to me for over a week. Finally i tried leaving my NeXTmail address as well as my phone number and only then did i get a reply. Somebody made a mistake when they misinformed me about my hardware, but the way I've been ignored since was no accident. I'm only asking for $100 back. You've already lost enough sales by me talking to people at school about your company to more than make up for it. I have every piece of what i initially got with my purchase down to the shrink wrap. Celeste the add: -------------------------------------------------------------------------- ----------------------- Path: agate!howland.reston.ans.net!gatech!udel!news.mathworks.com!uunet!in1.uu.n et!maple.enet.net!usenet From: terence@sfsc.com (Terence Liow) Newsgroups: comp.sys.next.marketplace,comp.sys.next.software Subject: $99 *new* NS 3.1 for 80x86 Date: 27 Feb 1995 03:49:51 GMT Organization: w Lines: 17 Message-ID: <3iri4v$pvd@maple.enet.net> NNTP-Posting-Host: sfschost.enet.net X-Newsreader: WinVN 0.92.5 Xref: agate comp.sys.next.marketplace:10594 comp.sys.next.software:18820 A store I have dealt with for a few mail-order purchases seems to have gotten its hand on NextStep 3.1 for Intel and they're selling them for only $99 a copy. Great deal for anyone who wants to try Next Step and don't want to pay full sticker for it. I've dealt with Access Comp. Tech. and they are reliable. SHipped monitors to Northern Virginia and Nevada for me. Address: 2225 El Camino Real Santa Clara CA (408) 247-4444 2080 El Camino Real Palp Alto CA (415) 813-0600 -------------------------------------------------------------------------- ------------------------ ------------ A friend of mine who is a NeXT computer owner then sent this on my behalf, after my additional responces were ignored: From: Jessica L Mosher <cooncat@wombat> Message-Id: <9505172056.AA03981@wombat.mills.edu> To: Rick_Collison@NeXT.COM Subject: Celeste Hutchins purchase of NeXTstep for Intel Cc: celesteh Status: R I have seen a copy of the email that Celeste Hutchins sent you about her copy. I was the assistant NeXT system administrator at Mills for three years. If you check your records, you will find that Mills purchased 20 NeXT computers from the period of 1989- 1992, and that some of the members of the tech support staff who are at NeXT have been to Mills for onsite maintenance. All of the systems are also registered with Bell Atlantic. The purpose of this email is to reassure you that gershwin, mingus, etc. are indeed college-owned NeXT computers. If you examined the header of the NeXTmail, you would realize that the release correlates to NS release 3.0, hence, I believe it is impossible for the mail to be the software that Ms. Hutchins purchased for the Intel platform. ----------- and that's it. I never heard from them again. So, if anyone is interested, i have a never-used copy of NS available for sale. Please send email. -- _______________________________________________________________ | | | Celesteh@mills.edu NeXTmail:Celesteh@gershwin.mills.edu | | http://www.mills.edu/~celesteh/celesteh.homepage.html | |_____________________________________________________________| -- _______________________________________________________________ | | | Celesteh@mills.edu NeXTmail:Celesteh@gershwin.mills.edu | | http://www.mills.edu/~celesteh/celesteh.homepage.html | |_____________________________________________________________|
From: David Kelman Newsgroups: comp.sys.next.bugs Subject: Re: NewsGrazer Crashes Date: 30 Aug 1995 23:22:14 GMT Organization: MountainNet, Inc. Morgantown WV 800.444.1458 Distribution: world Message-ID: <422rr7$4uu@news.mountain.net> References: <OTTO.95Aug29170341@tukki.jyu.fi> In article <OTTO.95Aug29170341@tukki.jyu.fi> otto@tukki.jyu.fi (Otto J. Makela) writes: > In article <41qj35$2qg@pdgm.com> ted@pdgm.com (Edward E. Murphy) writes: > > > NewsGrazer75 seems to crash after pressing the "Catch up" cell a few > > times. Anyone have a fix for this? > > I've given up on NewsGrazer and switched over to Alexandra, which is copyleft > and supports ISO 8859.1 character sets. > -- There is a new version of NewsGrazer, v77, that is quad fat, and I haven't had it crash on me once yet. It will also follow crossposts and mark as read posts in one group that were in another group that you "caught up' on. It is available at the Peanuts archive. David Kelman sacremon@access.mountain.net (NeXTMail preferred)
From: Newsgroups: comp.sys.next.bugs Subject: Re: lookupd Date: 30 Aug 1995 23:23:15 GMT Organization: Minnesota Regional Network (MRNet) Message-ID: <422rt3$dq4@news.mr.net> References: <41frag$9r6@usc.edu> <DDzBpz.xs@murdoch.acc.Virginia.EDU> > Yes, I would also like to know if next will release these patches to > non-premium support customers like most of us! > -Paul we only got our copy here because we have 50 licenses and we hit a bug that the patch fixes. you say you only have 1 license? dream on. perhaps the patch will find it's way to an anonymous ftp site near you. who can tell with these things? ben
From: Matthew Reichman Newsgroups: comp.sys.next.bugs Subject: Re: lookupd Date: 31 Aug 1995 08:10:57 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-d-19.usc.edu Message-ID: <423qqh$ecb@usc.edu> References: <41frag$9r6@usc.edu> <DDzBpz.xs@murdoch.acc.Virginia.EDU> <422rt3$dq4@news.mr.net> > perhaps the patch will find it's way to an anonymous ftp site near > you. who can tell with these things? Well, Ben...... can you tell? :) -- Matthew ====================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV | PGP key upon request "Brooding's the word. Saw him kick the rose bush, kick the green ferns by the porch, decide against kicking the apple tree. God made it too firm. There, he just jumped on a dandelion."
From: root@terra.crystalengine.com(Felipe A. Rodriguez) Newsgroups: comp.sys.next.bugs Subject: Re: lookupd Date: 31 Aug 1995 09:23:35 GMT Organization: Network Intensive Message-ID: <423v2n$okj@ni1.ni.net> References: <422rt3$dq4@news.mr.net> In article <422rt3$dq4@news.mr.net> writes: >> Yes, I would also like to know if next will release these patches to >> non-premium support customers like most of us! > >> -Paul > >we only got our copy here because we have 50 licenses and we hit a bug that >the patch fixes. you say you only have 1 license? dream on. > >perhaps the patch will find it's way to an anonymous ftp site near you. >who can tell with these things? > > >ben The fact that there is a major flaw in the OS due to lookupd and that they are withholding the available patch has really irked me of late. Come on NeXT release the lookupd patch. -- Felipe A. Rodriguez # ...it cannot be called ingenuity to kill Agoura Hills, CA # one's fellow citizens, to betray # friends, to be without faith, without root@terra.crystalengine.com # mercy, without religion; by these means (NeXTmail preferred) # one can acquire power but not glory. (MIMEmail welcome) # --Nicolo Machiavelli
From: robertn@seahawk.nwest.mccaw.com (Robert Nicholson) Newsgroups: comp.sys.next.bugs Subject: Re: lookupd Date: 31 Aug 1995 02:22:31 -0700 Organization: McCaw Cellular Communications, Inc. Sender: robertn@seahawk.nwest.mccaw.com Message-ID: <h2but6fmw8.fsf@seahawk.nwest.mccaw.com> References: <41frag$9r6@usc.edu> <DDzBpz.xs@murdoch.acc.Virginia.EDU> <422rt3$dq4@news.mr.net> In-reply-to: 's message of 30 Aug 1995 23:23:15 GMT > Yes, I would also like to know if next will release these patches to > non-premium support customers like most of us! > -Paul we only got our copy here because we have 50 licenses and we hit a bug that the patch fixes. you say you only have 1 license? dream on. perhaps the patch will find it's way to an anonymous ftp site near you. who can tell with these things? I simply do not understand the mentality behind no releasing a patch. Surely they will put something on nextanswers right? What reason could you give for them not doing this? ben -- The views and opinions expressed in this article are those of the poster and not his employer.
From: Garance A Drosehn <gad@eclipse.its.rpi.edu> Newsgroups: comp.sys.next.bugs Subject: Re: NewsGrazer Crashes Date: 31 Aug 1995 03:33:24 GMT Organization: Rensselaer Polytechnic Institute, Troy NY, USA Message-ID: <423ai4$js0@usenet.rpi.edu> References: <41qj35$2qg@pdgm.com> ted@pdgm.com (Edward E. Murphy) wrote: > Hi all, > NewsGrazer75 seems to crash after pressing the "Catch up" cell > a few times. Anyone have a fix for this? More specifically, pressing the "Catch Up" button will cause a crash if there are any articles in the newsgroup that it has not retrieved the header-info for. So, if you just make sure to wait for all headers to be displayed, you can click on "Catch Up" all you want without any problems. A much better solution, of course, is to get the newer version of Newsgrazer, which was just released in the past month. It fixes this bug, and a few other minor bugs. NewsGrazer still isn't a supported app, but if you want NewsGrazer you should use the new version. You can get the latest version off some page on the NeXT web site: http://www.next.com/ Or you can pick up the binaries via ftp: ftp://eclipse.its.rpi.edu/NeXT/usenet/NewsGrazer_f (which is a folder of archives, one quad-fat binary and four single-architecture binaries). My own recommendation is that you check out some of the other newsreaders that are now available. There's Alexandra, NutNews, Kiwi, and some others that I don't remember right now. You'd be better off with something that's actively supported... --- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
From: Robert Worne <rworne@primenet.com> Newsgroups: comp.sys.next.bugs Subject: NSFIP boot disks on www.next.com Date: 30 Aug 1995 16:51:10 GMT Organization: A Big, Black Box. Message-ID: <4224tu$7s0@nnrp3.primenet.com> References: <41frag$9r6@usc.edu> <DDzBpz.xs@murdoch.acc.Virginia.EDU> Has anyone successfully made the 3.3 driver floppy? I attempted to make this many times, but it just bombs with a broken PS/2 keyboard driver & hoses the machine when loading the initial drivers. The sum of the downloaded version does check out, and these were created on my NeXT Cube. The boot disk works flawlessly. -- //--------------------------------------------------------------------- // Robert Worne rworne@primenet.com OS/2-NeXT -=Starving CS Undergrad=- //--------------------------------------------------------------------- // Microsoft Network is prohibited from redistributing this work in // any form, in whole or in part. Copyright, Robert Worne, 1995. // License to distribute this post is available to Microsoft for $1000. // Posting without permission constitutes an agreement to these terms. // Please send notices of violation to rworne@primenet.com and // postmaster@microsoft.com
From: heller@attila.imo.physik.uni-muenchen.de (Helmut Heller) Newsgroups: comp.sys.next.bugs Subject: Re: Bug in NS 3.3 find (intel, not motorola) {find codes DB not working} Date: 31 Aug 1995 19:51:49 GMT Organization: Leibniz-Rechenzentrum, Muenchen (Germany) Distribution: world Message-ID: <4253sl$pi9@sparcserver.lrz-muenchen.de> References: <41onrn$p2a@news-e1a.megaweb.com> In article <41onrn$p2a@news-e1a.megaweb.com> pierre@nova.org (Thomas Piergallini) writes: > The find codes database is non-functional for me (NSFIP). I run updatedb > out of /usr/lib/find no problem, it creates the find.codes database in > /etc. But when you execute the find command, no output is returned. Any > ideas? On black boxes this is not a problem, and it works just fine. It > did not work in 3.2 on intel either. For some reason unknown to me, the (quick) find syntax is different between intel and m68k based machines: on m68k you can say: find blabla and it will return any file with blabla as a substring somewhere in its name or path on intel you have to type: find \*blabla\* to get the same behaviour (don't forget to escape the asterisk!) by default, intel-based NS tries to make an EXACT match of the argument, including the complete path! BTW: find \*blabla\* also works on m68k machines... -- Servus, Helmut (DH0MAD) ______________NeXT-mail welcome_________________ FAX: +49-89-2394-4607 "Knowledge must be gathered and cannot be given" heller@attila.imo.physik.uni-muenchen.de ZEN, one of BLAKES7 Phone: +49-89-2394-4565 ------------------------------------------------ Helmut Heller, Ludwig Maximilians University | G i b D O S | Institute for Medical Optics, Theoretical Biophysics Group | k e i n e |
From: steved@bankone.com Newsgroups: comp.sys.next.bugs Subject: Re: Bug in NS 3.3 find (intel, not motorola) {find codes DB not working} Date: 31 Aug 1995 18:50:04 GMT Organization: Bank One Distribution: world Message-ID: <42508s$dgm@natasha.bankone.com> References: <41onrn$p2a@news-e1a.megaweb.com> In article <41onrn$p2a@news-e1a.megaweb.com> pierre@nova.org (Thomas Piergallini) writes: > The find codes database is non-functional for me (NSFIP). I run updatedb > out of /usr/lib/find no problem, it creates the find.codes database in > /etc. But when you execute the find command, no output is returned. Any > ideas? On black boxes this is not a problem, and it works just fine. It > did not work in 3.2 on intel either. > Ditto NS/HPPA.... Steve
From: steved@bankone.com Newsgroups: comp.sys.next.bugs Subject: Re: Bug in NS 3.3 find (intel, not motorola) {find codes DB not working} Date: 31 Aug 1995 18:50:28 GMT Organization: Bank One Distribution: world Message-ID: <42509k$dgn@natasha.bankone.com> References: <41onrn$p2a@news-e1a.megaweb.com> In article <41onrn$p2a@news-e1a.megaweb.com> pierre@nova.org (Thomas Piergallini) writes: > The find codes database is non-functional for me (NSFIP). I run updatedb > out of /usr/lib/find no problem, it creates the find.codes database in > /etc. But when you execute the find command, no output is returned. Any > ideas? On black boxes this is not a problem, and it works just fine. It > did not work in 3.2 on intel either. > Ditto NS/HPPA.... Steve --- Steve Dieringer NeXTMail: steved@bankone.com Group Product Manager, Electronic Commerce Product & Technology Management, 1140 Bank One, Columbus, NA Columbus, OH 43271-1140 Phone: (614) 248-3019
From: steved@bankone.com Newsgroups: comp.sys.next.bugs Subject: Re: Bug in NS 3.3 find (intel, not motorola) {find codes DB not working} Date: 31 Aug 1995 18:50:59 GMT Organization: Bank One Distribution: world Message-ID: <4250aj$dgo@natasha.bankone.com> References: <41onrn$p2a@news-e1a.megaweb.com> In article <41onrn$p2a@news-e1a.megaweb.com> pierre@nova.org (Thomas Piergallini) writes: > The find codes database is non-functional for me (NSFIP). I run updatedb > out of /usr/lib/find no problem, it creates the find.codes database in > /etc. But when you execute the find command, no output is returned. Any > ideas? On black boxes this is not a problem, and it works just fine. It > did not work in 3.2 on intel either. > Ditto NS/HPPA.... Steve --- Steve Dieringer NeXTMail: steved@bankone.com Group Product Manager, Electronic Commerce Product & Technology Management, 1140 Bank One, Columbus, NA Columbus, OH 43271-1140 Phone: (614) 248-3019
From: rragner@panther.vm.iastate.edu (Rod Ragner) Newsgroups: comp.sys.next.bugs Subject: Re: lookupd Date: 31 Aug 1995 21:26:06 GMT Organization: Iowa State University, Ames, Iowa Message-ID: <4259de$ol9@news.iastate.edu> References: <h2but6fmw8.fsf@seahawk.nwest.mccaw.com> In article <h2but6fmw8.fsf@seahawk.nwest.mccaw.com> robertn@seahawk.nwest.mccaw.com (Robert Nicholson) writes: > > we only got our copy here because we have 50 licenses and we hit a bug that > the patch fixes. you say you only have 1 license? dream on. We have 41 (all be them academic) licenses and we have not received the patches. BUT, we do not pay anything for support (therefore we do not get any support), so... > > I simply do not understand the mentality behind no releasing a patch. > > Surely they will put something on nextanswers right? I also hope they put it on NeXTanswers. > > What reason could you give for them not doing this? > Maybe they are just too busy. I do not mean that cynically - from what I hear, they have a lot of work to do to deliver the products that they have promised in the next six months. -- Rod Ragner, UNIX Systems Administrator/NEXTSTEP Application Developer Veterinary Diagnostic Laboratory, College of Veterinary Medicine, Iowa State University, 2630 Vet. Med. Bldg., Ames, Iowa 50011 Voice: (515) 294-4751, FAX: (515) 294-6961 or 3564, (NeXT Mail accepted) Email: rragner@stallion.vm.iastate.edu or stryder@iastate.edu
Newsgroups: comp.sys.next.bugs From: shayman@Objectario.com (Steve Hayman) Subject: Re: Bug in NS 3.3 find (intel, not motorola) {find codes DB not working} Message-ID: <1995Sep1.025527.27823@objectario.com> Sender: news@objectario.com Organization: Steve Hayman + Associates / NeXTSTEP Consulting / Toronto References: <42509k$dgn@natasha.bankone.com> Date: Fri, 1 Sep 1995 02:55:27 GMT In article <42509k$dgn@natasha.bankone.com> steved@bankone.com writes: > In article <41onrn$p2a@news-e1a.megaweb.com> pierre@nova.org (Thomas > Piergallini) writes: > > The find codes database is non-functional for me (NSFIP). I run updatedb > > out of /usr/lib/find no problem, it creates the find.codes database in > > /etc. But when you execute the find command, no output is returned. I'm not sure I can explain why this is, but find foo Works on black hardware find '*foo*' Works on black, Intel, HP. So if you can get used to using the extra *'s, the fast-find database works OK. Hope this helps, Steve
Newsgroups: comp.sys.next.bugs From: dalpen@JSP.UMontreal.CA (DALPE Nathalie) Subject: HELP error message while printing Message-ID: <DE7J81.I9n@cc.umontreal.ca> Sender: news@cc.umontreal.ca (Administration de Cnews) Organization: Universite de Montreal Date: Fri, 1 Sep 1995 03:33:36 GMT I have post this problem to comp.sys.next.hardware but I didn't received any reply. Let's try in this group... I'm trying to print to a FUJITSU DL900 using the "EPSON 510 / NeXT" emulation from print manager. I ask print manager to print the 'Test page'. Printing start, everything is fine. But somewhere in the middle of the page (not always at the same place) the printer stop and I have the following error message that appear in the console: sz1: lost xmit HW interrupt There's no way I can restart the printer. Also when that appends, I have trouble shuting down my NeXTstation. I have to press the 'power' button twice and when I restart the computer it always do a complete HD check ?!? Here's my setup: NeXTstation 25Mhz 20Meg ram/ 406 megs HD N.S. 3.0 No network, no NeXT printer, no modems. I hope it is not and hardware problem... Thank you, Francois Lanciault
From: Gerald Wildgruber Newsgroups: comp.sys.next.bugs Subject: State of art in the NeXT dump/restore question ? Date: 3 Sep 1995 20:06:44 GMT Organization: Leibniz-Rechenzentrum, Muenchen (Germany) Distribution: world Message-ID: <42d1sk$hvb@sparcserver.lrz-muenchen.de> Hello, this is not for renewing the well known discussion on the dump/restore problem with NeXT. I just want to ask if any document meanwhile appeared to resume this question. I consulted NeXTAnswers and the whole of comp.sys.next.*, archived and indexed (since 89) on the Peanuts NeXT CD; in short, the problem was to read dump tapes with the restore command: a solution oftenly given was the mtset program wich fixes the blocksize of tape devices; in NeXTAnswers I then could read that the problem should definitely disappear with the SCSI tape driver ver.3.31. Another posting mentioned the newly implemented stblocksize program, shipped with NeXTSTEP 3.3. ; it is run from /etc/rc to atomatically fix the block size. Thus mtset seems no longer necessary as it has been before. Given all these hints, restore still doesn't work for me : I'm running NeXTSTEP 3.3 for Intel and I'm using the HP35470A DAT Streamer. With no mtset command given before (because of stblocksize run during boot time), I get the well known messages: root@ue801be:/tmp> /usr/etc/restore if /dev/nrst0 Tape/disk read error: I/O error I repeat this twenty times and I get: root@ue801be:/tmp> /usr/etc/restore if /dev/nrst0 Tape/disk block size (512) is not a multiple of dump block size (1024) Again I'm repeating this command about twenty times and suddenly restore begins to work with this message: root@ue801be:/tmp> /usr/etc/restore if /dev/nrst0 Note: Doing Quad swapping (...) restore > ls From there on restoring is possible but incredibly slow. I also tried Markus Wenzel's proposal: >Anyway, the better solution is to dump in variable size >mode (no mtset, or mtset without any option) and restore after doing a >'mtset -i', which means "ignore illegal length reports and continue >restore". It just works fine for me and my WangDAT but without improvement in my unhappy case. Another question is the length parameter -s for the dump command; I couldn't find any systematic answer on how to determine this number. Apparently backup with dump/restore works, but I imagine typing x times the whole command in single user mode (without history-function), desperately trying to restore the whole NeXT partition. Is there any solution beyond this perhaps-perhaps-not basis ? Any help would be greatly appreciated.. Gerald.
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 4 Sep 1995 04:15:10 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <42duge$10k@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
Newsgroups: comp.sys.next.bugs From: rbp@investor.pgh.pa.us (Bob Peirce #305) Subject: Re: NewsGrazer Crashes Message-ID: <1995Sep5.163223.12365@investor.pgh.pa.us> Date: Tue, 5 Sep 95 16:32:23 GMT References: <OTTO.95Aug29170341@tukki.jyu.fi> <422rr7$4uu@news.mountain.net> Organization: Cookson, Peirce & Co., Pittsburgh, PA In article <422rr7$4uu@news.mountain.net> sacremon@access.mountain.net writes: >In article <OTTO.95Aug29170341@tukki.jyu.fi> otto@tukki.jyu.fi (Otto J. >Makela) writes: >> In article <41qj35$2qg@pdgm.com> ted@pdgm.com (Edward E. Murphy) writes: >> >> > NewsGrazer75 seems to crash after pressing the "Catch up" cell a few >> > times. Anyone have a fix for this? >> >> I've given up on NewsGrazer and switched over to Alexandra, which is >copyleft >> and supports ISO 8859.1 character sets. >> -- > > There is a new version of NewsGrazer, v77, that is quad fat, and I >haven't had it crash on me once yet. It will also follow crossposts and >mark as read posts in one group that were in another group that you >"caught up' on. It is available at the Peanuts archive. > I was running an old version and it worked fine. Then, all of a sudden it started to crash. I got v77 and it does the same thing. I don't know why it worked so long and now it doesn't, but I too am using Alexandra. NG has some nice features, but not if the thing blow up! -- Bob Peirce Pittsburgh, PA 412-471-5320 rbp@investor.pgh.pa.us [OFFICE] me@venetia.pgh.pa.us [HOME (NeXT)] There is only one basic human right, the right to do as you damn well please. And with it comes the only basic human duty, the duty to take the consequences. -- P.J. O'Rourke
From: alancook@cyberstore.net Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: DEC XL590 & Panic Windows Date: Wed, 6 Sep 1995 11:34:05 GMT Organization: Cyberstore Systems, Inc. Message-ID: <950906043405.257AAD4E.alancook@marketel> Mime-Version: 1.0 (Generated by Eloquent) Content-Type: text/plain; charset=US-ASCII Does anyone have a DEC XL590 and is having the system crash on a regualr basis (either freezing or bringing upo panic windows). If so could you e- mail about it or if you have fixed the problem, tell me what you have done. Alan Cook alancook@cycberstore.net
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.sys.next.hardware Subject: Help, can't mount rootdev anymore! Date: Thu, 7 Sep 1995 16:10:42 +0200 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.950907160258.15744A-100000@hphalle6a.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello altogether, Due to a stupid bug in the Microsoft installation software, my seperate (although not mounted) NeXTSTEP root drive got formated for about 2-3%. After I broke the formatting procedure, I was still able to boot NEXTSTEP! YES!, but the system gets into panic if it tries to mount rootdev. So I picked up my boot disks, and CD-Rom, went into single user mode and tried everything I knew of. I tried fsck: Fail (no valid disk label) I tried disk for writing the label (fail, no valid disk label and I/O error) I tried disk to scan for superblocks (fail I/O error) But the data is still there! The system can read the configuration data etc.! Now here is my question: How can I access the hidden data on an NEXTSTEP drive without mounting it? It seems to me the last solution to retrieve the data from my disk is not mounting it. Any help would be very much appreciated, Many thanks in advance, Boerny. -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/peanuts/ scholz@c86501.rm.op.dlr.de http://www.leo.org/~scholz/
From: samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) Newsgroups: comp.sys.next.bugs Subject: Quantrix 2.0 major bug Date: 08 Sep 1995 18:02:42 GMT Organization: School of Computer Science, McGill Univ. Distribution: world Message-ID: <SAMURAI.95Sep8140242@maggie.cs.mcgill.ca> Summary: error in linking tables prevents some operations. (1) Create a new model. (2) Create a new table. (3) In the first table, for column B1, enter the following values: B1 A1 one A2 two A3 three (4) In the second table, enter the following formula: B1 = Table 1::B1 (5) This is what you get. B1 A1 one A2 one A3 one A4 one (6) What you should have gotten was this: B1 A1 one A2 two A3 three A4 <undefined> Again, not only is this completely wrong (it should be "one" "two" "three", as in the first table); another error is shown in that there's a mismatch in the number of rows in the B1 column. Thus, the B1:A4 element should be undefined, and the user notified. Lighthouse has been notified of the bug and states that it will be fixed in the first maintentance release. - db -- "Investing in tech stocks is a high risk." -- Bill Gates, Microsoft Chairman, after incurring a two day paper loss of 2 billion dollars on Microsoft stock.
From: cpayne@optical (Carl Payne) Newsgroups: comp.sys.next.bugs Subject: NetInfo...something new broken? Date: 9 Sep 1995 16:08:07 GMT Organization: Fibernet! Satisfy the need for speed Message-ID: <42se57$la@optical.fiber.net> 3.3 black hardware Using the save netinfo option shuts NetInfo down, never to return without a reboot. Also, why is a non-NetInfo-d computer insisting on waiting for another non NI'd computer to boot before it'll finish booting? They are NOT (underline that a million times) NetInfo connected. -- Carl Payne cpayne@fiber.net <--NeXTMAIL, MIME, ASCII all OK
From: caradoc@neta.com(John Groseclose) Newsgroups: control,comp.sys.next.bugs Subject: cmsg cancel <DEnrH5.2w2@goodnet.com> Control: cancel <DEnrH5.2w2@goodnet.com> Date: 11 Sep 1995 18:16:22 GMT Organization: Royal Society for the Prevention of Spam Sender: jgros@goodnet.com (John Grosclose) Message-ID: <cancel.DEnrH5.2w2@goodnet.com> Death to Spam!
From: art@cubicsol.com (Art Isbell) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: UUCP problem Date: 12 Sep 1995 04:04:10 GMT Organization: Sense Networking Seattle (www.oz.net) Distribution: world Message-ID: <4330rq$qlt@emerald.oz.net> References: <42vq8g$rkg@emerald.oz.net> In article <42vq8g$rkg@emerald.oz.net> art@cubicsol.com (Art Isbell) writes: > We've recently established 2 UUCP over PPP accounts with UUNET. Both > use NeXT's UUCP running in TCP mode over Steve Perkins' ppp2.2.0.1.9. The > PPP link comes up fine on both accounts. One account running NS 3.1 on a > NeXTstation works fine for exchanging email; the other running NS 3.2 on a > Canon object.station 41 never succeeds in exchanging mail. > wmesg 'S' Y > trddata expecting -50200576 bytes <--- check this out!! > trddata expecting 1836020294 bytes > cntrl - -1 > uucp uunet (9/9-20:16-2404) FAILED (conversation complete) Well, these outrageous numbers raised the same flag with a couple of others that it did with me - a byte order bug in the NS 3.2 Intel version of uucico. But I reasoned that such a bug would have been noticed by so many people that it would never have made it into the release. Wrongo!! Apparently, few NS 3.2 for Intel users use UUCP over TCP/IP which uses the 't' protocol instead of the much more common 'g' protocol (well, no one could use it because it flat doesn't work). I installed the NS 3.3 version of uucico which fixed the problem. Boy, did I waste a lot of time dealing with a bug that was never mentioned in the NS 3.3 Release Notes or anywhere else that I can find :-( --- Art Isbell NeXTmail: art@cubicsol.com NeXT Registered Consultant Voice/Fax: +1 408 335 2515 Trego Systems Voice Mail: +1 408 335 1154 CaseServ: NEXTSTEP managed care US Mail: Felton, CA 95018-9442 contract and case management solutions
From: dnelson@core.symnet.net (Dru Nelson) Newsgroups: comp.sys.next.bugs Subject: Urgent BUG, login problems, NS 3.0/lookupd dies soon after boot ???? Date: 11 Sep 1995 18:38:36 GMT Organization: S y m N e t - North Florida Internet Access (info@symnet.net) Message-ID: <431vnc$csk@tempest.symnet.net> References: <42umpg$50j@tempest.symnet.net> <431vcm$cqi@tempest.symnet.net> We are running Nextstep 3.0 on a Mono slab. After it boots and starts up, a minute or two will go by and then the system starts becoming very unfriendly and stops letting people log in. I was lucky enough to get in right after it first booted. (I can still get in now, as well). I noticed that the _lookupd_ process is going away and I cannot determine why. No changes have been made to the system. It leaves a nice core dump in the root directory. I go in with gdb but I cannot determine anything because there aren't any symbol tables. The system is used for shell accounts for users. I have checked all over the place for any modified binaries or scripts and I cannot detect any. This error is showing up more often and I don't believe I have solved it. I am also seeing others with this problem. What is this???? Dru Nelson SymNet
From: dnelson@core.symnet.net (Dru Nelson) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Urgent BUG, login problems, NS 3.0/lookupd dies soon after boot ???? Date: 11 Sep 1995 19:27:44 GMT Organization: S y m N e t - North Florida Internet Access (info@symnet.net) Message-ID: <4322jg$d4c@tempest.symnet.net> References: <42umpg$50j@tempest.symnet.net> <431vcm$cqi@tempest.symnet.net> More info on this. This machine is our mail server and name server (named). No messages are popping up in the logs about errors. This machine serves one NFS export. I am also running NIS (ypserv) to pass out the passwords to other machines on the network. Dru SymNet
From: Rocky Rockwell <rockwell@connect.net> Newsgroups: comp.sys.next.bugs Subject: Re: NetInfo...something new broken? Date: 12 Sep 1995 16:47:37 GMT Organization: Connection Technologies Message-ID: <434dj9$gm5@dallas1.connect.net> References: <42se57$la@optical.fiber.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: cpayne@optical cpayne@optical (Carl Payne) wrote: >3.3 black hardware > >Using the save netinfo option shuts NetInfo down, never to return >without a reboot. > Is this system a master NetInfo server? If it is, check the trusted network option in netinfomanger, as well as the serves property. I hope that helps helps! > >Also, why is a non-NetInfo-d computer insisting on waiting for another >non NI'd computer to boot before it'll finish booting? They are >NOT (underline that a million times) NetInfo connected. > What kind of computers and how are they connected! >-- >Carl Payne >cpayne@fiber.net <--NeXTMAIL, MIME, ASCII all OK -- ///////////////////////////////////////////////////////////// Martin "Rocky" Rockwell Ph: 214-393-4756 Wk: 214-518-6849 (till 8-30-95) Email: rockwell@connect.net (No NeXTMail yet!)
From: Rocky Rockwell <rockwell@connect.net> Newsgroups: comp.sys.next.bugs Subject: Re: NetInfo...something new broken? Date: 12 Sep 1995 16:48:16 GMT Organization: Connection Technologies Message-ID: <434dkg$gm5@dallas1.connect.net> References: <42se57$la@optical.fiber.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: cpayne@fiber.net cpayne@optical (Carl Payne) wrote: >3.3 black hardware > >Using the save netinfo option shuts NetInfo down, never to return >without a reboot. > Is this system a master NetInfo server? If it is, check the trusted network option in netinfomanger, as well as the serves property. I hope that helps helps! > >Also, why is a non-NetInfo-d computer insisting on waiting for another >non NI'd computer to boot before it'll finish booting? They are >NOT (underline that a million times) NetInfo connected. > What kind of computers and how are they connected! >-- >Carl Payne >cpayne@fiber.net <--NeXTMAIL, MIME, ASCII all OK -- ///////////////////////////////////////////////////////////// Martin "Rocky" Rockwell Ph: 214-393-4756 Wk: 214-518-6849 (till 8-30-95) Email: rockwell@connect.net (No NeXTMail yet!)
From: Dominik Westner <dominik@gowest.ppp.informatik.uni-muenchen.de> Newsgroups: comp.sys.next.bugs Subject: [intel] Fatal system crash during compile Date: 13 Sep 1995 00:13:20 GMT Organization: TeDoc, Munich, Germany Message-ID: <4357n0$9c@cube.ppp.informatik.uni-muenchen.de> Hi, today I encountered a fatal system crash while compiling mpeg.2.0 sources. The system hung and I had to push the reset button, reboot fsck failed, a manual fsck succeeded. I could not believe that this can be done by just compiling, so I tried again. Same result, but even the manual fsck in single user mode failed (just hung). I had to connect the disk to another system to do the fsck, which finally succeeded then. BUT, I discovered that in my root partition there are missing about 200MB and in my home partition it's about 450MB which are missing. So, is this a known problem? I'm running 3.3 User & Developer (+NT, +EOF1.1, no Developer.3.3-Patch applied). Especially is there a solution to reget the missing file space, besides a reinitialisation? Has anybody discovered the same? If anybody wants to try this :-) , the source of mpeg2.0 can be found at: <ftp://peanuts.leo.org/pub/comp/platforms/next/Graphics/movie/mpeg.2.0.s.ta r.gz> (and an additional patch in the same directory) so long -- Dominik --- Dominik Westner "Everything is possible in an infinite universe" _____________________________________Willi, the painter Munich, Germany (NOT :-)
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Bug in /bin/login Date: 12 Sep 1995 23:52:39 GMT Organization: Technical University of Berlin, Germany Message-ID: <4356g7$764@news.cs.tu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi, there's a bug in /bin/login which prevents telnet logins on some machines but not on others. The reason why it works on others is obscure if you look at the code. Here comes a short description: Just before /bin/login execlp's the users shell, it sets the process group and changes the process group of the tty. Unfortunately, the process group to which it sets and to which it makes the ioctl on the tty is the return value of a previously issues getpid-call (yes, getpid). This is totally wrong but seems to work on most occasions - we all can login (most of the time). I can even do rlogin on the machine which refused to let me in via telnet. Has anyone else experienced this? Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: Kernel thread bug Date: 13 Sep 1995 00:03:15 GMT Organization: Technical University of Berlin, Germany Message-ID: <435743$76k@news.cs.tu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi, while writing a loadable kernel server, I came across this one. I don't know if this is really a bug, but I suppose... When forking a thread in a loadable kernel server using kernel_thread the newly created thread does not have a default reply port. If the new thread subsequently issues a mach call, the reply port gets created by a call to mig_get_reply_port. This is correct as the new thread needs its own communication backchannel as it might be that serveral threads issue interleaved msg_rpc's. If the thread terminates itself later, this newly allocated port stays allocated. If you do this about 22000 times, all ports are in use and the kernel panics. I suspect the problem lies in the code which is responsible for terminating a thread. It should release the reply port but doesn't. I tried to get rid of the reply port in the thread itself, but as port_deallocate is a mach rpc call, it needs the reply port until it returns - so a call to port_deallocate results in a kernel panic as well. What is the right way around this obstacle (besides forking a thread only for deallocating all unneeded reply ports on behalf of the soon to be terminated threads)? Anyone dare to comment? Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: Re: Kernel thread bug Followup-To: comp.sys.next.bugs,comp.sys.next.programmer Date: 13 Sep 1995 00:07:22 GMT Organization: Technical University of Berlin, Germany Message-ID: <4357bq$79s@news.cs.tu-berlin.de> References: <435743$76k@news.cs.tu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Axel Habermann (kiwi@cs.tu-berlin.de) wrote: : while writing a loadable kernel server, I came across this : one. I don't know if this is really a bug, but I suppose... [...] Sorry for following up on my own post, but I forgot to mention another curious fact: In <mach/mach_user_internal.h> there's the following define #define thread_terminate thread_terminate_EXTERNAL thread_terminate_EXTERNAL does not exist in the kernel, but the kernel server linker does not complain about this symbol not being found on load of the kernel server and the kernel panics when I try to do the call. If I undef thread_terminate before I use it, it works as advertised. Hmmm.... Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 11 Sep 1995 04:15:15 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <430d4j$ecf@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
Newsgroups: comp.sys.next.bugs From: fred@improve.fdn.org (Frederic STARK) Subject: Re: Bug in /bin/login Message-ID: <1995Sep13.121227.4712@improve.fdn.org> Sender: news@improve.fdn.org Organization: improve SA - La Defense, Paris, France. References: <4356g7$764@news.cs.tu-berlin.de> Date: Wed, 13 Sep 1995 12:12:27 GMT In article <4356g7$764@news.cs.tu-berlin.de> kiwi@cs.tu-berlin.de (Axel Habermann) writes: > Hi, > > there's a bug in /bin/login which prevents telnet logins > on some machines but not on others. The reason why it works > on others is obscure if you look at the code. Here comes a > short description: > > Just before /bin/login execlp's the users shell, it sets > the process group and changes the process group of the tty. > > Unfortunately, the process group to which it sets and to > which it makes the ioctl on the tty is the return value > of a previously issues getpid-call (yes, getpid). > > This is totally wrong but seems to work on most occasions - > we all can login (most of the time). Why do you call this wrong ? I beleive (need to check) that it is the behaviour described in "The Design and Implementation of the 4.3BSD UNIX Operating System. Samuel J. Leffler, Marshall Kirk McKusick, Michael J. Karels, and John S. Quarterman. Addison-Wesley, 1989." -- fred ----------------------------------------------------------------- frederic stark -- fred@improve.fdn.org -----------------------------------------------------------------
From: chuck@its.com Newsgroups: comp.sys.next.bugs Subject: Re: NetInfo...something new broken? Date: 14 Sep 1995 02:51:18 GMT Organization: Information Technology Solutions, Inc. Message-ID: <4385b6$902@news.its.com> References: <42se57$la@optical.fiber.net> cpayne@optical (Carl Payne) wrote: > Also, why is a non-NetInfo-d computer insisting on waiting for another > non NI'd computer to boot before it'll finish booting? They are > NOT (underline that a million times) NetInfo connected. Because you can't ever turn NetInfo *off*, even on a standalone system or one that only uses standard Unix flatfile or YP/NIS networking. I think that NetInfo is really good at handling things such as distributing and maintaining user accounts, alias information, and similiar things. However, the resolver library routines (for DNS) never should have relied on getting info from a NetInfo server. NeXT did this as a shortcut so that people could get local-area nets up quickly and easily, but the result is that a good percentage of NEXTSTEP sites do not have their DNS databases configured correctly, if at all. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: Gerald Wildgruber Newsgroups: comp.sys.next.bugs Subject: multiple volume spanning with gnutar -M Date: 14 Sep 1995 06:49:58 GMT Organization: Leibniz-Rechenzentrum, Muenchen (Germany) Distribution: world Message-ID: <438jam$jmp@sparcserver.lrz-muenchen.de> Hello, Every time i'm trying to use the multiple volume spanning option -M of GNU tar (version 1.11.2) on my NeXTStep 3.3 for Intel Processors machine, either in writing such volumes or in reading volumes written on another computer, i meet the following problem: after having given a command such as: gnutar cMvf /dev/rfd0b "files" to write the data stream on a diskette as a raw device, the programm starts working properly, filling the diskette, but with continuous messages in /usr/adm/messages: Sep 13 20:01:38 ue801be mach: fd0: Sector 760 cmd = Write; DMA Over/underrun: RETRYING (etc...) After about fifty times repeating this warning, the programm exits with: gnutar: can't write to /dev/rfd0b : Invalid argument Same thing when trying to read an archive from a diskette via the raw device; in /usr/adm/messages: Sep 13 20:26:52 ue801be mach: fd0: Sector 1180 cmd = Read; DMA Over/underrun: RETRYING the programm starts reading back onto the hard drive, exiting then with the mentioned message: gnutar: read error on /dev/rfd0b : Invalid argument gnutar: Too many errors, quitting. Does someone know what the DMA Over/underrun messages mean in relation to the gnutar utility? I'm not sure if this is a NeXT specific problem or if i made mistakes with the gnutar options. Any help would be appreciated. Gerald.
From: mow@navigator.de (Markus Wenzel) Newsgroups: comp.sys.next.bugs Subject: Re: Quantrix 2.0 major bug Date: 11 Sep 1995 08:59:48 +0200 Organization: Navigator Distribution: world Message-ID: <430mp4$20l@marsu.navigator.de> References: <SAMURAI.95Sep8140242@maggie.cs.mcgill.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Darcy BROCKBANK (samurai@maggie.cs.mcgill.ca) wrote: > > Summary: error in linking tables prevents some operations. > > (1) Create a new model. > (2) Create a new table. > (3) In the first table, for column B1, enter the following values: > > B1 > A1 one > A2 two > A3 three > > (4) In the second table, enter the following formula: > > B1 = Table 1::B1 > > (5) This is what you get. > > B1 > A1 one > A2 one > A3 one > A4 one > > (6) What you should have gotten was this: > > B1 > A1 one > A2 two > A3 three > A4 <undefined> > > Again, not only is this completely wrong (it should be "one" "two" > "three", as in the first table); another error is shown in that there's > a mismatch in the number of rows in the B1 column. Thus, the B1:A4 > element should be undefined, and the user notified. > > Lighthouse has been notified of the bug and states that it will be fixed > in the first maintentance release. Gee! I already changed all my spreadsheets from multi to single cell references :-((( -- who? // Dipl.-Phys. Markus Wenzel work? // Navigator - IT Consulting & System administration mail? // mow@navigator.de
From: kiwi@sojus (kiwi@buran.fb10.tu-berlin.de (Axel Habermann)) Newsgroups: comp.sys.next.bugs Subject: Re: Bug in /bin/login Date: 14 Sep 1995 08:37:29 GMT Organization: Technical University of Berlin, Germany Message-ID: <438pk9$uv@news.cs.tu-berlin.de> References: <4356g7$764@news.cs.tu-berlin.de> <1995Sep13.121227.4712@improve.fdn.org> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Frederic STARK (fred@improve.fdn.org) wrote: : In article <4356g7$764@news.cs.tu-berlin.de> kiwi@cs.tu-berlin.de (Axel : Habermann) writes: : > Hi, : > : > there's a bug in /bin/login which prevents telnet logins : > on some machines but not on others. The reason why it works : > on others is obscure if you look at the code. Here comes a : > short description: : > : > Just before /bin/login execlp's the users shell, it sets : > the process group and changes the process group of the tty. : > : > Unfortunately, the process group to which it sets and to : > which it makes the ioctl on the tty is the return value : > of a previously issues getpid-call (yes, getpid). : > : > This is totally wrong but seems to work on most occasions - : > we all can login (most of the time). : Why do you call this wrong ? : I beleive (need to check) that it is the behaviour described in : "The Design and Implementation of the 4.3BSD UNIX Operating System. : Samuel J. Leffler, Marshall Kirk McKusick, Michael J. Karels, and John S. : Quarterman. Addison-Wesley, 1989." Ok. I just checked it - and you're right. The problem remains. The code described (which is correct according to Leffler et al) causes /bin/login to be terminated on execution of ioctl(0, TIOCSPGRP, &pgrp); which follows. This happens on some machines, everytime /bin/login is started by telnetd. It does not happen on other machines (which appear to be configured equally) even when /bin/login is started by telnetd. This leads to the conclusion, that the tty driver code must be buggy. Axel -- Axel Habermann \\|// "Wenn Du kiwi@buran.fb10.tu-berlin.de )o o( nicht weisst Fon: +49 30 45478986 Fax:4542296 \ | / was Du tust, D2: +49 172 3900348 \~/ mach's mit Eleganz!"
From: suckow@ bln.sel.alcatel.de (Ralf Suckow) Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs Subject: Selected Text in NXBrowser overwrites scroll bar Date: 14 Sep 1995 17:03:33 GMT Organization: Alcatel/Bell Distribution: world Message-ID: <439n95$ki6@btmpjg.god.bel.alcatel.be> Hi netters, here's a problem with NXBrowser that's new for me. Depending on the displayed string, some cells in a browser overwrite the scroll bar to the left of them, when selected. When they are unselected, the scroll bar is not restored. I tried [matrix sizeToCells] when loading, [browser sizeToFit] etc. no effect. Also I studied more than a hundred old postings to c.s.n.p concerning NXBrowser - nothing. Any ideas? Thank you very much. Ralf PS. I had the same problem with a text in a scroll view some weeks ago, but I could solve it in IB by resizing the content view. -- Ralf Suckow |------------------------ suckow@bln.sel.alcatel.de | All opinions are mine.
From: davisre@best.com (Robert Davis) Newsgroups: comp.sys.next.bugs Subject: Similar Improv Bug (was: Re: Quantrix 2.0 major bug) Date: 14 Sep 1995 12:58:46 -0700 Organization: independent Message-ID: <43a1hm$mbk@shell1.best.com> References: <SAMURAI.95Sep8140242@maggie.cs.mcgill.ca> <430mp4$20l@marsu.navigator.de> >Darcy BROCKBANK (samurai@maggie.cs.mcgill.ca) wrote: >> >> Again, not only is this completely wrong (it should be "one" "two" >> "three", as in the first table); another error is shown in that there's >> a mismatch in the number of rows in the B1 column. Thus, the B1:A4 >> element should be undefined, and the user notified. >> Strangely enough, a very similar bug exists in Improv (1.0, v108x0): 1. New doc. 2. Make the table look like this: B1 B2 B3 A1 one two three A2 A3 3. Create a second table that looks like this: B1 B2 B3 A1 A2 A3 4. In the second table, enter this formula: B1 = Worksheet1::A1 Notice that Improv fills every cell with "one" and incorrectly warns of a mismatch. It's such a serendipitous occurrence to find virtually the same bug in both of NeXTSTEP's multidimensional spreadsheets that I'm convinced today is the day to sell Netscape stock. :-) Run run, Rob
Newsgroups: comp.sys.next.bugs From: alex@genoa.com (Alex Blakemore) Subject: Re: Similar Improv Bug (was: Re: Quantrix 2.0 major bug) Message-ID: <DEy7uJ.Dy@genoa.com> Sender: alex@genoa.com (Alex Blakemore) Organization: Genoa Software Systems References: <43a1hm$mbk@shell1.best.com> Date: Fri, 15 Sep 1995 13:23:07 GMT Robert Davis writes > It's such a serendipitous occurrence to find virtually the same bug in > both of NeXTSTEP's multidimensional spreadsheets that I'm convinced > today is the day to sell Netscape stock. :-) gee, its almost as if they both use the same third party computation engine. -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: wiley@gibraltar (Wiley Hodges) Newsgroups: comp.sys.next.bugs Subject: Re: Similar Improv Bug (was: Re: Quantrix 2.0 major bug) Date: 15 Sep 1995 17:16:43 GMT Organization: Lighthouse Design, Ltd. Message-ID: <43ccdr$adk@lighthouse.lighthouse.com> References: <43a1hm$mbk@shell1.best.com> <DEy7uJ.Dy@genoa.com> In article <DEy7uJ.Dy@genoa.com>, you wrote: > Robert Davis writes > > It's such a serendipitous occurrence to find virtually the same bug in > > both of NeXTSTEP's multidimensional spreadsheets that I'm convinced > > today is the day to sell Netscape stock. :-) > gee, its almost as if they both use the same third party computation engine. > -- > Alex Blakemore > alex@genoa.com NeXT, MIME and ASCII mail accepted Actually, they were both written independently. The Improv engine was written in C++, and originally ran on various minicomputers as a research project for years before Improv was "productized" by Lotus. The Quantrix computation engine was written by Pete Murray, Dan Fish, and Alan Eyzaguirre, who were all Lighthouse engineers at the time. The engine is written in Objective-C, with some parts written in vanilla procedural C for the sake of performance. We've never had any assistance, or indeed, any relationship with Lotus on this product. I suspect that the coincidence of the bugs is more due to the common problems which faced independent engineering teams. You should know, however, that all past/present/future members of the Quantrix engineering team are required to be freemasons, owing to the quantity of fnords which are liberally used in the comments in the code. ;-) Best, --Wiley
From: rjrjr@tuna (Ray Ryan) Newsgroups: comp.sys.next.bugs Subject: cmsg cancel <43d3jf$dt5@lighthouse.lighthouse.com> Control: cancel <43d3jf$dt5@lighthouse.lighthouse.com> Date: 16 Sep 1995 00:13:06 GMT Organization: Lighthouse Design, Ltd. Message-ID: <43d4qi$dt5@lighthouse.lighthouse.com> cancel
Newsgroups: comp.sys.next.bugs From: alex@genoa.com (Alex Blakemore) Subject: Re: Similar Improv Bug (was: Re: Quantrix 2.0 major bug) Message-ID: <DEzD2o.107@genoa.com> Sender: alex@genoa.com (Alex Blakemore) Organization: Genoa Software Systems References: <43ccdr$adk@lighthouse.lighthouse.com> Date: Sat, 16 Sep 1995 04:13:36 GMT I wrote: > its almost as if they both use the same third party computation engine. Wiley Hodges replied > Actually, they were both written independently. Sorry for the erroneous implication. Someone told me that a third party developed the engine, which was later licensed by Lotus for Improv and Quantrix. Just goes to show you shouldn't believe everything you hear or that some bozo posts on the net :-) -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) Newsgroups: comp.sys.next.bugs Subject: Not in my Improv (Re: Similar Improv Bug) Date: 17 Sep 1995 02:10:47 GMT Organization: School of Computer Science, McGill Univ. Message-ID: <SAMURAI.95Sep16221047@maggie.cs.mcgill.ca> References: <43a1hm$mbk@shell1.best.com> <DEy7uJ.Dy@genoa.com> <43ccdr$adk@lighthouse.lighthouse.com> In-reply-to: wiley@gibraltar's message of 15 Sep 1995 17:16:43 GMT <wiley@gibraltar> writes: >In article <DEy7uJ.Dy@genoa.com>, you wrote: >>Robert Davis writes >>> It's such a serendipitous occurrence to find virtually the same bug in >>> both of NeXTSTEP's multidimensional spreadsheets that I'm convinced >>> today is the day to sell Netscape stock. :-) >>gee, its almost as if they both use the same third party computation engine. >Actually, they were both written independently. The Improv engine was written in C++, >and originally ran on various minicomputers as a research project for years before Improv >was "productized" by Lotus. >The Quantrix computation engine was written by Pete Murray, Dan Fish, and Alan >Eyzaguirre, who were all Lighthouse engineers at the time. The engine is written in >Objective-C, with some parts written in vanilla procedural C for the sake of performance. The presense of the bug in Improv is surprising to me. I found the bug in Quantrix because I tried to open one of my Improv spreadsheets which used inter-worksheet formulas. The Quantrix one failed miserably, while the Improv one worked perfectly. I distilled the problem down into the simple case which I posted here, for the sake of making a bug report. I'm surprised that my complicated reference works in Improv and the simple case is supposed to fail... perhaps I've just got a later release of Improv. I know there was one enhanced version (well I think so). You can probably figure out which one you have by control-alt-command clicking on the empty area in the info panel. If you see a goofy face, you've got the older version. Mine doesn't do it, so I assume it's the newer version. Sure enough, when I test my simple case on my version of Improv, it works as I described it should. I get one two three Error in the B1 column of the target spreadsheet. v1.0 108x0. - db -- "Investing in tech stocks is a high risk." -- Bill Gates, Microsoft Chairman, after incurring a two day paper loss of 2 billion dollars on Microsoft stock.
From: davisre@best.com (Robert Davis) Newsgroups: comp.sys.next.bugs Subject: Re: Not in my Improv (Re: Similar Improv Bug) Date: 16 Sep 1995 21:08:12 -0700 Organization: independent Message-ID: <43g6vc$5gn@shell1.best.com> References: <43a1hm$mbk@shell1.best.com> <DEy7uJ.Dy@genoa.com> <43ccdr$adk@lighthouse.lighthouse.com> <SAMURAI.95Sep16221047@maggie.cs.mcgill.ca> Darcy BROCKBANK <samurai@maggie.cs.mcgill.ca> wrote: > >Sure enough, when I test my simple case on my version of Improv, it >works as I described it should. > Yes -- you'll have to try *my* slightly different recipe to see the Improv bug. Hence the word "similar." I said: > >Strangely enough, a very similar bug exists in Improv (1.0, v108x0): > >1. New doc. >2. Make the table look like this: > > B1 B2 B3 >A1 one two three >A2 >A3 > >3. Create a second table that looks like this: > > B1 B2 B3 >A1 >A2 >A3 > >4. In the second table, enter this formula: > >B1 = Worksheet1::A1 > > >Notice that Improv fills every cell with "one" and incorrectly warns >of a mismatch. > By the way, I've heard (can't verify it) that the goofy face in the info panel doesn't appear in some versions of NeXTSTEP (post 3.0?) -- hmm, not sure. Rob -- Robert Davis (davis@coaster.com) "Look up, Hannah."
From: trout@denizen.com (Paul Kim) Newsgroups: comp.sys.next.bugs Subject: Re: Not in my Improv (Re: Similar Improv Bug) Date: 17 Sep 1995 18:47:08 GMT Organization: a2i network Message-ID: <43hqfc$rqf@hustle.rahul.net> References: <SAMURAI.95Sep16221047@maggie.cs.mcgill.ca> In article <SAMURAI.95Sep16221047@maggie.cs.mcgill.ca> samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) writes: > perhaps I've just got a later release of Improv. I know there was one > enhanced version (well I think so). > > You can probably figure out which one you have by control-alt-command > clicking on the empty area in the info panel. If you see a goofy face, > you've got the older version. Mine doesn't do it, so I assume it's the > newer version. > I think the existence of a "newer" version of Improv is more of a rumour. If one did exist, it was only given to a few sites. I could be wrong though. As for the goofy face not appearing, it seems to be more an indication of what OS revision you have and not the version of Improv. I was able to call up the face under NS 2.X but have been unable to do it since upgrading to NS 3.X. For those who want to see the face under 3.X, you'll have to "seghoark" the image out of the binary. It's called something like "Glenn.tiff". I don't know who Glenn is but I'm guessing that it was the programmer's boss. I posted a .dir.tiff and .opendir.tiff to the next-icon list some time ago and should be in the next-icon archives. BTW, I've heard that Improv has an easter egg involving the toilet paper dispenser where it generates the names of the programmer's ex-girlfriends. Anyone know how to conjure this up? I did a 'strings' on the binary but didn't see anything suspect (albeit, I didn't scan it very meticulously). paul trout@denizen.com
From: samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) Newsgroups: comp.sys.next.bugs Subject: Re: Not in my Improv (Re: Similar Improv Bug) Date: 17 Sep 1995 19:41:51 GMT Organization: School of Computer Science, McGill Univ. Message-ID: <SAMURAI.95Sep17154151@maggie.cs.mcgill.ca> References: <43a1hm$mbk@shell1.best.com> <DEy7uJ.Dy@genoa.com> <43ccdr$adk@lighthouse.lighthouse.com> <SAMURAI.95Sep16221047@maggie.cs.mcgill.ca> <43g6vc$5gn@shell1.best.com> In-reply-to: davisre@best.com's message of 16 Sep 1995 21:08:12 -0700 <davisre@best.com> writes: >Darcy BROCKBANK <samurai@maggie.cs.mcgill.ca> wrote: >> >>Sure enough, when I test my simple case on my version of Improv, it >>works as I described it should. >> >Yes -- you'll have to try *my* slightly different recipe to see the >Improv bug. Hence the word "similar." >I said: >> >>Strangely enough, a very similar bug exists in Improv (1.0, v108x0): >> >>1. New doc. >>2. Make the table look like this: >> >>B1 B2 B3 >>A1 one two three >>A2 >>A3 >> >>3. Create a second table that looks like this: >> >>B1 B2 B3 >>A1 >>A2 >>A3 >> >>4. In the second table, enter this formula: >> >>B1 = Worksheet1::A1 >> >> >>Notice that Improv fills every cell with "one" and incorrectly warns >>of a mismatch. >> Ahhh, I didn't read it all the way. OK, so I did this, and saw what you did. Now, if you want the REALLY odd stuff, go and change the "B" category name to "C" then the "A" category name to "B". The spreadsheet re-calcs and the results are correct!!! - db -- "Investing in tech stocks is a high risk." -- Bill Gates, Microsoft Chairman, after incurring a two day paper loss of 2 billion dollars on Microsoft stock.
From: kparks@its.brooklyn.cuny.edu (Kevin Parks) Newsgroups: comp.sys.next.bugs Subject: loginwindow MoveWhenIdle YES- thingy Date: 19 Sep 1995 00:45:31 -0400 Organization: Brooklyn College Message-ID: <43lhtc$3o9@atrium16.its.brooklyn.cuny.edu> Hi all, here's a dumb question: I running NeXTStep 3.2 on a NeXTStation and i would like my login panel to bounce around when no one is logged in. I sometimes run really big jobs that go for 30 or 40 hours and i want to be able to logout and not have the login panel burn in. I have looked for hours and hours for info on how to do this in the on line documents, But to no avail. I know tha under 3.0 it was something like: % dwrite loginwindow MoveWhenIdle YES % dwrite loginwindow MovementRate 0.05 % dwrite loginwindow MovementScale 3 % dwrite loginwindow MovementTimeout 60.0 But that didn't work for me. Anyone know how to do this under3.2? Many thanks, kevin
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 18 Sep 1995 04:15:10 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <43iroe$e4p@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
Newsgroups: comp.sys.next.bugs From: magnus@darwin.uchicago.edu (Magnus Nordborg) Subject: Mail.app (NS 3.3) eats .mailalias Message-ID: <DF5ns5.LF1@midway.uchicago.edu> Sender: news@midway.uchicago.edu (News Administrator) Organization: University of Chicago -- Networking Services Date: Tue, 19 Sep 1995 13:50:29 GMT Mail.app has a bad habit of arbitrarily truncating long entries under "Private Groups" so that data is lost. This bug existed in 3.2 as well. --- Magnus Nordborg Department of Ecology & Evolution The University of Chicago 1101 E. 57th St. Chicago, IL 60637-1573 USA magnus@darwin.uchicago.edu (NeXT Mail, MIME) +1.312.702-1093 phone (lab) +1.312.667-5331 phone (home) +1.312.702-9740 fax
From: richard@runner.uucp@usc.edu (Richard Ruth) Newsgroups: comp.sys.next.bugs Subject: Re: loginwindow MoveWhenIdle YES- thingy Date: 19 Sep 1995 19:38:15 -0700 Organization: runner Message-ID: <43nuqn$5ii@runner.uucp> References: <43lhtc$3o9@atrium16.its.brooklyn.cuny.edu> Kevin Parks (kparks@its.brooklyn.cuny.edu) wrote: : Hi all, : here's a dumb question: : I running NeXTStep 3.2 on a NeXTStation and i would like my login panel to : bounce around when no one is logged in. I sometimes run really big jobs that : go for 30 or 40 hours and i want to be able to logout and not have the login : panel burn in. ... : kevin Try dwriting the following as root: loginwindow ScreenSaverEnabled YES loginwindow TimeToDim 60 -- Richard richard%runner.uucp@usc.edu (100K bytes max -- ok to send NeXTMail)
From: franke@hamlet.dbag.ulm.DaimlerBenz.COM (Juergen Franke) Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs,comp.sys.next.software Subject: debugging problems on hp Next Stations Date: 20 Sep 1995 07:58:21 GMT Organization: debis Network Services GmbH Distribution: world Message-ID: <43ohit$gps@news.sns-felb.debis.de> Keywords: BuildInDebuggerInterface I have problems with the Build in Debugger Interface on my hp Next Station. The hierarchie of methods is reverse to the hierarchie on the intel machines. If I want to have launched the code of the 5th method in the stack I have to double-click the (N-5)th stack line. For example [ImageOps process:] [ImageOps test:] [Application send Event:] [Application run] main If I want to launch the cde of [ImageOps process] I have to click on the line of main. Please send the answer to J rgen Franke Text Understanding Department F3M/T Daimler-Benz AG, Research Center Ulm P.O. Box 2360 89013 Ulm Germany Telephone Germany 731 505 2355 Fax Germany 731 505 4113 e-mail: franke@dbag.ulm.DaimlerBenz.COM I will post the summary. -- NewsGrazer, a NeXTstep(tm) news reader, posting -- M>UQR=&8P7&%N<VE[7&9O;G1T8FQ<9C!<9FUO9&5R;B!#;W5R:65R.WT*7&UA M<F=L,3(P"EQM87)G<C$R,`I<<&%R9%QT>#$Q-3)<='@R,S`T7'1X,S0U-EQT M>#0V,#A<='@U-S8P7'1X-CDQ,EQT>#@P-C1<='@Y,C$V7'1X,3`S-CA<='@Q M,34R,%QF,%QB,%QI,%QU;&YO;F5<9G,R-%QF8S!<8V8P($D@:&%V92!P<F]B M;&5M<R!W:71H('1H92!"=6EL9"!I;B!$96)U9V=E<B!);G1E<F9A8V4@;VX@ M;7D@:'`@3F5X="!3=&%T:6]N+B!4:&4@:&EE<F%R8VAI92!O9B!M971H;V1S M(&ES(')E=F5R<V4@=&\@=&AE(&AI97)A<F-H:64@;VX@=&AE(&EN=&5L(&UA M8VAI;F5S+B!)9B!)('=A;G0@=&\@:&%V92!L875N8VAE9"!T:&4@8V]D92!O M9B!T:&4@-71H(&UE=&AO9"!I;B!T:&4@<W1A8VL@22!H879E('1O(&1O=6)L M92UC;&EC:R!T:&4@*$XM-2ET:"!S=&%C:R!L:6YE+B!<"EP*1F]R(&5X86UP M;&4@7`I;26UA9V5/<',@<')O8V5S<SI=7`I;26UA9V5/<',@=&5S=#I=7`I; M07!P;&EC871I;VX@<V5N9"!%=F5N=#I=7`I;07!P;&EC871I;VX@<G5N75P* M;6%I;EP*7`I)9B!)('=A;G0@=&\@;&%U;F-H('1H92!C9&4@;V8@6TEM86=E M3W!S('!R;V-E<W-=($D@:&%V92!T;R!C;&EC:R!O;B!T:&4@;&EN92!O9B!M M86EN+EP*7`I0;&5A<V4@<V5N9"!T:&4@86YS=V5R('1O(%P*"EQP87)D7'1X M-S`P7'1X,30R,%QT>#(Q,C!<='@R.#0P7'1X,S4T,%QT>#0R-C!<='@T.3@P M7'1X-38X,%QT>#8T,#!<='@W,3`P7&9C,%QC9C`@2O9R9V5N($9R86YK95P* M5&5X="!5;F1E<G-T86YD:6YG($1E<&%R=&UE;G0@1C--+U1<"D1A:6UL97(M M0F5N>B!!1RP@4F5S96%R8V@@0V5N=&5R(%5L;5P*4"Y/+B!";W@@,C,V,%P* M.#DP,3,@56QM7`I'97)M86YY7`I<"E1E;&5P:&]N92!'97)M86YY(#<S,2`U M,#4@,C,U-5P*1F%X("`@("`@($=E<FUA;GD@-S,Q(#4P-2`T,3$S7`IE+6UA M:6PZ(&9R86YK94!D8F%G+G5L;2Y$86EM;&5R0F5N>BY#3TU<"EP*22!W:6QL 7('!O<W0@=&AE('-U;6UA<GDN7`H*?0IM `
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: little WM bug Date: Wed, 20 Sep 1995 19:44:06 -0400 Organization: Princeton University Message-ID: <Pine.NXT.3.91.950920194133.5727C-100000@capitalist.princeton.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Here's a funny little item. I can't assign the key "D" (uppercase) to any menu item in the Workspace Manager. I've assigned it via dwrite WM NXCommandKeys, and it shows up on the menu panel, but <command>-<shift>-d gives me nothing but a system beep. tested a few times on 3.2 m68k TjL -- Timothy J. Luoma 476tjl@ptsmail.ptsem.edu (ASCII Only!) NS 3.2 m68k
From: Politics@usa.com Newsgroups: comp.sys.next.bugs Subject: Proposed State & Federal Regulations for the INTERNET! Date: 23 Sep 1995 06:57:02 GMT Organization: Wisconsin Internet Services Message-ID: <440b3u$6id@newsie.wis.com> Originator: Politics@usa.com Originator: glasrud@squeaky.free.org My name is Scott Glasrud, and I am running for the New Mexico State Senate during the 1996 elections. One of the reasons I have chosen to run is to combat the proposed state and federal regulations of the Internet. As you know, the Internet was never designed to be regulated! It was designed to allow communications in the event of anuclear war or a major catastrophe. I OPPOSE REGULATION, and if elected will fight to preserve your constitutional rights. HOWEVER, I NEED YOUR HELP! I am asking each person who reeives this message to send $5.00 to the Scott Glasrud Campaign Committee. If we pull together, we CAN protect our first amendment rights! HELP ME show the politicians the POWER behind this important NETWORK. Please send contributions to: The Scott Glasrud Campaign Committee 11024 Montgomery Blvd. NE, Suite 179 Albuquerque, New Mexico 87111 Thank you!
Newsgroups: comp.sys.next.bugs From: fozztexx@nvc.cc.ca.us (Chris Osborn) Subject: POSIX getcwd is broken in NS 3.3 Message-ID: <DFDpzy.M9L@nvc.cc.ca.us> Sender: news@nvc.cc.ca.us Organization: Napa Valley College Date: Sat, 23 Sep 1995 22:19:10 GMT The POSIX function getcwd appears to be broken under 3.3. It always returns NULL with an errno of 2. -- Chris Osborn, Network Administrator Voice: 707 253 3130 Napa Valley College Fax: 707 253 3063 2277 Napa-Vallejo Hwy., Napa, CA, 94558 <fozztexx@nvc.cc.ca.us> MIME ok, NeXTMail tolerated
From: fischer-michael@cs.yale.edu (Michael Fischer) Newsgroups: comp.sys.next.bugs Subject: Re: POSIX getcwd is broken in NS 3.3 Date: 23 Sep 1995 22:49:35 GMT Organization: Yale University Computer Science Dept., New Haven, CT 06520-8285 Message-ID: <4422u0$k0i@babyblue.cs.yale.edu> References: <DFDpzy.M9L@nvc.cc.ca.us> Chris Osborn (fozztexx@nvc.cc.ca.us) wrote: : The POSIX function getcwd appears to be broken under 3.3. It always : returns NULL with an errno of 2. I can add a little more. It does work when the working directory is on a local disk, but it fails as you describe for directories on NFS-mounted disks. ================================================== | Michael Fischer <fischer-michael@cs.yale.edu> | ==================================================
From: hansen@berkelynetcentral.com Newsgroups: comp.sys.next.bugs Subject: Supra FAX modem won't fax Date: 24 Sep 1995 04:17:07 GMT Organization: Best Internet Communications Message-ID: <442m43$nea@shellx.best.com> Keywords: supra fax Anyone ever get a Supra FAx modem 144LC to work faxing? Have 040cube 3.2
From: sfoy@srtb0511a12.resnet.ubc.ca (Shaun Patrick Foy) Newsgroups: comp.sys.next.bugs Subject: Re: Supra FAX modem won't fax Date: 24 Sep 1995 09:58:45 GMT Organization: PhilosVille Message-ID: <443a4l$ab@nntp.ucs.ubc.ca> References: <442m43$nea@shellx.best.com> Isn't that a class1 fax modem?? On 24 Sep 1995 04:17:07 GMT hansen@berkelynetcentral.com wrote: >Anyone ever get a Supra FAx modem 144LC to work faxing? >Have 040cube 3.2 -- \ o / o __| \ / |__ o \ o / | -/\ ___\o \ o | o / o/___ /\- | / \ | \ /) | ( \ /o\ / ) | (\ / | / \ <A HREF="http://philos.resnet.ubc.ca/~sfoy">sfoy@bcu.ubc.ca</A>
Newsgroups: comp.sys.next.bugs From: mark@bohica.net (Mark Gregory Salyzyn) Subject: Re: little WM bug Message-ID: <DFEyKA.Mxq@bohica.net> Organization: UNIX drivers'R'Us References: <Pine.NXT.3.91.950920194133.5727C-100000@capitalist.princeton.edu> Date: Sun, 24 Sep 1995 14:21:45 GMT "Timothy J. Luoma" <luomat@capitalist.princeton.edu> writes: >Here's a funny little item. I can't assign the key "D" (uppercase) to >any menu item in the Workspace Manager. > >I've assigned it via dwrite WM NXCommandKeys, and it shows up on the menu >panel, but <command>-<shift>-d gives me nothing but a system beep. System NXCommandKeys "Check for Disks,D" Works for me on 3.1, 3.2 and 3.3 on Intel. Ciao -- Mark Salyzyn
Date: 24 Sep 1995 11:13:54 GMT From: Politics@usa.com Newsgroups: comp.sys.next.bugs Message-ID: <cancel.440b3u$6id@newsie.wis.com> Control: cancel <440b3u$6id@newsie.wis.com> Subject: cmsg cancel <440b3u$6id@newsie.wis.com> EMP/ECP (aka SPAM) cancelled by clewis@ferret.ocunix.on.ca. See news.admin.net-abuse.announce, report 19950924.01 for further details
Control: cancel <440b3uid@newsie.wis.com> Newsgroups: comp.sys.next.bugs From: Politics@usa.com Subject: cmsg cancel <440b3uid@newsie.wis.com> Followup-To: poster Sender: Politics@usa.com Organization: Taronga Park BBS Message-ID: <cancel.440b3uid@newsie.wis.com> Date: Sun, 24 Sep 1995 16:26:24 GMT Spam cancelled by cancelbot@taronga.com
From: ccwf@thales.klab.caltech.edu (Charles Fu) Newsgroups: comp.sys.next.bugs Subject: Re: POSIX getcwd is broken in NS 3.3 Date: 25 Sep 1995 01:13:35 GMT Organization: California Institute of Technology, Pasadena Message-ID: <444vnv$3c2@gap.cco.caltech.edu> References: <DFDpzy.M9L@nvc.cc.ca.us> <4422u0$k0i@babyblue.cs.yale.edu> >Chris Osborn (fozztexx@nvc.cc.ca.us) wrote: >: The POSIX function getcwd appears to be broken under 3.3. It always >: returns NULL with an errno of 2. No, it's more fundamental than that. It's the POSIX readdir() function that's broken, as I have posted here previously, several months ago. I believe this is bug is documented in KBNS, although I haven't checked it recently. In article <4422u0$k0i@babyblue.cs.yale.edu>, Michael Fischer <fischer-michael@cs.yale.edu> wrote: >I can add a little more. It does work when the working directory is >on a local disk, but it fails as you describe for directories on >NFS-mounted disks. One manifestation of the bug occurs when you do a readdir() of a directory in which disks have been mounted on directories which did not exist prior to mounting (something you can not do on many UNIXes). The problem is not restricted to NFS mounts. Since many people spend much of their time in subdirectories of such directories (e.g., /Net for folks on networks and / when you automount a floppy or CD) and since many functions like getcwd() call readdir(), the problem is fairly pervasive. There may also be other manifestations of the bug. I don't know why this bug wasn't detected and fixed, or a public patch distributed. The fix is quite simple--get working source for readdir() from a POSIX system (for example from 4.4BSD) and use that instead of NeXT's readdir() function. You can just include this in your sources and link against it to prevent NeXT's broken function from being used. If you want to be more daring, you can remove the broken function from libposix.a (it's a normal library, not shared, so you can manipulate it with standard tools), and replace it with a fixed version. Since the library is not shared, this will not fix POSIX programs which have already been linked with the broken function. It only helps with programs you create. -ccwf
From: ccwf@thales.klab.caltech.edu (Charles Fu) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: CERN httpd patch for NEXTSTEP (was Re: Omni vs Netscape) Date: 25 Sep 1995 02:53:20 GMT Organization: California Institute of Technology, Pasadena Message-ID: <4455j0$894@gap.cco.caltech.edu> References: <41s023$f3s@rzsun02.rrz.uni-hamburg.de> <41ulht$3jn@usc.edu> <MAGNAN.95Aug30114017@maths1.mathcn.umontreal.ca> <442bav$o91@gap.cco.caltech.edu> In article <442bav$o91@gap.cco.caltech.edu>, Charles Fu <ccwf@thales.klab.caltech.edu> wrote: > The most commonly used caching HTTP daemon is CERN's. Note that the > pre-compiled NEXTSTEP version available from CERN contains a date > bug. (Getting the date right is important if you want the httpd cache to > automatically detect when a cache entry is out of date.... I suppose I can > post the necessary patch if anyone is interested. Since several people have mailed me asking for the patch, I am posting it below. I should probably also point out that the precompiled NEXTSTEP httpd at CERN does not incorporate their latest patches (also available from CERN). Again, this patch is necessary for correct operation of conditional GETs and proper refresh of the cache. I reported this bug to CERN when I found it. It should not be necessary to report it again. Unfortunately, the official source distribution has not been updated in quite some time. :-( -ccwf =============================================================================== --- WWW/Daemon/Implementation/HTSUtils.c Sun Sep 25 06:52:57 1994 +++ - Sun Sep 24 19:33:43 1995 @@ -81,7 +81,7 @@ PUBLIC time_t parse_http_time ARGS1(char *, str) { char * s; - struct tm tm; + struct tm tm = {0}; time_t t; #ifdef ISC3 /* Lauren */ struct tm * gorl; /* GMT or localtime */ @@ -166,6 +166,7 @@ /* BSD, have tm_gmtoff */ #if defined(SIGTSTP) && !defined(AIX) && !defined(__sgi) && !defined(_AUX) && ! defined(__svr4__) +#ifndef __NeXT__ { time_t cur_t = time(NULL); struct tm * local = localtime(&cur_t); @@ -173,6 +174,7 @@ CTRACE(stderr,"TimeZone.... %02d hours from GMT\n", (int)local->tm_gmtoff / 3600); } +#endif /* !__NeXT__ */ #else /* SysV or VMS */ { #ifdef VMS
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 25 Sep 1995 04:15:12 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <445acg$gqd@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
Newsgroups: comp.sys.next.bugs From: c008213@next498.pcp.com (Royce Howland) Subject: Re: little WM bug Message-ID: <1995Sep25.150858.23777@pcp.ca> Sender: news@pcp.ca Organization: PanCanadian Petroleum Ltd. References: <Pine.NXT.3.91.950920194133.5727C-100000@capitalist.princeton.edu> <DFEyKA.Mxq@bohica.net> Date: Mon, 25 Sep 95 15:08:58 GMT > "Timothy J. Luoma" <luomat@capitalist.princeton.edu> writes: > >Here's a funny little item. I can't assign the key "D" (uppercase) to > >any menu item in the Workspace Manager. > > > >I've assigned it via dwrite WM NXCommandKeys, and it shows up on the menu > >panel, but <command>-<shift>-d gives me nothing but a system beep. I've experienced this intermittently on NS 3.1, 3.2 and 3.3. I habitually define Cmd-i to bring up Tools->Inspector..., but every now and then it does nothing but produce the system beep. Log out and back in and it works. I used to define Cmd-D to bring up Disk->Check for Disks, but got tired of it not working when I wanted it, and so have quit trying to use that particular short-cut. Royce -- Royce Howland, Object Systems Group howlandr@cadvision.com (home, NeXT mail OK) Howland_Royce@pcp.ca (PanCanadian Petroleum, NeXT & MIME mail OK) I speak, but not necessarily for OSG or PCP.
From: robert@steffi.accessone.com (Robert Nicholson) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: CERN httpd patch for NEXTSTEP (was Re: Omni vs Netscape) Date: 25 Sep 1995 09:31:34 +0100 Organization: me organized? That's a joke! Sender: robert@steffi.accessone.com Message-ID: <e3fdlqzve.fsf@steffi.accessone.com> References: <41s023$f3s@rzsun02.rrz.uni-hamburg.de> <41ulht$3jn@usc.edu> <MAGNAN.95Aug30114017@maths1.mathcn.umontreal.ca> <442bav$o91@gap.cco.caltech.edu> <4455j0$894@gap.cco.caltech.edu> In-reply-to: ccwf@thales.klab.caltech.edu's message of 25 Sep 1995 02:53:20 GMT To: ccwf@thales.klab.caltech.edu (Charles Fu) Does CERN's 3.0 source distribution reflect this patch? -- "Mary ate a little lamb and punk rock isn't dead" (PGP key: send email with Subject: request pgp key)
From: Studio Malaguti Ass. Prof.Le <MC1116@mclink.it> Newsgroups: comp.sys.next.bugs Subject: Problems With Nfsd Date: Mon, 25 Sep 95 18:08:26 CET Organization: MC-link - Italian Dial-Up Online Service Message-ID: <812048906.MC1116@mclink.it> I'm running NS 3.1 on a black cube. I'm not able to mount any exported NFS-volume on my PC/DOS with Lantastic for TCP/IP from Artisoft. Artisoft claims that NFS server needs to be compatible with PCNFSD 2.0 specifications from Sun. Is there anybody knowing that newer versions of NeXTstep are compatible with such specification ? Mirco Malaguti Ingenia s.r.l. mc1116@mclink.it
From: pp000738@interramp.com Newsgroups: comp.sys.next.bugs Subject: Free classified advertising database on the Web Date: Mon, 25 Sep 1995 21:14:14 EDT Organization: PSI Public Usenet Link Message-ID: <211414.pp000738@ipiws-001.interramp.com> Content-Type: TEXT/PLAIN; charset=US-ASCII OVERDRIVE Overdrive has a Worldwide Classified Advertising Database that is available on the Worldwide Web. It works just like a newspaper classified advertising section and is FREE for all to use. You can post an advertisement of your own, update and delete it. Sections include: Announcements and Services Training and Education Employment Financial Pets and Livestock Computer & Communications Merchandise General Merchandise Real Estate for Rent Real Estate for Sale Manufactured Homes Transportation When you enter an Ad, you are prompted for a password. When the record is created, you receive the record id for your ad. Later, you may update or delete that ad using the combination of your password and record id. Web travelers can search for Ads based on Classified Section, price, manufacturer, model, etc. OVERDRIVE can be found at http://www.ipworld.com
From: bmaher@ixc.net(Bruce Maher) Newsgroups: comp.sys.newton.programmer,comp.sys.next.advocacy,comp.sys.next.announce,comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.marketplace Subject: Free Spam Program, Does it Work?? Date: 25 Sep 1995 23:48:48 -0400 Organization: Internet Exchange Carrier Sender: bmaher@ixc.ixc.net Message-ID: <447t70$s4a@ixc.ixc.net> Sorry folks, but I just couldn't resist trying this. Found the script on comp.lang.perl.misc. Does it work? Guess so <gr>. HERE IT IS: ABSOLUTELY FREE Spam as many groups as you want with one command. Can post same message to 14,000 groups in just a few hours. Must have UNIX shell account, the attached script, and create two ascii text files called groups.txt and message.txt. The file groups.txt should contain a list of all groups you want to spam, one per line. If you want to hit everything, you can just copy your .newsrc file, but you'll have to remove all index numbers and end of line punctuation. The message.txt file is the actual message you want to spam. It must contain the subject header on the first line, ie: Subject: Spamming is Fun. You may add in other headers, such as Organization, Paths, Reply to, etc, but none are required. Do NOT put in a newsgroup header. The script does that automatically. There must be a blank line between the last header and the start of the actual message. Then, download those two files plus the following script (call it spamming.fun) to your home directory on the Unix shell account, type "perl spamming.fun", and sit back to watch the fireworks. -----------------------------------Cut Here-------------------- #!perl #Assumes both Perl and Inews are accessible through #your home directory. If not, either place them in your path #or adjust the script. #You must also have a ascii text file called groups.txt, which #contains a list of each group you want to spam. One to a line. #No punctuation at end of line. #Your spam message must be called message.txt, and it should #be in ascii. The first line must be you subject header: # For example: Subject: this is a spam. # You may add in other headers if you wish, but there must be a #blank line before your actual message begins. #run program by typing "perl spam.pl. # # #!perl print "Running...\n"; $newsrc = "$ENV{'HOME'}/testgrps.txt"; open(GROUPS, "$newsrc"); while($group=<GROUPS>) {print "Posting to: $group"; open(NEWS,"|inews -h"); print NEWS "Newsgroups: $group"; open(MESSAGE, "message.txt"); while($line=<MESSAGE>) { print NEWS $line; } close NEWS; if($?==0) { $success++ } else { $fail++ }}$total = $success + $fail;print "Tried to post to $total groups.\n$success OK, $fail failed.\n";
Date: 26 Sep 1995 05:27:09 GMT From: bmaher@ixc.net(Bruce Maher) Sender: bmaher@ixc.ixc.net Message-ID: <cancel.447t70$s4a@ixc.ixc.net> Newsgroups: comp.sys.newton.programmer,comp.sys.next.advocacy,comp.sys.next.announce,comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.marketplace Subject: cmsg cancel <447t70$s4a@ixc.ixc.net> Control: cancel <447t70$s4a@ixc.ixc.net> Spam cancelled by clewis@ferret.ocunix.on.ca
From: alvin@cse.ucsc.edu (Alvin Jee) Newsgroups: comp.sys.next.bugs Subject: NeXT Dev Documentation Error Date: 26 Sep 1995 05:56:08 GMT Organization: University of California, Santa Cruz Message-ID: <4484lo$g9j@darkstar.UCSC.EDU> Hello! Has anybody else noticed that the on-line docs for the Application class and the Action Cell class are virtually the same document? I even checked the Dev CD and they are the same. Sort of defeats the purpose of on-line docs :) I'm running NS3.3-Intel. NS3.2-Intel didn't have this problem. -- Alvin Jee alvin@cse.ucsc.edu NeXTMail gleefully accepted! Using the Internet since 1984 http://www.cse.ucsc.edu/~alvin
From: bmaher@ixc.net(Bruce Maher) Newsgroups: comp.sys.next.bugs Subject: Free Spam Program, Does it Work?? Date: 26 Sep 1995 00:50:03 -0400 Organization: Internet Exchange Carrier Sender: bmaher@ixc.ixc.net Message-ID: <4480pr$hh@ixc.ixc.net> Sorry folks, but I just couldn't resist trying this. Found the script on comp.lang.perl.misc. Does it work? Guess so <gr>. HERE IT IS: ABSOLUTELY FREE Spam as many groups as you want with one command. Can post same message to 14,000 groups in just a few hours. Must have UNIX shell account, the attached script, and create two ascii text files called groups.txt and message.txt. The file groups.txt should contain a list of all groups you want to spam, one per line. If you want to hit everything, you can just copy your .newsrc file, but you'll have to remove all index numbers and end of line punctuation. The message.txt file is the actual message you want to spam. It must contain the subject header on the first line, ie: Subject: Spamming is Fun. You may add in other headers, such as Organization, Paths, Reply to, etc, but none are required. Do NOT put in a newsgroup header. The script does that automatically. There must be a blank line between the last header and the start of the actual message. Then, download those two files plus the following script (call it spamming.fun) to your home directory on the Unix shell account, type "perl spamming.fun", and sit back to watch the fireworks. -----------------------------------Cut Here-------------------- #!perl #Assumes both Perl and Inews are accessible through #your home directory. If not, either place them in your path #or adjust the script. #You must also have a ascii text file called groups.txt, which #contains a list of each group you want to spam. One to a line. #No punctuation at end of line. #Your spam message must be called message.txt, and it should #be in ascii. The first line must be you subject header: # For example: Subject: this is a spam. # You may add in other headers if you wish, but there must be a #blank line before your actual message begins. #run program by typing "perl spam.pl. # # #!perl print "Running...\n"; $newsrc = "$ENV{'HOME'}/testgrps.txt"; open(GROUPS, "$newsrc"); while($group=<GROUPS>) {print "Posting to: $group"; open(NEWS,"|inews -h"); print NEWS "Newsgroups: $group"; open(MESSAGE, "message.txt"); while($line=<MESSAGE>) { print NEWS $line; } close NEWS; if($?==0) { $success++ } else { $fail++ }}$total = $success + $fail;print "Tried to post to $total groups.\n$success OK, $fail failed.\n";
Date: 26 Sep 1995 06:35:41 GMT From: bmaher@ixc.net(Bruce Maher) Sender: bmaher@ixc.ixc.net Message-ID: <cancel.4480pr$hh@ixc.ixc.net> Newsgroups: comp.sys.next.bugs Subject: cmsg cancel <4480pr$hh@ixc.ixc.net> Control: cancel <4480pr$hh@ixc.ixc.net> Spam cancelled by clewis@ferret.ocunix.on.ca
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: Problems With Nfsd Date: 26 Sep 1995 06:48:40 GMT Organization: Technical University of Berlin, Germany Message-ID: <4487o8$h4r@news.cs.tu-berlin.de> References: <812048906.MC1116@mclink.it> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Studio Malaguti Ass. Prof.Le (MC1116@mclink.it) wrote: : I'm running NS 3.1 on a black cube. : I'm not able to mount any exported NFS-volume on my PC/DOS : with Lantastic for TCP/IP from Artisoft. : Artisoft claims that NFS server needs to be compatible : with PCNFSD 2.0 specifications from Sun. : Is there anybody knowing that newer versions of NeXTstep : are compatible with such specification ? : Take a look at /usr/etc/rpc.pcnfsd e.g. just start it up in your /etc/rc.local Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: ccwf@thales.klab.caltech.edu (Charles Fu) Newsgroups: comp.sys.next.bugs Subject: Re: Problems With Nfsd Date: 26 Sep 1995 07:32:25 GMT Organization: California Institute of Technology, Pasadena Message-ID: <448aa9$1c2@gap.cco.caltech.edu> References: <812048906.MC1116@mclink.it> In article <812048906.MC1116@mclink.it>, Studio Malaguti Ass. Prof.Le <MC1116@mclink.it> wrote: >Artisoft claims that NFS server needs to be compatible >with PCNFSD 2.0 specifications from Sun. >Is there anybody knowing that newer versions of NeXTstep >are compatible with such specification ? They are "compatible," but the PC-NFS daemon NeXT ships is a 1.x version. You can FTP the sources for a 2.0 daemon from Sun. I don't recall any notable gotchas with compilation, installation, and administration. Be aware, however, that the configuration and command line arguments are slightly different. -ccwf
From: Paul Lynch <paul@griffin.plsys.co.uk> Newsgroups: comp.sys.next.bugs Subject: Re: Problems With Nfsd Date: Tue, 26 Sep 1995 08:25:34 GMT Organization: P & L Systems, Ltd Sender: usenet@griffin.plsys.co.uk Message-ID: <DFI7EM.EEq@plsys.co.uk> References: <4487o8$h4r@news.cs.tu-berlin.de> In article <4487o8$h4r@news.cs.tu-berlin.de> kiwi@cs.tu-berlin.de (Axel Habermann) writes: > Studio Malaguti Ass. Prof.Le (MC1116@mclink.it) wrote: > > : I'm not able to mount any exported NFS-volume on my PC/DOS > : with Lantastic for TCP/IP from Artisoft. > : Artisoft claims that NFS server needs to be compatible > : with PCNFSD 2.0 specifications from Sun. > : Is there anybody knowing that newer versions of NeXTstep > : are compatible with such specification ? > : > > Take a look at > /usr/etc/rpc.pcnfsd > e.g. just start it up in your /etc/rc.local This one is only PCNFSD 1.0 capable. This isn't a problem, as most NFS clients provide source for a 2.0 pcnfsd, which sould compile with few problems. Paul -- Paul Lynch (NeXTmail) paul@plsys.co.uk Tel: (01494)432422 P & L Systems Fax: (01494)432478 http://ww.plsys.co.uk/~paul
Date: 26 Sep 1995 10:51:44 GMT From: pp000738@interramp.com Newsgroups: comp.sys.next.bugs Message-ID: <cancel.211414.pp000738@ipiws-001.interramp.com> Control: cancel <211414.pp000738@ipiws-001.interramp.com> Subject: cmsg cancel <211414.pp000738@ipiws-001.interramp.com> EMP/ECP (aka SPAM) cancelled by clewis@ferret.ocunix.on.ca. See news.admin.net-abuse.announce, report 19950926.01 for further details
From: gclem@dannug.dk Newsgroups: comp.sys.next.bugs Subject: Re: little WM bug Date: 25 Sep 1995 07:21:13 GMT Organization: Danish NeXT User Group Distribution: world Message-ID: <445l99$8sd@snaps.dannug.dk> References: <DFEyKA.Mxq@bohica.net> System NXCommandKeys "Check for Disks,D" gives Beep on my 3.3 Intel as well. Using T instead of D makes it work, though (I guess D in some implicit way is used for something else). Geert PS: You can also use Preferences.app to do the dwrites Mark Gregory Salyzyn writes > "Timothy J. Luoma" <luomat@capitalist.princeton.edu> writes: > > >Here's a funny little item. I can't assign the key "D" (uppercase) to > >any menu item in the Workspace Manager. > > > >I've assigned it via dwrite WM NXCommandKeys, and it shows up on the menu > >panel, but <command>-<shift>-d gives me nothing but a system beep. > > System NXCommandKeys "Check for Disks,D" > > Works for me on 3.1, 3.2 and 3.3 on Intel. > > Ciao -- Mark Salyzyn
From: ir003093@interramp.com (Eric Bloom) Newsgroups: comp.sys.next.bugs Subject: compresshelp bug Date: 28 Sep 1995 23:29:47 GMT Organization: Bacchus, Inc. Message-ID: <44fb5b$ijk@usenet4.interramp.com> According to the "Known Problems" section of the NS 3.3 user release notes, when running NS dev. 3.2 with NS 3.3 user on a NeXT machine, you cannot use compresshelp to create the help.store file. The release notes claimed that after accessing the Help, the application may become unusable. They recommend that you use an Intel to create the Help.store file. After compiling an application on a NeXT machine running NS user 3.3 and NS dev. 3.2 we did not notice any problems after accessing the Help (we also did not have any problems accessing any of the Help sections). We then, as recommended in the release notes, created the Help.store file using compresshelp on an Intel running NS 3.3 user. When we accessed certain Help sections, our application quit, i.e., crashed. Does anyone have any additional information on this problem ? Has this problem been fixed when using NS 3.3 user and NS 3.3 developer on either a NeXT or an Intel ? Thanks in advance, Eric Bloom Bacchus, Inc. eric@bacchus.com
From: J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) Newsgroups: comp.text.tex,comp.sys.next.software,comp.sys.next.bugs Subject: Problems with TeXView [NS 3.2 on intel] Followup-To: comp.sys.next.software Date: 29 Sep 1995 15:58:42 GMT Organization: Theory Group, Physics, QMW College Distribution: world Message-ID: <J.M.FIGUEROA.95Sep29165842@strings7.ph.qmw.ac.uk> Hi, For some time I have been living with a very annoying problem: TeXView.app used to behave like a charm with embedded Postscript, but since I upgraded to NS 3.2 it doesn't work at all: it either displays the embedded Postscript in a weird position [some sort of inversion about the origin] or it simply crashes. Has anyone seen this problem? And more importantly, can someone tell me the solution? [Will upgrading to 3.3 do it???] Many thanks in advance, Jose -- +----------------------------------+----------------------------------+ | Dr Jose M Figueroa-O'Farrill | Vox: +44.171.975-5055 | | Department of Physics (Rm. 227) | Fax: +44.181.981-7465 | | Queen Mary and Westfield College | mailto:J.M.Figueroa@qmw.ac.uk | | Mile End Road, London E1 4NS, UK | | | and | | http://stringswww.ph.qmw.ac.uk/~jmf/jmf.html | +----------------------------------+----------------------------------+
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: daa6s@hebb.med.Virginia.EDU (David A. August) Subject: Problem with large user ID #s Message-ID: <DFopsr.KJr@murdoch.acc.Virginia.EDU> Sender: usenet@murdoch.acc.Virginia.EDU Organization: University of Virginia Date: Fri, 29 Sep 1995 20:48:27 GMT I am the sysadmin for a small cluster of NeXT's in a laboratory at a university. Every so often, undergraduate students rotate through our lab. When giving them new accounts, I set their UNIX Uid # on our local network to be the same as the uid on their university mainframe accounts. Recently, however, the new band of youngsters has presented me with a problem: Some of the incoming students have ID #'s on the university mainframe that are larger than that magic quantity, 32768. When I try to assign them this large ID # on our own network, it does a two's complement (?), and starts counting backwards from -32768. So, their directory and files end up being owned by, say, -12345 rather than 53191. To avoid this, I have just assigned these students the default NeXT uid # (e.g, 109), rather than trying to match their university uid #. That's fine for doing local work on our net, but seems to cause problems when the students are using two machines at once. Maybe some handshaking protocol relies on the uid being the same? For example, one student had trouble using an X-Windows simulator (on our machines) to run MatLab (on the university computer). Has anyone out there had similar problems, or know of a fix for this? Any ideas would be greatly appreciated. Thanks, --David August daa6s@virginia.edu
From: hansen@berkeleynetcentral.com Newsgroups: comp.sys.next.bugs Subject: Re: Supra FAX modem won't fax Date: 29 Sep 1995 21:01:50 GMT Organization: Best Internet Communications Message-ID: <44hmru$8l0@shellx.best.com> References: <443a4l$ab@nntp.ucs.ubc.ca> In article <443a4l$ab@nntp.ucs.ubc.ca> sfoy@srtb0511a12.resnet.ubc.ca (Shaun Patrick Foy) writes: > Isn't that a class1 fax modem?? > > On 24 Sep 1995 04:17:07 GMT hansen@berkelynetcentral.com wrote: > >Anyone ever get a Supra FAx modem 144LC to work faxing? > > >Have 040cube 3.2 > > > -- > \ o / o __| \ / |__ o \ o / > | -/\ ___\o \ o | o / o/___ /\- | > / \ | \ /) | ( \ /o\ / ) | (\ / | / \ > > <A HREF="http://philos.resnet.ubc.ca/~sfoy">sfoy@bcu.ubc.ca</A> I checked. Indeed it claims to be Class 1. (By giving a certain AT command it responds with its specifications.)
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 2 Oct 1995 04:15:09 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <44np0d$qgn@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: ccwf@thales.klab.caltech.edu (Charles Fu) Newsgroups: comp.sys.next.bugs Subject: Re: POSIX getcwd is broken in NS 3.3 Date: 2 Oct 1995 23:57:39 GMT Organization: California Institute of Technology, Pasadena Message-ID: <44pu9j$9qi@gap.cco.caltech.edu> References: <DFDpzy.M9L@nvc.cc.ca.us> <4422u0$k0i@babyblue.cs.yale.edu> <444vnv$3c2@gap.cco.caltech.edu> In article <444vnv$3c2@gap.cco.caltech.edu>, Charles Fu <ccwf@thales.klab.caltech.edu> wrote: > It's the POSIX readdir() function that's broken, as I have posted here > previously, several months ago. I believe this is bug is documented in > KBNS.... [It is.] > The fix is quite simple--get working source for readdir() from a POSIX system > (for example from 4.4BSD) and use that instead of NeXT's readdir() > function. You can just include this in your sources and link against it to > prevent NeXT's broken function from being used. If you want to be more > daring, you can remove the broken function from libposix.a (it's a normal > library, not shared, so you can manipulate it with standard tools), and > replace it with a fixed version. Since the library is not shared, this will > not fix POSIX programs which have already been linked with the broken > function. It only helps with programs you create. The last time I actually did this patch was quite some time ago, and I forgot about the direct vs dirent issue (getdirentries returns data which needs to be "reformatted" before putting it into a dirent). This is a relatively minor and straightforward issue, but it might trip up folks not used to doing system programming and who just want a drop-in replacement. Here is my replacement readdir() function I have been using on those rare occasions when I am forced to use the POSIX library. I have not observed any problems with it, but I do not claim it is bug free. Suggestions for improvements are welcome. Note that the return value is not handled in the same way as NeXT's readdir() function. I return the value in a static buffer instead of using (a dynamically allocated) dd_buf. If, for some bizarre reason, your supposedly POSIX-compliant program relies upon this undocumented behavior, it will break. I would be happy if NeXT would release an official replacement readdir() function. OTOH, if they are installing enough BSD4.4 internals for NEXTSTEP 4.0, the problem should just go away (BSD4.4 is POSIX compliant). So they may be unwilling to deal with it. -ccwf =============================================================================== /* patched readdir() function for use with POSIX -ccwf */ extern int getdirentries(int fd, char *buf, int nbytes, long *basep); /* * Copyright (c) 1983, 1993 * The Regents of the University of California. All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. All advertising materials mentioning features or use of this software * must display the following acknowledgement: * This product includes software developed by the University of * California, Berkeley and its contributors. * 4. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. */ #if defined(LIBC_SCCS) && !defined(lint) static char sccsid[] = "@(#)readdir.c 8.1 (Berkeley) 6/4/93"; #endif /* LIBC_SCCS and not lint */ #include <sys/param.h> #include <dirent.h> /* * get next entry in a directory. */ struct dirent * readdir(dirp) register DIR *dirp; { register struct direct *dp; #ifdef __NeXT__ static struct dirent de; #endif for (;;) { if (dirp->dd_loc == 0) { dirp->dd_size = getdirentries(dirp->dd_fd, #ifdef __NeXT__ dirp->dd_buf, dirp->dd_bsize, &dirp->dd_bbase); #else dirp->dd_buf, dirp->dd_len, &dirp->dd_seek); #endif if (dirp->dd_size <= 0) return NULL; } if (dirp->dd_loc >= dirp->dd_size) { dirp->dd_loc = 0; continue; } #ifdef __NeXT__ dp = (struct direct *)(dirp->dd_buf + dirp->dd_loc); #else dp = (struct dirent *)(dirp->dd_buf + dirp->dd_loc); #endif if ((int)dp & 03) /* bogus pointer check */ return NULL; if (dp->d_reclen <= 0 || #ifdef __NeXT__ dp->d_reclen > dirp->dd_bsize + 1 - dirp->dd_loc) #else dp->d_reclen > dirp->dd_len + 1 - dirp->dd_loc) #endif return NULL; dirp->dd_loc += dp->d_reclen; if (dp->d_ino == 0) continue; #ifdef __NeXT__ de.d_off = dirp->dd_loc; de.d_fileno = dp->d_fileno; de.d_reclen = dp->d_reclen; de.d_namlen = dp->d_namlen; strcpy(de.d_name, dp->d_name); return &de; #else return (dp); #endif } }
From: Howard Ship Newsgroups: comp.sys.next.bugs Subject: Terminal: permissions suspicious? Date: 3 Oct 1995 15:49:41 GMT Organization: Stratus Computer Inc, Marlboro MA Distribution: us Message-ID: <44rm2l$7b3@transfer.stratus.com> Any idea what the following error is all about? I get it when I attempt to invoke Terminal services: Permissions on /net/zion/cac1/hship/.NeXT/Terminal.svcs look suspicious. Terminal won't be able to export any services. Service Insert .signature could not be provided: There doesn't seem to be a list of services for Terminal. I have no clue why this suddenly started to occur. I am unaware of any event which may have caused it. I've been on 3.3 for some time. I fiddled with the permissions of ~/.Next/Terminal.svcs and ~/.Next/services/Terminal, adding and removing likely culrpits (such as group and other access), to no avail. I've done a chown on all of my files to make sure they were all the same group (wheel). In terms of ownership and permissions, I have Librarian services setup that still work correctly. ~/.NeXT -rwxrwxrwx 1 hship wheel 295 Oct 3 11:30 Terminal.svcs* ~/.NeXT/services -rwxrwxrwx 1 hship wheel 771 Oct 3 11:29 Librarian* -rwxrwxrwx 1 hship wheel 1118 Oct 3 11:30 Terminal* Any clues?
From: J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) Newsgroups: comp.sys.next.bugs Subject: Re: Problems with TeXView [NS 3.2 on intel] Followup-To: comp.sys.next.bugs Date: 03 Oct 1995 23:17:20 GMT Organization: Theory Group, Physics, QMW College Distribution: world Message-ID: <J.M.FIGUEROA.95Oct4001720@strings7.ph.qmw.ac.uk> References: <J.M.FIGUEROA.95Sep29165842@strings7.ph.qmw.ac.uk> In-reply-to: J.M.Figueroa@QMW.ac.uk's message of 29 Sep 1995 15:58:42 GMT This is a follow-up of my own post of last week: Me> TeXView.app used to behave like a charm with embedded Postscript, but Me> since I upgraded to NS 3.2 it doesn't work at all: ... Here is a more detailed description of the problem: The embedded PostScript file gets translated to the upper left hand corner of the display instead of being displayed where it should be. Sometimes TeXview just crashes, I have not been able to determine a pattern for this behaviour and I can't reproduce it. At the same time I get a bunch of errors in the console. Here is a recent example: (As usual, all lines are prefaced by the host, the time and TeXview["pid"]): DPS client library error: PostScript program error, DPSContext 2ec50 %%[ Error: undefined; OffendingCommand: @beginspecial ]%% DPS client library error: PostScript program error, DPSContext 2ec50 %%[ Error: undefined; OffendingCommand: @llx ]%% DPS client library error: PostScript program error, DPSContext 2ec50 %%[ Error: undefined; OffendingCommand: @lly ]%% DPS client library error: PostScript program error, DPSContext 2ec50 %%[ Error: undefined; OffendingCommand: @urx ]%% DPS client library error: PostScript program error, DPSContext 2ec50 %%[ Error: undefined; OffendingCommand: @ury ]%% DPS client library error: PostScript program error, DPSContext 2ec50 %%[ Error: undefined; OffendingCommand: @rwi ]%% DPS client library error: PostScript program error, DPSContext 2ec50 %%[ Error: undefined; OffendingCommand: @setspecial ]%% DPS client library error: PostScript program error, DPSContext 2ec50 %%[ Error: undefined; OffendingCommand: @endspecial ]%% DPS client library error: Invalid tag in returned object, DPSContext 2ec50, data 503076 DPS client library error: Invalid tag in returned object, DPSContext 2ec50, data 503092 DPS client library error: Invalid tag in returned object, DPSContext 2ec50, data 503108 DPS client library error: Invalid tag in returned object, DPSContext 2ec50, data 503124 At the same time, dvips works fine. I should also mention that I have four NS machines available to me right now: two intel and two Motorola, all four running NS 3.2. The Motorola machines work fine, but both intel machines give the same kind of errors. Any ideas??? A million thanks in advance, Jose -- +----------------------------------+----------------------------------+ | Dr Jose M Figueroa-O'Farrill | Vox: +44.171.975-5055 | | Department of Physics (Rm. 227) | Fax: +44.181.981-7465 | | Queen Mary and Westfield College | mailto:J.M.Figueroa@qmw.ac.uk | | Mile End Road, London E1 4NS, UK | | | and | | http://stringswww.ph.qmw.ac.uk/~jmf/jmf.html | +----------------------------------+----------------------------------+
From: Estascy@euphoria.com Newsgroups: comp.sys.next.bugs Subject: Euphoria Tape...Try it with a Lover! Date: 4 Oct 1995 12:41:59 GMT Organization: Internet Direct, Inc. Message-ID: <44tven$cha@globe.indirect.com> Originator: marcus@bud.indirect.com I.C.P., Suite 578-101, 5360 E. Bay Dr., Clwr., FL 34624 The Euphoria Tape Do The Euphoria Tape alone the first time, and then with a lover. It's a phenomenon you will feel within the first 3 minutes of use -- what you feel the rest of the time is unbelievable. The Euphoria Tape. It was first whispered about on college campuses, and only experimented with in seclusion, or among close friends in private settings, but it has now become a cult rush and is considered the next passion of our Cyber lifestyles. The Euphoria Tape. Not a drug, not a chemical, but an audio stimulation of brain centers controlling pleasure. And agony. And ecstacy. The buzz. An audio mesmer signal strokes the base of the mind, choking out pain sensations and producing the violent flow of... euphoria. The Euphoria Tape. Copies have changed hands from friends to friends to other friends. It is not regulated, there are no laws yet restricting use of The Euphoria Tape. The Euphoria Tape is rushing the masses. ************************************************* Stereo audio cassette. Scientific treatment included. Shipped immediately upon receipt of $11.88 personal check or money order. Absolute satisfaction guaranteed (and experienced) or full purchase price refunded. Offered only to adults 18 years or older. ******************* IntraNet Cyber Products Suite 578-101 5360 E. Bay Dr. Clearwater, FL 34624 ******************* Tape Duplication prohibited. $2.00 for Scientific Treatment. Copyright 1995 Intranet Cyber Products, All Rights Reserved. -------------------------------------------------------
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <476TJL@ptsmail.ptsem.edu> From: 476TJL@ptsmail.ptsem.edu (TIMOTHY LUOMA @ TEMPLE * LUOMA,TIMOTHY) Subject: unable to login Message-ID: <608A723001A23A7C@-SMF-> Date: 04 Oct 95 09:22:00 EDT Sometimes (I can't seem to figure out what leads to it) when I logout I can't login -- or at least, I can't stay logged in. I logout and it seems normal, and then when I try to login the next time, everything seems to be OK and then my LogoutHook runs and I'm back at the login panel. I can login as root and I can login as 'console' but not as my normal self. Last time this happened, I ran a 'ps -auxwwww' loop while I tried to login, and the only strange thing I saw is this below: ---- This appeared only once ---- root 2930 54.3 0.6 1.59M 208K co S 0:02 - std.9600 console (getty) --- the next time through, these 3 lines show up ---- root 2915 43.6 7.9 20.0M 2.52M ? S 0:03 - console (WindowServer) root 2916 37.5 5.6 4.02M 1.80M ? S 0:02 - console (loginwindow) root 2917 0.0 0.7 2.66M 224K ? S 0:00 - console (WindowServer) So it looks like there are two windowservers, and perhaps the first one is not letting the second one take over? Does anyone have any comments or suggestions relating to this perplexing event? Thanks! TjL -- Timothy J. Luoma <476tjl@ptsmail.ptsem.edu> ASCII Email only! No NeXTMail or MIME please NOTE: My stupid DOS mail reader strips off REPLY-TO lines. Please make sure to include your email address within the text of your message.
Date: 4 Oct 1995 12:47:21 GMT From: Estascy@euphoria.com Newsgroups: comp.sys.next.bugs Message-ID: <cancel.44tven$cha@globe.indirect.com> Control: cancel <44tven$cha@globe.indirect.com> Subject: cmsg cancel <44tven$cha@globe.indirect.com> EMP/ECP (aka SPAM) cancelled by clewis@ferret.ocunix.on.ca. See news.admin.net-abuse.announce, report 19951004.04 for further details
From: Mark Crispin <mrc@CAC.Washington.EDU> Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: Wed, 4 Oct 1995 14:47:43 -0700 Organization: Networks & Distributed Computing Message-ID: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII NNTP-Posting-User: peter I finally managed to catch this bug during a period when nothing else could be blamed; specifically, a freshly-rebooted system which was then left alone for a week. It's a memory leak in Preferences.app. It had gobbled down over 60MB of virtual memory and was still going strong. To be sure, I tried with a fresh reboot and started close monitoring of Preferences.app. Sure enough, it slowly but steadily grows over real time, gaining about 1MB every 3 hours. -- Mark -- DoD #0105, R90/6 pilot, FAX: (206) 685-4045 ICBM: N 47 39'35" W 122 18'39" Science does not emerge from voting, party politics, or public debate.
From: J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) Newsgroups: comp.sys.next.bugs Subject: Re: Problems with TeXView [NS 3.2 on intel] Followup-To: comp.sys.next.bugs Date: 04 Oct 1995 22:20:10 GMT Organization: Theory Group, Physics, QMW College Distribution: world Message-ID: <J.M.FIGUEROA.95Oct4232010@strings7.ph.qmw.ac.uk> References: <J.M.FIGUEROA.95Sep29165842@strings7.ph.qmw.ac.uk> <J.M.FIGUEROA.95Oct4001720@strings7.ph.qmw.ac.uk> In-reply-to: J.M.Figueroa@QMW.ac.uk's message of 03 Oct 1995 23:17:20 GMT This is the third post in the series. I have now identified the problem I've been having with TeXview. TeXview is simply not reading the prologue file /usr/lib/tex/ps/special.pro. If I happen to place a copy of special.pro in the directory with the .dvi file, then things work smoothly. Otherwise I get the errors I listed in my previous post. Does anyone know the correct environment variable that TeXview looks up to find the prologue directory? I think this should about wrap this whole thing up. To Mark Doyle, Henning Heinze, Raf Schietekat and Juergen Sell, who answered my posts and offered many helpful suggestions, many thanks! Cheers, Jose -- +----------------------------------+----------------------------------+ | Dr Jose M Figueroa-O'Farrill | Vox: +44.171.975-5055 | | Department of Physics (Rm. 227) | Fax: +44.181.981-7465 | | Queen Mary and Westfield College | mailto:J.M.Figueroa@qmw.ac.uk | | Mile End Road, London E1 4NS, UK | | | and | | http://stringswww.ph.qmw.ac.uk/~jmf/jmf.html | +----------------------------------+----------------------------------+
Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy From: mcbinc@world.std.com (Monty C Brandenberg) Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Message-ID: <DFyJLB.BHL@world.std.com> Followup-To: comp.sys.next.bugs,comp.sys.next.advocacy Organization: MCB, Incorporated. Cambridge, MA References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> Date: Thu, 5 Oct 1995 04:10:23 GMT Mark Crispin (mrc@CAC.Washington.EDU) wrote: : I finally managed to catch this bug during a period when nothing else : could be blamed; specifically, a freshly-rebooted system which was then : left alone for a week. : It's a memory leak in Preferences.app. It had gobbled down over 60MB of : virtual memory and was still going strong. Not to detract from your effort but I reported that one in c.s.n.bugs at least six months ago looking for verification from other users. Not a bloody peep from *anyone*. Does anyone actually use the damned software? Best workaround: kill Preferences from the command line. Do *not* unhide it and quit. That just thrashes the VM. I'd suggest a crontab entry to kill and restart a new Preferences. I'd really like to kick someone in the head for this one... monty -- Monty Brandenberg Software Consultant MCB, Inc. mcbinc@world.std.com 617.864.6907
Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy From: G.C.Th.Wierda@AWT.NL (Gerben Wierda) Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Message-ID: <DFz55x.38q@AWT.NL> Sender: news@AWT.NL Organization: Adviesraad voor het Wetenschaps- en Technologiebeleid References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> Mime-Version: 1.0 Date: Thu, 5 Oct 1995 11:56:21 GMT In article <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washingto n.EDU>, mrc@CAC.Washington.EDU says... > >I finally managed to catch this bug during a period when nothing else >could be blamed; specifically, a freshly-rebooted system which was then >left alone for a week. > >It's a memory leak in Preferences.app. It had gobbled down over 60MB of >virtual memory and was still going strong. > >To be sure, I tried with a fresh reboot and started close monitoring of >Preferences.app. Sure enough, it slowly but steadily grows over real >time, gaining about 1MB every 3 hours. There must be a way to narrow this down. I haven't seen this behaviour on machines which are left on for weeks. So what part of Preferences is to blame? Any non-standard modules, any special usage? Let's look into this. --Gerben
From: art@cubicsol.com (Art Isbell) Newsgroups: comp.sys.next.bugs Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 5 Oct 1995 15:27:47 GMT Organization: Sense Networking Seattle (www.oz.net) Distribution: world Message-ID: <450thj$hd@emerald.oz.net> References: <DFz55x.38q@AWT.NL> In article <DFz55x.38q@AWT.NL> G.C.Th.Wierda@AWT.NL (Gerben Wierda) writes: > There must be a way to narrow this down. I haven't seen this behaviour on > machines which are left on for weeks. So what part of Preferences is to blame? > Any non-standard modules, any special usage? > I started Preferences just as Workspace does so that it is opened hidden, doesn't display its panel, and so doesn't load any modules. Depending on which clock the icon displays, the memory leak rate varies. For the clock-calendar combination which updates only once per minute, the leak rate ranged from only 10 - 80 bytes/minute. But for those clocks that update each second, the leak rate was from 1400 - 2200 bytes/minute, or about 3 MB/day max which is less than Mark observed, but still bad. I didn't spend a lot of time observing these leaks, so the above rates are only approximate. Mark's machine also runs an hourly chime using, I believe, Scott Hess' chime utility that writes, plays, and deletes a score file. I run chime 4 times/hour and have noticed a possible memory leak when scores are played in general (nothing to do with Scott's implementation). So maybe this is contributing to Mark's observed leak rate. --- Art Isbell NeXTmail: art@cubicsol.com NeXT Registered Consultant Voice/Fax: +1 408 335 2515 Trego Systems Voice Mail: +1 408 335 1154 CaseServ: NEXTSTEP managed care US Mail: Felton, CA 95018-9442 contract and case management solutions
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <476TJL@ptsmail.ptsem.edu> From: 476TJL@ptsmail.ptsem.edu (TIMOTHY LUOMA @ TEMPLE * LUOMA,TIMOTHY) Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Message-ID: <6403743001A23A7C@-SMF-> Date: 05 Oct 95 12:41:00 EDT Ok, stupid problem, but could you use the 3.2 Preferences.app? TjL, who's still using 3.2 and doesn't know any better -- Timothy J. Luoma <476tjl@ptsmail.ptsem.edu> ASCII Email only! No NeXTMail or MIME please NOTE: My stupid DOS mail reader strips off REPLY-TO lines. Please make sure to include your email address within the text of your message.
From: robin@pencom.com Newsgroups: comp.sys.next.bugs Subject: Strange FileViewer bug... Date: 5 Oct 1995 19:21:41 GMT Organization: Pencom Software Message-ID: <451b85$ild@digdug.pencom.com> I logged into my NeXTstation Turbo this morning after being away for 2 days. I'm running NEXTSTEP 3.3. I tried to copy files from my Mail.app to my FileViewer window (into my Home directory), and it wouldn't let me. If I tried to copy files into "/tmp" or some other directory that I don't own -- it didn't seem to have a problem. It was as if FileViewer didn't know who I was... I couldn't even remove files, or create a new folder in a directory that I owned (and had approproate permissions for). Rebooting cleared the problem. Has anyone else seen this? Does anyone have any ideas as to what caused it? -- ----------------------------------------------------------------------------- *** These are my opinions... Mine! All Mine! Minemineminemineminemine! *** ----------------------------------------------------------------------------- Robin D. Wilson robin@pencom.com Pencom Software 701 Canyon Bend Dr. 9050 Capital of Texas Hwy Pflugerville, TX 78660 Austin, TX 78759 (512) 251-1737 (512) 343-6666
From: Garance A Drosehn <gad@eclipse.its.rpi.edu> Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 5 Oct 1995 00:18:31 GMT Organization: Rensselaer Polytechnic Institute, Troy NY, USA Message-ID: <44v88n$gcj@usenet.rpi.edu> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> Mark Crispin <mrc@CAC.Washington.EDU> wrote: > I finally managed to catch this bug during a period when nothing > else could be blamed; specifically, a freshly-rebooted system > which was then left alone for a week. > It's a memory leak in Preferences.app. It had gobbled down over > 60MB of virtual memory and was still going strong. I wonder what it depends on. Presumably it depends on which clock you have running in the icon, as I can't think what else in preferences would be running constantly. The clock I have set is the one with the time in LED's at the top of the icon, and the date looking like it's on a pad of white paper on the bottom two-thirds of it. > To be sure, I tried with a fresh reboot and started close > monitoring of Preferences.app. Sure enough, it slowly but > steadily grows over real time, gaining about 1MB every 3 hours. You're measuring via 'ps', right? In my case I have been logged in for 12 days, and the process for Preferences.app was at 7.99 meg. I killed and restarted it, and it's at 7.64 meg. I wonder if there's something else which would cause the difference besides the setting of the clock icon. You're on NeXT hardware, right? My NS-3.3 machine is NS/Intel, so my results might be different because of that. Or is it that you're letting Preferences run, but you have to watch the swapfile (instead of 'ps' output) to see the growth? --- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
From: Mark Crispin <mrc@CAC.Washington.EDU> Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: Thu, 5 Oct 1995 13:27:51 -0700 Organization: Networks & Distributed Computing Message-ID: <Pine.NXT.3.92.951005132520.1474C-100000@Tomobiki-Cho.CAC.Washington.EDU> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFz55x.38q@AWT.NL> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII NNTP-Posting-User: hellman To: Gerben Wierda <G.C.Th.Wierda@AWT.NL> In-Reply-To: <DFz55x.38q@AWT.NL> On Thu, 5 Oct 1995, Gerben Wierda wrote: > There must be a way to narrow this down. I haven't seen this behaviour on > machines which are left on for weeks. So what part of Preferences is to blame? > Any non-standard modules, any special usage? No non-standard modules. Apparently, if you have the clock that updates every second, Preferences in 3.3 leaks 2200 bytes each minute, which translates into 3MB/day. -- Mark -- DoD #0105, R90/6 pilot, FAX: (206) 685-4045 ICBM: N 47 39'35" W 122 18'39" Science does not emerge from voting, party politics, or public debate.
From: hendryj@Mercury.mcs.com (Jonathan W. Hendry) Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Followup-To: comp.sys.next.bugs,comp.sys.next.advocacy Date: 5 Oct 1995 21:51:43 GMT Organization: MCSNet Internet Services Message-ID: <451k1f$m2d@News1.mcs.net> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <44v88n$gcj@usenet.rpi.edu> Garance A Drosehn (gad@eclipse.its.rpi.edu) wrote: : > To be sure, I tried with a fresh reboot and started close : > monitoring of Preferences.app. Sure enough, it slowly but : > steadily grows over real time, gaining about 1MB every 3 hours. : You're measuring via 'ps', right? In my case I have been logged : in for 12 days, and the process for Preferences.app was at 7.99 : meg. I killed and restarted it, and it's at 7.64 meg. : I wonder if there's something else which would cause the difference : besides the setting of the clock icon. You're on NeXT hardware, : right? My NS-3.3 machine is NS/Intel, so my results might be : different because of that. : Or is it that you're letting Preferences run, but you have to : watch the swapfile (instead of 'ps' output) to see the growth? I just used leaks on Preferences.app. First, without having used it. Then after opening it, and 'walking' through the various modules. I also switched to each of the clock styles: uptime: 5:49pm up 2:02, 1 user, load average: 0.16, 0.38, 0.44 Process 239: 323 nodes malloced Process 239: 1 leaks Leak: 0x0004d0c0 size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x43800000 0x43080000 0x00000060 0x00000000 0x0004d68c 0x00000000 0x00000000 0x00000000 0x00000430 Then after using Preferences.app: > leaks 239 Process 239: 3503 nodes malloced Process 239: 26 leaks Leak: 0x0009bab8 size=32 instance of 'NXStringTable' 0x04018190 0x0000000a 0x0509f824 0x0509f820 0x0000000f 0x000a2bec 0x00000002 0x00000140 Leak: 0x0009ccd8 size=12 string 'English' Leak: 0x0009d0fc size=12 string 'German' Leak: 0x0009d1bc size=12 string 'Spanish' Leak: 0x00097f5c size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x00000000 0x00000000 0x00000160 0x00000000 0x00097f88 0x00000000 0x00000000 0x00000000 0x00000090 Leak: 0x0009d1d4 size=12 string 'Swedish' Leak: 0x00098174 size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x00000000 0x00000000 0x00000160 0x00000000 0x000981a0 0x00000000 0x00000000 0x00000000 0x00000090 Leak: 0x0009d114 size=12 string 'Italian' Leak: 0x000a1f90 size=304 0x69570000 0x00000120 0x00000120 0x0005059c 0x00000005 0x0005bb84 0x00000000 0xffffffff 0x00000000 0x00000000 0x00000000 0x00050378 0x00000010 0x0005bb84 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x0005059c 0x00000005 0x0005bb84 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00050378 0x0000000a 0x0005bb84 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x0005059c 0x00000003 0x0005bb84 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00050378 0x00000004 0x0005bb84 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x0005059c 0x00000008 0x0005bb84 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00050378 0x00000011 0x0005bb84 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00050378 0x00000001 0x000a142c 0x00000000 0xffffffff 0x00000000 0x00000000 0x00000000 0x000001f0 Leak: 0x00053870 size=36 string 'Objects/Menus/Main/Info.rtfd' Leak: 0x0009d0f0 size=12 string 'French' Leak: 0x00098b30 size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x00000000 0x00000000 0x00000160 0x00000000 0x00098b5c 0x00000000 0x00000000 0x00000000 0x00000090 Leak: 0x0005ad10 size=28 instance of 'NXStringTable' 0x04018190 0x00000000 0x0004864e 0x00048650 0x00000001 0x00050798 0x00000040 Leak: 0x00098c90 size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x00000000 0x00000000 0x00000160 0x00000000 0x00098cbc 0x00000000 0x00000000 0x00000000 0x00000090 Leak: 0x00098010 size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x00000000 0x00000000 0x00000160 0x00000000 0x0009803c 0x00000000 0x00000000 0x00000000 0x00000090 Leak: 0x00180330 size=56 instance of 'NXSoundOut' 0x061879f0 0x06126cdc 0x00000021 0x00000025 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x001803e8 0x00000000 0x00000000 0x00000030 Leak: 0x00097eac size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x00000000 0x00000000 0x00000160 0x00000000 0x00097ed8 0x00000000 0x00000000 0x00000000 0x00000090 Leak: 0x0009d1c8 size=12 string 'Italian' Leak: 0x0009d0e4 size=12 string 'English' Leak: 0x0004d0c0 size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x43800000 0x43080000 0x00000060 0x00000000 0x0004d68c 0x00000000 0x00000000 0x00000000 0x000001a0 Leak: 0x0009d184 size=44 string '/NextApps/Preferences.app/English.lproj' Leak: 0x0009d120 size=12 string 'Swedish' Leak: 0x0009d108 size=12 string 'Spanish' Leak: 0x0009cce4 size=16 string 'French' Leak: 0x00098be0 size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x00000000 0x00000000 0x00000160 0x00000000 0x00098c0c 0x00000000 0x00000000 0x00000000 0x00000090 Leak: 0x000980c0 size=44 instance of 'NXImage' 0x06186c80 0x00000000 0x00000000 0x00000000 0x00000160 0x00000000 0x000980ec 0x00000000 0x00000000 0x00000000 0x00000090
From: hendryj@Mercury.mcs.com (Jonathan W. Hendry) Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 5 Oct 1995 21:58:52 GMT Organization: MCSNet Internet Services Message-ID: <451kes$m2d@News1.mcs.net> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <44v88n$gcj@usenet.rpi.edu> <451k1f$m2d@News1.mcs.net> How about a little doodad which runs leaks on all processes? Anyone want to whip up a quick shellscript? - Jon
From: deniseh@shell1.best.com (Denise Howard) Newsgroups: comp.sys.next.bugs Subject: Re: Strange FileViewer bug... Date: 6 Oct 1995 00:03:45 GMT Organization: Best Internet Communications Message-ID: <451rp1$c8d@shellx.best.com> References: <451b85$ild@digdug.pencom.com> robin@pencom.com wrote: : I logged into my NeXTstation Turbo this morning after being away for 2 days. : I'm running NEXTSTEP 3.3. : I tried to copy files from my Mail.app to my FileViewer window (into my Home : directory), and it wouldn't let me. If I tried to copy files into "/tmp" or : some other directory that I don't own -- it didn't seem to have a problem. It : was as if FileViewer didn't know who I was... I couldn't even remove files, or : create a new folder in a directory that I owned (and had approproate permissions : for). : Rebooting cleared the problem. : Has anyone else seen this? Does anyone have any ideas as to what caused it? Similar symptoms can be caused by adding a new disk to the network and moving peoples' home directories to it without rebooting the machines on the network. Until they are rebooted they will act like they've forgotten who you are. If you are still logged in you'll see a question mark as your home directory icon. Denise -- Denise Howard | PROGRAM, tr. v., An activity similar to Mountain View, CA | banging one's head against a wall, but deniseh@best.com | with fewer opportunities for reward. NeXTMail welcome! | http://www.best.com/~deniseh
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 6 Oct 1995 03:45:06 GMT Organization: Express Access Online Communications, USA Message-ID: <4528o2$psi@news4.digex.net> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> Mark Crispin <mrc@CAC.Washington.EDU> wrote: > It's a memory leak in Preferences.app. It had gobbled down over > 60MB of virtual memory and was still going strong. Great going Mark! Thanks very much for finding it. The first step to a cure is finding out what is wrong. Now I guess the next step (npi) is to get NeXT to fix it...although, maybe 4.0 will take care of this (while hopefully not introducing new tricky bugs). -- Thanks, take care, later, John Kheit )^> %^) monoChrome, Inc. | New York Law School NEXTSTEP Developer | Opinions expressed represent me only... Telepathy, It's coming | MIME & NeXTmail OK--PPP (3P) jkheit@cnj.digex.net | http://cnj.digex.net/~jkheit
From: Garance A Drosehn <gad@eclipse.its.rpi.edu> Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 5 Oct 1995 16:13:47 GMT Organization: Rensselaer Polytechnic Institute, Troy NY, USA Message-ID: <45107r$bjo@usenet.rpi.edu> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFyJLB.BHL@world.std.com> mcbinc@world.std.com (Monty C Brandenberg) wrote: > Mark Crispin (mrc@CAC.Washington.EDU) wrote: > : I finally managed to catch this bug during a period when nothing > : else could be blamed; specifically, a freshly-rebooted system > : which was then left alone for a week. > : It's a memory leak in Preferences.app. It had gobbled down over > : 60MB of virtual memory and was still going strong. > Not to detract from your effort but I reported that one in > c.s.n.bugs at least six months ago looking for verification from > other users. Not a bloody peep from *anyone*. Does anyone > actually use the damned software? Not to detract from your sarcasim, but I didn't peep at the time because I most certainly do not see this behavior. The question is why you'd see it and others do not. > Best workaround: kill Preferences from the command line. Do > *not* unhide it and quit. That just thrashes the VM. I'd suggest > a crontab entry to kill and restart a new Preferences. I don't understand this part. If you quit preferences right away, right after logging on, then I wouldn't expect any problems with thrashing. Better yet, I wonder if you could change the Workspace preferences so that Preferences.app is not autostarted. --- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
From: e9026354@fbma.tuwien.ac.at (Wolfgang Holzinger) Newsgroups: comp.sys.next.bugs Subject: Re: NeXT Dev Documentation Error Date: 6 Oct 1995 12:41:45 GMT Organization: Vienna University of Technology, Austria Message-ID: <453869$i1n@news.tuwien.ac.at> References: <4484lo$g9j@darkstar.UCSC.EDU> Alvin Jee (alvin@cse.ucsc.edu) wrote: : Hello! : Has anybody else noticed that the on-line docs for the Application class : and the Action Cell class are virtually the same document? I even checked the : Dev CD and they are the same. Sort of defeats the purpose of on-line docs :) Glad to hear that. I was getting suspicious on my filesystem... But where do we get the missing file from? Ciao Wolfgang
From: madler@cco.caltech.edu (Mark Adler) Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 6 Oct 1995 17:04:07 GMT Organization: California Institute of Technology, Pasadena Message-ID: <453ni7$1ou@gap.cco.caltech.edu> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFyJLB.BHL@world.std.com> Perhaps more bug-hunting could determine which .preferences module in Perferences.app is the guilty party (on the optimistic assumption it's not the main part of the application). Then the offending one could be removed unless and until needed. mark p.s. I never thought there might be a vm suck since I use Mathematica which has always been a vm suck since NeXTstep 1.0.
From: eric.hermanson@attws.com (Eric Hermanson) Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 6 Oct 1995 19:11:57 GMT Organization: AT&T Wireless Services, Inc. Distribution: world Message-ID: <453v1t$bu0@nwestmail.nwest.mccaw.com> References: <451k1f$m2d@News1.mcs.net> In article <451k1f$m2d@News1.mcs.net> hendryj@Mercury.mcs.com (Jonathan W. Hendry) writes: > I just used leaks on Preferences.app. > First, without having used it. Then after opening it, and 'walking' > the various modules. I also switched to each of the clock styles: > Process 239: 323 nodes malloced > Process 239: 1 leaks > Then after using Preferences.app: > Process 239: 3503 nodes malloced > Process 239: 26 leaks There are also serious memory leaks in HeaderViewer: Edit.app (after running 7 hours with heavy use) 1,304,868 Malloced Bytes 198 Occupied Pages 8780 Used Nodes, 413 Free Nodes HeaderViewer.app (after running for about 2 hours with heavy use) 35,806,996 Malloced Bytes 5082 Occupied Pages 1171638 Used Nodes, 43607 Free Nodes NeXT acknowledges that HeaderViewer malloc's memory when a new header is viewed, but it fails to free the memory. So the more you use HeaderViewer (and probably the more you use Preferences) the more leaks you will have. Eric
From: katzlbt@vuse.vanderbilt.edu (Thomas Katzlberger) Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Followup-To: comp.sys.next.bugs,comp.sys.next.advocacy Date: 7 Oct 1995 17:41:09 GMT Organization: a black NeXT Message-ID: <456e3l$2k7@news.vanderbilt.edu> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFyJLB.BHL@world.std.com> <453ni7$1ou@gap.cco.caltech.edu> Mark Adler (madler@cco.caltech.edu) wrote: : Perhaps more bug-hunting could determine which .preferences module in : Perferences.app is the guilty party (on the optimistic assumption it's : not the main part of the application). Then the offending one could : be removed unless and until needed. As far as I know all modules get loaded after the first activation (double click) on Preferences.app. So if this swapfile leak shows up also after an autolaunch without double-click on Preferences.app it might be a leak related to the clock(?). Well, just wild guesses. Cat. -- _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ _/_/ _/_/ _/_/ Thomas Katzlberger _/_/ _/_/ katzlbt@vuse.vanderbilt.edu _/_/ _/_/ @aBlackNeXT.called.garfield _/_/ _/_/ http://www.vuse.vanderbilt.edu/~katzlbt/ _/_/ _/_/ _/_/ _/_/ "You can tune a file system, but you can't tune a fish." _/_/ _/_/ UNIX man page for tunefs. _/_/ _/_/ _/_/ _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy From: karl@trapac.com (Karl Kraft) Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Message-ID: <DG3CKr.wq@trapac.com> Organization: Trans Pacific Container Service Corporation References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFz55x.38q@AWT.NL> <Pine.NXT.3.92.951005132520.1474C-100000@Tomobiki-Cho.CAC.Washington.EDU> Date: Sat, 7 Oct 1995 18:26:51 GMT In article <Pine.NXT.3.92.951005132520.1474C-100000@Tomobiki-Cho.CAC.Washington.EDU> Mark Crispin <mrc@CAC.Washington.EDU> writes: >On Thu, 5 Oct 1995, Gerben Wierda wrote: >> There must be a way to narrow this down. I haven't seen this behaviour on >> machines which are left on for weeks. So what part of Preferences is to blame? >> Any non-standard modules, any special usage? > >No non-standard modules. > >Apparently, if you have the clock that updates every second, Preferences >in 3.3 leaks 2200 bytes each minute, which translates into 3MB/day. > Preferences doesn't actually leak the memory. It just mallocs more and more unneeded memory without freeing it, while keeping a reference to it. In such a case, even a garbage collector for the swapfile would not solve the problem. Also note that simply quitting Preference will free the memory to be reused by other processes. Using a clock that updates 1 per second, 1 node per second is being allocated by Preferences. Through clever usage of gdb (left as an exercise for the reader), you can see that the node is an NSDate, created by the method +date of the NXDate class cluster. +date is called by the -updateClock method of the IconClock class. This works out to 3.1 meg per day if your clock updates once per second, or about 52K per day if your clock updates once per minute. For those who want a detailed example of the problem: myhost> dread -o Preferences Preferences IconClockType 4 Preferences NXColorPanelMode 0 Preferences "NXWindow Frame NXColorPanel" "580 336 196 298 " Preferences NXFontPanelPreviewFrame "0 0 281 47" Preferences NXSwatchSize 22.000000 Preferences "NXWindow Frame NXFontPanel" "108 129 299 330 " Preferences HideClockType 4 Preferences NXColorPickerListDefaults 1NeXT Preferences NXColorPickerSlidersDefaults 0 myhost>/NextApps/Preferences.app/Preferences & [1] 13220 myhost>while (1) while>leaks 13220 | grep "nodes malloced" while>sleep 5 while>end Process 13220: 2435 nodes malloced Process 13220: 2437 nodes malloced Process 13220: 2444 nodes malloced Process 13220: 2450 nodes malloced Process 13220: 2457 nodes malloced Process 13220: 2465 nodes malloced Process 13220: 2471 nodes malloced Process 13220: 2476 nodes malloced Process 13220: 2481 nodes malloced Process 13220: 2487 nodes malloced Process 13220: 2492 nodes malloced Process 13220: 2496 nodes malloced Process 13220: 2500 nodes malloced Process 13220: 2505 nodes malloced Process 13220: 2510 nodes malloced Process 13220: 2516 nodes malloced Process 13220: 2520 nodes malloced Process 13220: 2524 nodes malloced Process 13220: 2529 nodes malloced Process 13220: 2534 nodes malloced Process 13220: 2540 nodes malloced Process 13220: 2544 nodes malloced Process 13220: 2548 nodes malloced Process 13220: 2553 nodes malloced Process 13220: 2558 nodes malloced -- Karl Kraft Karl_Kraft@trapac.com Karl_Kraft@ensuing.com [My opinions are my own]
From: dicosmo@verveine.ens.fr (Roberto DiCosmo) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: HELP: vfs_mountroot fails after 3.3 upgrade! Date: 07 Oct 1995 13:48:14 GMT Organization: Ecole Normale Superieure Distribution: world Message-ID: <DICOSMO.95Oct7144814@verveine.ens.fr> I have just upgrade (over the network) my portable to 3.3, and after the upgrade (that needed to be restarted several times) and the following reboot, everything works fine in the 3.3 boot process *up to* the phase of mounting the /dev/hd0a device: there, the system panics with the error: vfs_mountroot: 19 (in /usr/include/bsd/sys/errno.h: NO SUCH DEVICE) I tried to boot using the 486 boot floppy from the archives (boot single user on a 3.2 bare single user system), and I can mount the hard disk, but I can do nothing (ls from the hard disk works, cat does not) interesting. On the other hand, booting with hd()mach_kernel rootdev=fd0a, seems to work (this means, boot from the hard disk, but mount root on the floppy), but then I *cannot* mount the hard disk! Is there a simple solution (changing some mode of / on the hard disk or something similar that has *not* been done by the upgrader) to recover the disk without re-installing the system? Any nice person at NeXT would like to lend a hand (like explaining the cryptic format of the Release3.3.uplan file)? Thank you! (btw, I am quite desperate :-( ) -- Roberto Di Cosmo <dicosmo@dmi.ens.fr>, http://www.ens.fr/users/dicosmo/index.html LIENS Ecole Normale Superieure 45, Rue d'Ulm 75005 Paris FRANCE
From: Mark Crispin <mrc@CAC.Washington.EDU> Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: Sat, 7 Oct 1995 16:37:26 -0700 Organization: Networks & Distributed Computing Message-ID: <Pine.NXT.3.92.951007162208.4004C-100000@Tomobiki-Cho.CAC.Washington.EDU> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFz55x.38q@AWT.NL> <Pine.NXT.3.92.951005132520.1474C-100000@Tomobiki-Cho.CAC.Washington.EDU> <DG3CKr.wq@trapac.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII NNTP-Posting-User: koma In-Reply-To: <DG3CKr.wq@trapac.com> On Sat, 7 Oct 1995, Karl Kraft wrote: > Using a clock that updates 1 per second, 1 node per second > is being allocated by Preferences. Through clever usage of gdb > (left as an exercise for the reader), you can see that the node is > an NSDate, created by the method +date of the NXDate class cluster. > +date is called by the -updateClock method of the IconClock class. > > This works out to 3.1 meg per day if your clock updates once per > second, or about 52K per day if your clock updates once per minute. This problem is new in 3.3; it does not happen in 3.0. I guess that some bright boy at NeXT decided to "improve" working code and screwed up. Probably wanted to use the latest and greatest fancy classes. It's curious how they have time to do stupid stuff like this, but never any time to fix bugs in the kernel and Unix utils. Of course, if Steve Jobs hadn't been such a jerk about system software source code, we could have fixed the bugs ourselves. Those of you who were around in 1989 will remember his reasons: 1) To force all NeXT systems to be bug-for-bug identical, so there would be a guarantee of lots of third-party software developers writing shrinkwrap software. SUN, according to Jobs, was on its way out because those evil customers with source code could modify their system software. 2) If customers had source code, they would clone it on their old SUN3's and not buy any more 68030 cubes. The software absolutely had to be locked to a proprietary architecture. These reasons seem ironic today. -- Mark -- DoD #0105, R90/6 pilot, FAX: (206) 685-4045 ICBM: N 47 39'35" W 122 18'39" Science does not emerge from voting, party politics, or public debate.
Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy From: mcbinc@world.std.com (Monty C Brandenberg) Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Message-ID: <DG3sG6.Mz3@world.std.com> Followup-To: comp.sys.next.bugs,comp.sys.next.advocacy Organization: MCB, Incorporated. Cambridge, MA References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFz55x.38q@AWT.NL> <Pine.NXT.3.92.951005132520.1474C-100000@Tomobiki-Cho.CAC.Washington.EDU> <DG3CKr.wq@trapac.com> <Pine.NXT.3.92.951007162208.4004C-100000@Tomobiki-Cho.CAC.Washington.EDU> Date: Sun, 8 Oct 1995 00:09:42 GMT Mark Crispin (mrc@CAC.Washington.EDU) wrote: : On Sat, 7 Oct 1995, Karl Kraft wrote: : > This works out to 3.1 meg per day if your clock updates once per : > second, or about 52K per day if your clock updates once per minute. : This problem is new in 3.3; it does not happen in 3.0. I guess that some : bright boy at NeXT decided to "improve" working code and screwed up. : Probably wanted to use the latest and greatest fancy classes. Agreed. Preferences is fine up to and including 3.2. : It's curious how they have time to do stupid stuff like this, but never : any time to fix bugs in the kernel and Unix utils. Not to mention the chutzpah of trying to charge $5+K for this and a compiler... :-( While we're in here, how about a pointer to 'leaks.' This is a tool I haven't seen before. Rather liked what I saw. -- Monty Brandenberg Software Consultant MCB, Inc. mcbinc@world.std.com 617.864.6907
From: felix.rauch@harka.limmat.net.ch (Felix Rauch) Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Followup-To: comp.sys.next.bugs,comp.sys.next.advocacy Date: 7 Oct 1995 18:01:50 GMT Organization: Private NEXTSTEP-site Distribution: world Message-ID: <456fae$9q@harka.limmat.net.ch> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> Strange, but on my system it seems to be the other way around. Shortly after booting and logging in, it looks like this (ps output): USER PID %CPU %MEM VSIZE RSIZE TT STAT TIME COMMAND xilef 215 0.0 5.0 5.94M 1.59M ? SW 0:01 /NextApps/Preferences.app/P After some hours: xilef 215 0.0 4.2 5.94M 1.34M ? SW 0:04 /NextApps/Preferences.app/P Before I shut the machine down again (uptime: 7:57): xilef 215 0.0 4.0 5.94M 1.29M ? SW 0:04 /NextApps/Preferences.app/P The RSIZE gets smaller over the time! I wonder whether I could reduze its memory-usage to 0 if I let the machine on forever ;-) This is on a black colorstation with NS3.3. - Felix -- Felix Rauch, CS-Student @ ETH Zurich, Switzerland. E-mail: felix@nice.ch NeXT/MIME-mail welcome. Member of NiCE - NeXT User Group Homepage: http://nice.ethz.ch/~felix (includes pgp public key)
From: gideon@csarc.otago.ac.nz (Gideon King) Newsgroups: comp.sys.next.bugs Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 8 Oct 1995 21:36:50 GMT Organization: University of Otago, Dunedin, NZ Distribution: world Message-ID: <459g9i$9kv@celebrian.otago.ac.nz> References: <450thj$hd@emerald.oz.net> Could someone please tell me what leaks does and where to get it? Thanks. --- Gideon King | Phone +64-3-479 8347 University of Otago | Fax +64-3-479 8529 Computer Science Applied | e-mail gideon@csarc.otago.ac.nz Research Centre | NeXT mail, MIME ok Department of Computer Science | P.O. Box 56 | There are 3 kinds of people: those who Dunedin | can count & those who can't. New Zealand | Do witches run spell checkers? WWW access: http://www.csarc.otago.ac.nz:805/PersonalHomePages/Gideon.htmld/
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 9 Oct 1995 04:15:09 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <45a7kd$9q2@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: Jeffrey T Eaton <jeaton+@andrew.cmu.edu> Newsgroups: comp.sys.next.bugs Subject: Termcap bug? Date: Mon, 9 Oct 1995 15:58:25 -0400 Organization: Junior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <wkSLvVG00iV2Q6z7cM@andrew.cmu.edu> I noticed a quite annoying problem when telnetting into my NS/Intel 3.3 box from a Sun Sparc 1+ running Solaris 2.3 or from a Sparc 1+ running SunOS 4.1.2. (I'm telnetted from within an xterm, TERM set to xterm, , using tcsh on both systems...) It seems that my NS box will occasionally drop the carrige return and only send a line feed. The problem seems to be rather random, as it happens at different times -- sometimes right away, other times partway through a listing, other times not at all. This happens in many different programs (ls, cat, et. al.) Any suggestions? Transcript attached below... zoo:~:ls -la total 383 drwxr-xr-x 15 jeaton 2048 Oct 5 19:49 ./ drwxr-xr-x 15 root 1024 Sep 22 01:16 ../ drwxr-xr-x 5 jeaton 1024 Apr 5 1995 .AppInfo/ -rw-r--r-- 1 jeaton 103 Sep 11 00:29 .AstScores drwxr-xr-x 6 jeaton 1024 Oct 9 02:43 .NeXT/ -rw-r--r-- 1 jeaton 145 Oct 4 21:09 .TYScores srwxrwxrwx 1 jeaton 0 Jun 27 00:46 .TeXview_Pipe= -rw-r--r-- 1 jeaton 2198 Feb 1 1995 .addressbook -rw-r--r-- 1 jeaton 240 Mar 4 1992 .commanddict -rw-r--r-- 1 jeaton 0 Apr 17 19:25 .dir.newsbase -rw-r--r-- 1 jeaton 192 Aug 16 03:58 .dir3_0.wmd -rw-r--r-- 1 jeaton 1490 Oct 3 14:34 .gatorftprc -rw------- 1 jeaton 609 Oct 9 15:43 .history -rw-r--r-- 1 jeaton 885 Jul 15 1991 .indent.pro -rw-r--r-- 1 jeaton 49152 May 10 22:04 .index.store -rw-r--r-- 1 jeaton 182 Oct 21 1994 .login -rw-r--r-- 1 jeaton 55 Oct 21 1994 .logout -rw-r--r-- 1 jeaton 19 Oct 21 1994 .mailrc -rw-r--r-- 1 jeaton 153563 Apr 17 19:25 .newsbaserc -rw-r--r-- 1 jeaton 126746 Sep 25 22:23 .newsrc -rw-r--r-- 1 jeaton 69 May 1 17:34 .pipedict -rw-r--r-- 1 jeaton 36 Apr 19 20:29 .plan -rw------- 1 jeaton 651 Aug 7 13:31 .rhosts -rw-r--r-- 1 jeaton 138 Sep 17 16:46 .signature -rw-r--r-- 1 jeaton 842 Aug 16 02:01 .tcshrc drwxr-xr-x 3 jeaton 1024 Sep 23 19:25 .www/ drwxr-xr-x 2 jeaton 1024 Sep 25 03:55 Apps/ drwxr-xr-x 2 jeaton 1024 Oct 5 19:50 Archive/ drwxr-xr-x 5 jeaton 1024 Sep 25 04:03 Classes/ drwxr-xr-x 6 jeaton 1024 Sep 25 23:23 Development/ drwxr-xr-x 6 jeaton 3072 Oct 5 19:13 Incoming/ drwxr-xr-x 23 jeaton 1024 May 31 03:15 Library/ drwxr-xr-x 16 jeaton 1024 Sep 25 12:38 Mailboxes/ drwxr-xr-x 2 jeaton 1024 Oct 4 17:07 Random/ drwxr-xr-x 2 jeaton 1024 Oct 5 19:20 Stuff/ drwxr-xr-x 5 jeaton 1024 Jul 28 00:44 ZooStuff/ -rw-r--r-- 1 jeaton 1128 Aug 7 11:50 ezmail.ini zoo:~: zoo:~:ls -la total 383 drwxr-xr-x 15 jeaton 2048 Oct 5 19:49 ./ drwxr-xr-x 15 root 1024 Sep 22 01:16 ../ drwxr-xr-x 5 jeaton 1024 Apr 5 1995 .AppInfo/ -rw-r--r-- 1 jeaton 103 Sep 11 00:29 .AstScores drwxr-xr-x 6 jeaton 1024 Oct 9 02:43 .NeXT/ -rw-r--r-- 1 jeaton 145 Oct 4 21:09 .TYScores srwxrwxrwx 1 jeaton 0 Jun 27 00:46 .TeXview_Pipe= -rw-r--r-- 1 jeaton 2198 Feb 1 1995 .addressbook -rw-r--r-- 1 jeaton 240 Mar 4 1992 .commanddict -rw-r--r-- 1 jeaton 0 Apr 17 19:25 .dir.newsbase -rw-r--r-- 1 jeaton 192 Aug 16 03:58 .dir3_0.wmd -rw-r--r-- 1 jeaton 1490 Oct 3 14:34 .gatorftprc -rw------- 1 jeaton 609 Oct 9 15:43 .history -rw-r--r-- 1 jeaton 885 Jul 15 1991 .indent.pro -rw-r--r-- 1 jeaton 49152 May 10 22:04 .index.store -rw-r--r-- 1 jeaton 182 Oct 21 1994 .login -rw-r--r-- 1 jeaton 55 Oct 21 1994 .logout -rw-r--r-- 1 jeaton 19 Oct 21 1994 .mailrc -rw-r--r-- 1 jeaton 153563 Apr 17 19:25 .newsbaserc -rw-r--r-- 1 jeaton 126746 Sep 25 22:23 .newsrc -rw-r--r-- 1 jeaton 69 May 1 17:34 .pipedict -rw-r--r-- 1 jeaton 36 Apr 19 20:29 .plan -rw------- 1 jeaton 651 Aug 7 13:31 .rhosts -rw-r--r-- 1 jeaton 138 Sep 17 16:46 .signature -rw-r--r-- 1 jeaton 842 Aug 16 02:01 .tcshrc drwxr-xr-x 3 jeaton 1024 Sep 23 19:25 .www/ drwxr-xr-x 2 jeaton 1024 Sep 25 03:55 Apps/ drwxr-xr-x 2 jeaton 1024 Oct 5 19:50 Archive/ drwxr-xr-x 5 jeaton 1024 Sep 25 04:03 Classes/ drwxr-xr-x 6 jeaton 1024 Sep 25 23:23 Development/ drwxr-xr-x 6 jeaton 3072 Oct 5 19:13 Incoming/ drwxr-xr-x 23 jeaton 1024 May 31 03:15 Library/ drwxr-xr-x 16 jeaton 1024 Sep 25 12:38 Mailboxes/ drwxr-xr-x 2 jeaton 1024 Oct 4 17:07 Random/ drwxr-xr-x 2 jeaton 1024 Oct 5 19:20 Stuff/ drwxr-xr-x 5 jeaton 1024 Jul 28 00:44 ZooStuff/ -rw-r--r-- 1 jeaton 1128 Aug 7 11:50 ezmail.ini zoo:~: Happened the first time, but not the second... I'm lost here... -jeaton --- Jeffrey T. Eaton jeaton@andrew.cmu.edu Junior, Computer Science/Information Systems at Carnegie Mellon University ---
Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs From: bfbaker@netcom.com (Technom Enterprises Inc) Subject: EOF: multiple database connections? Message-ID: <bfbakerDG7ApF.HL5@netcom.com> Organization: Operations Technologies Date: Mon, 9 Oct 1995 21:36:50 GMT Sender: bfbaker@netcom17.netcom.com Greetings, We're having some major difficulties with EOF. We can't seem to talk to more than one database at a time. In the following code example, I've tried to create two EODatabaseDataSources, connected to two different databases, and fetch objects from each. When I try each section of code separately, they work fine. But if I try to use both, the second one fails during the fetch, with the following console output: ** otto> EOFTest ** Got 2 objects from data source 1 ** Oct 9 15:09:38 [578] Sybase Server:unknown Procedure:none Line:1 ** Message:Invalid object name 'CustodianInfo'. ** No array returned from data source 2 ** otto> 'CustodianInfo' is a valid object, since access to data source 2 works perfectly by itself. Both of these databases live on a Sybase server. I have run the same experiment with two QuickBase databases, with the same results. We are running NS3.3, and EOF 1.1. Any suggestions on what we can do to fix this will be most appreciated. Many thanks in advance, Brian Baker. bbaker@technom.com The code: #import <stdio.h> #import <foundation/NSString.h> #import <foundation/NSArray.h> #import <eoaccess/EODatabaseDataSource.h> #import <eoaccess/EOQualifier.h> void main(int argc, char *argv[]) { EODatabaseDataSource *dataSource1, *dataSource2; EOQualifier *qualifier1, *qualifier2; NSArray *array1, *array2; /* access data source one */ if (dataSource1 = [[EODatabaseDataSource alloc] initWithModelName: @"UMFData" entityName: @"Locations"]) { qualifier1 = [[EOQualifier alloc] initWithEntity: [dataSource1 entity] qualifierFormat: @"%A = '%@'", @"name", @"Input"]; [dataSource1 setQualifier: qualifier1]; if (array1 = [dataSource1 fetchObjects]) { printf("Got %d objects from data source 1\n", [array1 count]); } else { fprintf(stderr, "No array returned from data source 1\n"); } } else { fprintf(stderr, "Couldn't create data source 1\n"); } /* access data source two */ if (dataSource2 = [[EODatabaseDataSource alloc] initWithModelName: @"UMFInfo" entityName: @"CustodianInfo"]) { qualifier2 = [[EOQualifier alloc] initWithEntity: [dataSource2 entity] qualifierFormat: @"%A = %@", @"maxEngines", @"1"]; [dataSource2 setQualifier: qualifier2]; if (array2 = [dataSource2 fetchObjects]) { printf("Got %d objects from data source 2\n", [array2 count]); } else { fprintf(stderr, "No array returned from data source 2\n"); } } else { fprintf(stderr, "Couldn't create data source 2\n"); } exit(0); }
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: Termcap bug? Date: 9 Oct 1995 23:40:03 GMT Organization: Information Technology Solutions, Inc. Message-ID: <45cbsj$hjg@news.its.com> References: <wkSLvVG00iV2Q6z7cM@andrew.cmu.edu> Jeffrey T Eaton <jeaton+@andrew.cmu.edu> wrote: > I noticed a quite annoying problem when telnetting into my NS/Intel 3.3 > box from a Sun Sparc 1+ running Solaris 2.3 or from a Sparc 1+ running > SunOS 4.1.2. > (I'm telnetted from within an xterm, TERM set to xterm, , using tcsh on > both systems...) Two suggestions: 1) Set the TERM/TERMCAP variables to "vt100" on the NeXT side of things. 2) "stty -extproc" -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: pisul_cj@cowley.uwlax.edu (Charles Pisula S92) Newsgroups: comp.sys.next.bugs Subject: Re: NeXT Dev Documentation Error Date: 10 Oct 1995 00:14:36 GMT Organization: University of Wisconsin - La Crosse Message-ID: <45cdtc$dii@alfred.acs.uwlax.edu> References: <453869$i1n@news.tuwien.ac.at> In article <453869$i1n@news.tuwien.ac.at> e9026354@fbma.tuwien.ac.at (Wolfgang Holzinger) writes: > Alvin Jee (alvin@cse.ucsc.edu) wrote: > : Hello! > > : Has anybody else noticed that the on-line docs for the Application class > : and the Action Cell class are virtually the same document? I even checked the > : Dev CD and they are the same. Sort of defeats the purpose of on-line docs :) > > > Glad to hear that. I was getting suspicious on my filesystem... > But where do we get the missing file from? > > Ciao Wolfgang I believe you can just swap in the correct files from 3.2 although they will of course not be the most current. If the real 3.3 version is out there, I would like to know too. -- Chuck Pisula ------------------------------------------------- Steve Jobs quote From UnixWorld, April 1993 "If we give people an alternative to Microsoft... it will have been a greater good." ------------------------------------------------- ***
From: J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) Newsgroups: comp.sys.next.bugs Subject: Re: Problems with TeXView [solved!] Followup-To: comp.sys.next.bugs Date: 10 Oct 1995 09:45:34 GMT Organization: Theory Group, Physics, QMW College Distribution: world Message-ID: <J.M.FIGUEROA.95Oct10104534@strings7.ph.qmw.ac.uk> References: <J.M.FIGUEROA.95Sep29165842@strings7.ph.qmw.ac.uk> <J.M.FIGUEROA.95Oct4001720@strings7.ph.qmw.ac.uk> <J.M.FIGUEROA.95Oct4232010@strings7.ph.qmw.ac.uk> In-reply-to: J.M.Figueroa@QMW.ac.uk's message of 04 Oct 1995 22:20:10 GMT Finally, thanks to a suggestion of Raf Schietekat, I found the problem. The culprit was my .dvipsrc file. I still don't understand why it is that it works on Motorola and not on Intel and why it is that dvips can find the *.pro files but TeXview couldn't. Maybe this ought to be fixed??? Anyhow, I'm happy this is over. Thanks again and sorry for the wasted bandwidth, if any. Cheers, Jose -- +----------------------------------+----------------------------------+ | Dr Jose M Figueroa-O'Farrill | Vox: +44.171.975-5055 | | Department of Physics (Rm. 227) | Fax: +44.181.981-7465 | | Queen Mary and Westfield College | mailto:J.M.Figueroa@qmw.ac.uk | | Mile End Road, London E1 4NS, UK | | | and | | http://stringswww.ph.qmw.ac.uk/~jmf/jmf.html | +----------------------------------+----------------------------------+
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <alex@conextions.com> Message-ID: <9510101523.AA11935@conextions.com> Content-Type: text/plain Mime-Version: 1.0 (NeXT Mail 3.3 v118.2) From: Alexander Elkin <alex@conextions.com> Date: Tue, 10 Oct 95 11:23:11 -0400 Subject: Re: EOF: multiple database connections? Cc: comp-sys-next-bugs@antigone.com >> We're having some major difficulties with EOF. We can't seem to talk to more than one database at a time. In the following code example, I've tried to create two EODatabaseDataSources, connected to two different databases, and fetch objects from each. When I try each section of code separately, they work fine. But if I try to use both, the second one fails during the fetch << Hi, Brian, You should have no problems with multiple databases if you use different names for the database and its contents. Just use EODatabaseDataSources' method: -initWithModelName:entityName:databaseName:contextName: instead of -initWithModelName:entityName:. Alex Elkin Conextions Inc. channelName:
From: robin@pencom.com Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 10 Oct 1995 14:24:18 GMT Organization: Pencom Software Message-ID: <45dvmi$fta@digdug.pencom.com> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <456fae$9q@harka.limmat.net.ch> felix.rauch@harka.limmat.net.ch (Felix Rauch) wrote: : Strange, but on my system it seems to be the other way around. Shortly : after booting and logging in, it looks like this (ps output): : USER PID %CPU %MEM VSIZE RSIZE TT STAT TIME COMMAND : xilef 215 0.0 5.0 5.94M 1.59M ? SW 0:01 /NextApps/Preferences.app/P : After some hours: : xilef 215 0.0 4.2 5.94M 1.34M ? SW 0:04 /NextApps/Preferences.app/P : Before I shut the machine down again (uptime: 7:57): : xilef 215 0.0 4.0 5.94M 1.29M ? SW 0:04 /NextApps/Preferences.app/P : The RSIZE gets smaller over the time! I wonder whether I could reduze : its memory-usage to 0 if I let the machine on forever ;-) : This is on a black colorstation with NS3.3. The RSIZE will eventually drop to 'just enough' to run the clock (so long as you don't open the app and play with the preferences...). On my NeXT Turbo (mono), that is 1024Kbytes. The reason you are seeing the RSIZE value drop is because portions of the app are being paged out over time. If you started a _huge_ app, it would probably page _all_ of the Preferences app out -- with the exception of the clock. The real interesting number is the VSIZE. Why is this different on different machines? Mine is 6.15MB (not enough to get worked up about -- but still ~160Kb different than yours)... -- ----------------------------------------------------------------------------- *** These are my opinions... Mine! All Mine! Minemineminemineminemine! *** ----------------------------------------------------------------------------- Robin D. Wilson robin@pencom.com Pencom Software 701 Canyon Bend Dr. 9050 Capital of Texas Hwy Pflugerville, TX 78660 Austin, TX 78759 (512) 251-1737 (512) 343-6666
From: robin@pencom.com Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 10 Oct 1995 14:17:36 GMT Organization: Pencom Software Message-ID: <45dva0$fta@digdug.pencom.com> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFyJLB.BHL@world.std.com> mcbinc@world.std.com (Monty C Brandenberg) wrote: : Mark Crispin (mrc@CAC.Washington.EDU) wrote: : : I finally managed to catch this bug during a period when nothing else : : could be blamed; specifically, a freshly-rebooted system which was then : : left alone for a week. : : It's a memory leak in Preferences.app. It had gobbled down over 60MB of : : virtual memory and was still going strong. : Not to detract from your effort but I reported that one in c.s.n.bugs : at least six months ago looking for verification from other users. : Not a bloody peep from *anyone*. Does anyone actually use the damned : software? : Best workaround: kill Preferences from the command line. Do *not* : unhide it and quit. That just thrashes the VM. I'd suggest a crontab : entry to kill and restart a new Preferences. : I'd really like to kick someone in the head for this one... How long does it take to notice this? I've been running this system for 4 days, and have been logged in all of that time. Yet Preferences.app does seem to be leaking for me: ~> uptime 9:09am up 4 days, 19:21, 1 user, load average: 0.70, 0.33, 0.28 ~> proc Pref USER PID %CPU %MEM VSIZE RSIZE TT STAT TIME COMMAND robin 226 0.0 3.1 6.15M 1024K ? SW 0:41 /NextApps/Preferences.app/Preferences -NXAutoLaunch YES -MachLaunch 2 31 ('proc' is just a script that does a 'ps' command and greps out stuff...) See that my 'VSIZE' == 6.15M, while the RSIZE is only 1024K. Also, the following suggests that there are _0_ pagein's as a result of this process... ~> ps -vaxww | egrep "PAG|Pref" | grep -v "grep" PID TT STAT TIME SL RE PAGEIN VSIZE RSIZE LIM TSIZ TRS %CPU %MEM COMMAND 226 ? SW 0:41 0 0 0 6.15M 1024K 0 0K 0 0.0 3.1 Preferences Which would indicate that this is _not_ a memory leak for the swapfile (you'd think that it would have to pagein if it were going to use some of the swapfile...). So, what's the definitive answer? -- ----------------------------------------------------------------------------- *** These are my opinions... Mine! All Mine! Minemineminemineminemine! *** ----------------------------------------------------------------------------- Robin D. Wilson robin@pencom.com Pencom Software 701 Canyon Bend Dr. 9050 Capital of Texas Hwy Pflugerville, TX 78660 Austin, TX 78759 (512) 251-1737 (512) 343-6666
From: robin@pencom.com Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 10 Oct 1995 19:25:02 GMT Organization: Pencom Software Message-ID: <45ehae$gf8@digdug.pencom.com> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFz55x.38q@AWT.NL> <Pine.NXT.3.92.951005132520.1474C-100000@Tomobiki-Cho.CAC.Washington.EDU> <DG3CKr.wq@trapac.com> karl@trapac.com (Karl Kraft) wrote: : Using a clock that updates 1 per second, 1 node per second : is being allocated by Preferences. Through clever usage of gdb : (left as an exercise for the reader), you can see that the node is : an NSDate, created by the method +date of the NXDate class cluster. : +date is called by the -updateClock method of the IconClock class. So, is +date a dynamically loaded method, is it coming from the shared library -- therefore the fix for _all_ associated leaks would be to ship a new library? When can I get mine? (I wonder if this problem is freaking out other programs too?) --- robin -- "Out the 10Base-T, through the router, down the T1, over the leased line, off the bridge, past the firewall...nothing but Net." -- ----------------------------------------------------------------------------- *** These are my opinions... Mine! All Mine! Minemineminemineminemine! *** ----------------------------------------------------------------------------- Robin D. Wilson robin@pencom.com Pencom Software 701 Canyon Bend Dr. 9050 Capital of Texas Hwy Pflugerville, TX 78660 Austin, TX 78759 (512) 251-1737 (512) 343-6666
From: srosie@umich.edu (Steven S. Rosenblum) Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 10 Oct 1995 13:26:53 GMT Organization: University of Michigan Message-ID: <45dsat$a8i@lastactionhero.rs.itd.umich.edu> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <456fae$9q@harka.limmat.net.ch> In article <456fae$9q@harka.limmat.net.ch>, Felix Rauch <felix.rauch@nice.ch> wrote: >[ps output showing that Preferences.app uses less memory over time chomped] Okay, I'm officially confused.... I've read in this forum that ps is not as reliable at measuring memory as vm_stat, but Felix's results don't seem to make sense. Does anyone know precisely how ps comes up with its statistics? thanks in advance, rosie
From: felix.rauch@harka.limmat.net.ch (Felix Rauch) Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Followup-To: comp.sys.next.bugs,comp.sys.next.advocacy Date: 10 Oct 1995 18:43:59 GMT Organization: Private NEXTSTEP-site Distribution: world Message-ID: <45eetf$fi@harka.limmat.net.ch> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <456fae$9q@harka.limmat.net.ch> <45dvmi$fta@digdug.pencom.com> robin@pencom.com wrote: [...] > The real interesting number is the VSIZE. Why is this > different on different machines? Mine is 6.15MB (not enough to get > worked up about -- but still ~160Kb different than yours)... Thanks for thouse explanations, I had a missunderstanding about RSIZE/VSIZE. The different VSIZE-values could be because of the different clocks. I use the one with the digital digits on top (without seconds) an the calendar on the bottom. - Felix -- Felix Rauch, CS-Student @ ETH Zurich, Switzerland. E-mail: felix@nice.ch NeXT/MIME-mail welcome. Member of NiCE - NeXT User Group Homepage: http://nice.ethz.ch/~felix (includes pgp public key)
From: jbn@mystery-train.cu-online.com (J.B. Nicholson-Owens) Newsgroups: comp.sys.next.bugs Subject: Re: Termcap bug? Date: 11 Oct 1995 01:46:40 GMT Organization: Organizing Organisms & Organs Message-ID: <45f7m0$nf6@ill.msilink.com> References: <wkSLvVG00iV2Q6z7cM@andrew.cmu.edu> <45cbsj$hjg@news.its.com> Chuck Swiger <chuck@its.com> wrote: > 1) Set the TERM/TERMCAP variables to "vt100" on the NeXT side of things. I suggest setting the TERM (not TERMCAP) environment variable to 'xterm' since not all xterms are 100% compatible with vt100s (and thus might cause software to break). If you don't have an xterm entry, give vt100 a try, but try finding an xterm entry (from any termcap library). I recommend against sticking "xterm" or "vt100" in the TERMCAP variable because that variable is meant for pointing to another termcap library, either a definition stored in the variable or the path of a file where a termcap library can be found. Sticking just "xterm" or "vt100" in TERMCAP would be erroneous. Recommended sources: (1) "termcap & terminfo" Strang/Mui/O'Reilly ISBN #0-937175-22-6 (2) the Usenet newsgroup comp.terminals and the references provided therein
From: fischer-michael@cs.yale.edu (Michael Fischer) Newsgroups: comp.sys.next.bugs Subject: Re: Problems with TeXView [NS 3.2 on intel] Date: 12 Oct 1995 18:39:50 GMT Organization: Yale University Computer Science Dept., New Haven, CT 06520-8285 Distribution: world Message-ID: <45jndm$l0@babyblue.cs.yale.edu> References: <J.M.FIGUEROA.95Sep29165842@strings7.ph.qmw.ac.uk> <J.M.FIGUEROA.95Oct4232010@strings7.ph.qmw.ac.uk> Jose Miguel Figueroa-O'Farrill (J.M.Figueroa@QMW.ac.uk) wrote: : This is the third post in the series. I have now identified the : problem I've been having with TeXview. TeXview is simply not reading : the prologue file /usr/lib/tex/ps/special.pro. If I happen to place a : copy of special.pro in the directory with the .dvi file, then things : work smoothly. Otherwise I get the errors I listed in my previous : post. : Does anyone know the correct environment variable that TeXview looks : up to find the prologue directory? Yes. Try setting TEXCONFIG=/usr/lib/tex/ps -- ================================================== | Michael Fischer <fischer-michael@cs.yale.edu> | ==================================================
Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy From: tom@basil.icce.rug.nl (Tom Hageman) Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Message-ID: <DGB170.8ur@basil.icce.rug.nl> Originator: uubasil@obelix.icce.rug.nl Sender: news@basil.icce.rug.nl (NEWS pusher) Organization: Warty Wolfs References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFz55x.38q@AWT.NL> <Pine.NXT.3.92.951005132520.1474C-100000@Tomobiki-Cho.CAC.Washington.EDU> <DG3CKr.wq@trapac.com> <45ehae$gf8@digdug.pencom.com> Date: Wed, 11 Oct 1995 22:01:48 GMT robin@pencom.com wrote: > karl@trapac.com (Karl Kraft) wrote: > : Using a clock that updates 1 per second, 1 node per second > : is being allocated by Preferences. Through clever usage of gdb > : (left as an exercise for the reader), you can see that the node is > : an NSDate, created by the method +date of the NXDate class cluster. > : +date is called by the -updateClock method of the IconClock class. > So, is +date a dynamically loaded method, is it coming from the shared library > -- therefore the fix for _all_ associated leaks would be to ship a new library? Nope. +[NSDate date] is a FoundationKit method, which is _supposed_ to create a new (autoreleased) instance of NSDate each time it is invoked. From Karl's description it looks like somebody at NeXT forgot to change Preferences' main.m to use EOApplication (an Application subclass which cleans up autoreleased objects in its event loop) instead of Application. BTW does anyone know if this is fixed in the much-rumoured NS3.3 patch? -- __/__/__/__/ Tom Hageman <tom@basil.icce.rug.nl> [NeXTmail/Mime OK] __/ __/_/ IC Group <tom@icgned.nl> (work) __/__/__/ "...to baldly go where no one has gone before." __/ _/_/ -- star trek TNG
From: rputtkam@iiic.ethz.ch (Roman Puttkammer) Newsgroups: comp.sys.next.bugs Subject: tip crashes NeXT Date: 13 Oct 1995 10:42:28 GMT Organization: Dept. Informatik, Swiss Federal Institute of Technology Message-ID: <45lfqk$8rn@neptune.ethz.ch> hello I have at last managed to use tip (Intel, 3.3). But there happen a few things, which worry me. When I use the tip which comes with the NeXT CD, making a connection works fine. After entering the loginname, nothings happens anymore. I know, that the remote machine prompts for a password ("Password:"). As workaround, I have compiled another tip with integrated x/y/zmodem (tipx). This works fine. Except: trying to receive bigger files is impossible (I use /dev/cufa and have the modem configured for hardware handshake). AND!!! when dropping the connection with '~.', and typing ctrl-c or ctrl-z without waiting for tip to exit, THE WHOLE SYSTEM STANDS STILL !@#$$%! I don't know, if the system has crashed or not. But the shell does not react anymore to any input and I cannot use the mouse anymore (no pointer). The HOME-key, which is configured to prompt for 'turn of computer?', does neither show any reactions... (reset, reboot, fsck, fsck, fsck, fsck.... :( Does anybody know, what this is? Some kind of race condition bug??? I know, Intel-based computer are some sort of crap, but I thought NeXT to run stable on them!?!?! Please tell me, which terminal program to use. -- ----------------------------------------------------------------------------- Roman Puttkammer, CS-Student @ Swiss Federal Institute of Technology, Zuerich E-Mail: rputtkam@iiic.ethz.ch
From: http@cs.nemostate.edu Newsgroups: comp.sys.next.hardware,comp.sys.next.bugs,comp.sys.next.software,comp.sys.next.misc Subject: Mach 64 Video Driver and IBM 16/4 ISA Token Ring Driver conflict resolution?? Date: 13 Oct 1995 20:07:22 GMT Organization: Northeast Missouri State University Message-ID: <45mgtq$18oa@noc.nemostate.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit We need help resolving this conflict: Running NeXTSTEP 3.3 on a Gateway 2000 P5-90 we've encountered a Rom addressing conflict (a similar conflict is documented by Solaris 2.4 with their Token Ring driver) with the NeXT Token Ring driver (beta version 3.31) and the Mach 64 Video driver (2 Meg). A) We have been able to run the Token Ring driver with an IBM Auto 16/4 ISA Token Ring card 92G3774 (?) using the default VGA video driver. B)We've also been able to configure the token ring card with the s series video driver on an IBM 330 pentium. C)Solaris reports a conflict with the Token Ring plug and play card wanting Rom address CC000 which is already used by some VGA cards (in particular the Mach 64 card). We have as yet been unsuccessfull in 1) forcing the plug and play card to take Rom address D2000 and Ram D4000 as recommended in the similar Solaris 2.4 x86 token ring card faqs from Sun. 2) Moving around the video memory (all 2+ megs of it!) to avoid CC000. Any suggestions? No, we can't use either net instead in the near future!:-)
From: bludvigs@ulrik.uio.no (B|rre Ludvigsen) Newsgroups: comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Can't log in as root Date: 14 Oct 1995 09:52:31 GMT Organization: University of Oslo, Norway Message-ID: <45o18v$gcl@hermod.uio.no> Or rather when I've logged in the workspace immediatly quits, declaring "workspace manager error". How in the world to I get around that one? (Barre Ludvigsen - please leave this line in reply for my news reading agent) - Barre
From: root@I_should_put_my_domain_in_etc_NNTP_INEWS_DOMAIN (Eren Kotan) Newsgroups: comp.sys.next.bugs Subject: AArgH! 3.3 window server is eating all my CPU time Date: 14 Oct 1995 11:20:32 GMT Organization: . Message-ID: <45o6e0$gsu@bright.ecs.soton.ac.uk> Hi, after recently "upgrading" to 3.3 from 3.2 I seem to notice that the computer (A Cube with an ND board) sometimes slows to an almost standstill. If I do a ps -aux from a shell window I see that the -console (Window Server) process is using about 45% of CPU time. Is this a bug? This did not use to happen with 3.2. Any ideas? Thanks for your help. Eren Kotan NeXTmail root@darkside.aladdin.co.uk
From: mjantti@aton.abo.fi (Markus J{ntti STA) Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.misc Subject: CD:s generate "unexpected kernel page fault" Followup-To: comp.sys.next.bugs Date: 14 Oct 1995 12:47:57 GMT Organization: Abo Akademi University Distribution: world Message-ID: <MJANTTI.95Oct14144757@aton.abo.fi> When I try to read a CD other than the NeXT User and Developer my system crashes, i.e., it throws plenty of garbage in a small monitor on the screen and will only reboot from the large monitor. I have a non-Turbo NeXTstaion running NS 3.2 and SONY CD-ROM CDU-55S Rev 1.0t. The error message looks something like this unexpected kernel page fault failure trap: type ox410 f fcode rw 3 faultdel 0x0 trap: pc 0x40021e2 sp 0x3ffff0 sr 0x2000 trap: cpu 0 th 0x1017556e0 proc 0x1017ae40 pid 1460 pcb 0x1118d010 traceback: fp 0x1118db48 called from pc 0x040961b6 fp +x1118db5c 4-args 1119a27c 00000000 1118db94 and so on This happens with all kinds of CD:s, the latest one is the Lighthouse Academic CD. I can't imagine the problem is with the CD:s. Any suggestions? -- Markus Jantti | Department of Economics markus.jantti@abo.fi | Abo Akademi University http://www.abo.fi/~mjantti | FIN 20500 Turku, Finland 358-0-643 747 (Home/Voice) | 358-21-2654 161 (Office/Voice) 358-0-3489 233 (Home/Fax) | 358-21-2654 677 (Office/Fax)
From: berezaw@river.it.gvsu.edu (Bill Bereza) Newsgroups: comp.sys.next.bugs Subject: Re: tip crashes NeXT Date: 14 Oct 1995 19:43:50 GMT Organization: My Bedroom, Ravenna, MI Message-ID: <45p3tm$98i@news.it.gvsu.edu> References: <45lfqk$8rn@neptune.ethz.ch> In article <45lfqk$8rn@neptune.ethz.ch>, Roman Puttkammer <rputtkam@iiic.ethz.ch> wrote: >Does anybody know, what this is? Some kind of race condition bug??? > >I know, Intel-based computer are some sort of crap, but I thought NeXT >to run stable on them!?!?! Get the latest serial drivers from the Next web site. I used to have this same problem, but it's gone away since I upgraded to the 3.33 drivers. -- Bill Bereza berezaw@river.it.gvsu.edu <NeXT/MIME> Beware of all enterprises that require new clothes.
From: Patrick Stein <jolly@cis.uni-muenchen.de> Newsgroups: comp.os.mach,comp.sys.next.bugs,comp.sys.next.programmer Subject: Q: Bug ? exec() and multithreading weirdness Date: 16 Oct 1995 10:13:12 GMT Organization: Institut fuer Informatik der Universitaet Muenchen Distribution: world Message-ID: <45tb7o$9ng@arcadia.informatik.uni-muenchen.de> Keywords: mach, thread, multithreaded, systemcall, foundationkit Hiho ! I have a program using multiple threads. At one time I want to quit the program and start another one using the UNIX exec(2) ( in this case execle ) system call. At that point I have all other remaining threads waiting in a : thread_switch(THREAD_NULL,SWITCH_OPTION_WAIT,sometime); mach call. The exec just works fine, but the new program context has still the threads waiting and whenever one of them awakes the program exits ( without getting a unix signal(3C) ). So I finally found out about that behaviour and killed the other threads in prior to the exec call. My question now is, if that's the right behaviour or shouldn't a new program begin with just one thread, cause it doesn't make any sense to me to keep them running. Btw. I killed the threads using the following hack ( knowing that running threads don't fork ): { int i; thread_array_t thread_list; unsigned int thread_count; task_threads(task_self(),&thread_list,&thread_count); for(i=0;i<thread_count;i++) { if(thread_list[i]!=thread_self()) { thread_suspend(thread_list[i]); thread_abort(thread_list[i]); thread_terminate(thread_list[i]); } } } what happens when a thread exit's before it get's aborted ? Do I just get a !kern_success ? Thanx for you time. System used: NeXTSTEP 3.3, Foundation Kit -- gimme a smile :) - jolly =================================================================== Jolly alias Patrick Stein jolly@cis.uni-muenchen.de office : ++49 +89 -211 06 70 Centrum fuer Informations private: ++49 +89 -950 57 34 und Sprachverarbeitung der Ludwig-Maximilians Universitaet Muenchen =================================================================== " Hickory, dickory, dock, The mouse ran up the clock. The clock struck one, The mouse ran down, Hickory, dickory, dock. " - nursery rhyme ===================================================================
From: ayis@esg.com Newsgroups: comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Can't log in as root Date: 16 Oct 1995 09:15:50 GMT Organization: University of Delaware Message-ID: <45t7s6$mg4@news.udel.edu> References: <45o18v$gcl@hermod.uio.no> bludvigs@ulrik.uio.no (B|rre Ludvigsen) wrote: > Or rather when I've logged in the workspace immediately quits, declaring > "workspace manager error". How in the world to I get around that one? > (Barre Ludvigsen - please leave this line in reply for my news reading agent) > - Barre This may be a sign that your configuration files in .NeXT are corrupt. (I am assuming that this error is not a result of the WorkSpace Manager being corrupt.) To set the files in .NeXT to a pristine state do the following: 1) At the log in window type in console, then log in as root (or just su). 2) Backup the /.NeXT directory then copy the files from the default template in /usr/template/user/.NeXT to / 3) Try logging in as root. I hope this helps. -Ayis //////////////////////////////////////////////////////////////////////////// Ayis Theseas Pyrros E-Mail: ayis@esg.com NeXT Mail and MIME Welcomed \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
From: mpaque@aol.com (Mpaque) Newsgroups: comp.sys.next.bugs Subject: Re: AArgH! 3.3 window server is eating all my CPU time Date: 16 Oct 1995 06:13:10 -0400 Organization: America Online, Inc. (1-800-827-6364) Sender: root@newsbf02.news.aol.com Message-ID: <45tb7m$7rh@newsbf02.news.aol.com> References: <45o6e0$gsu@bright.ecs.soton.ac.uk> In article <45o6e0$gsu@bright.ecs.soton.ac.uk>, root@I_should_put_my_domain_in_etc_NNTP_INEWS_DOMAIN (Eren Kotan) writes: >Hi, after recently "upgrading" to 3.3 from 3.2 I seem to notice that the >computer (A Cube with an ND board) sometimes slows to an almost standstill. >If >I do a ps -aux from a shell window I see that the -console (Window Server) >process is using about 45% of CPU time. Is this a bug? This did not use to >happen with 3.2. The odds are very goods that some application is stuck in a loop, pumping PostScript at the Window Server. After your system has been in this state for several minutes, from a remote login (to avoid getting a false reading from Terminal.app) do a 'ps aux'. See which applications are running with a fairly high CPU usage. One of these is likely to be the culprit. Quit or kill these apps one at a time until the system load drops. I'd be interested in hearing what you find the culprit to be. Mike Paquette
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <flexus!rfschtkt@maze.ruca.ua.ac.be> Date: Sun, 15 Oct 95 20:03:38 +0100 From: Raf Schietekat <flexus!rfschtkt@maze.ruca.ua.ac.be> Message-ID: <9510151903.AA01913@flexus> Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! OK? Anyone have a log number for Bug_NeXT? /NextApps/Preferences.app Preferences.app itself >>>>> KBNS.33.1.037_c3.3o NSDate instance leaked for each clock update (best guess for this problem) Description: NSDate is a new OpenStep class, which is instantiated in autoreleased state. But Preferences.app was apparently not equipped with the machinery to autorelease things (an NSAutoreleasePool that is emptied on every iteration of the event loop), so now those NSDate instances accumulate as garbage. After a while, the application can grow a lot (at one NSDate per second.), making the swapfile grow (e.g., 3MB/day according to Mark Crispin). How to verify: See c.s.n.bugs conversation. Urgency: Does no damage (other than a total system crash if a login session is left open for a very long time), is not immediately obvious, workaround available. Confirmed: See c.s.n.bugs conversation (this is a summary by Raf_Schietekat). Workaround: From time to time, quit and relaunch Preferences.app (how to do this automatically?). (Note that the swapfile will probably not shrink back, but it will be a while before it starts growing again.) Cure (for NeXT): Install the autorelease machinery. Report history: c.s.n.bugs 1995-10-(04-11) ``Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!!'', started by Mark Crispin, with many participants. Contains some very strange statements, and something about a utility called ``leaks''. (And according to himself, an isolated message by Monty Brandenberg at least 6 months earlier.) <<<<< Latest publication of KBNS ``Known Bugs in NeXT Software'': ftp://ftp.cs.orst.edu/software/NeXT/documents/KBNS.33.0.rtf Raf Schietekat, RfSchtkt@maze.ruca.ua.ac.be (NeXTmail), Flanders, Belgium If I don't answer: my mail relay can't handle !, % or .uucp, I think *** The year 2000 will be the last year of the 20th century. ***
From: Alex Blakemore <alex@genoa.com> Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 13 Oct 1995 04:39:25 GMT Organization: Genoa Software Systems Message-ID: <45kqht$9ud@saturn.genoa.com> References: <Pine.NXT.3.92.951004143919.583A-100000@Tomobiki-Cho.CAC.Washington.EDU> <DFz55x.38q@AWT.NL> <Pine.NXT.3.92.951005132520.1474C-100000@Tomobiki-Cho.CAC.Washington.EDU> <DG3CKr.wq@trapac.com> <Pine.NXT.3.92.951007162208.4004C-100000@Tomobiki-Cho.CAC.Washington.EDU> <DG3sG6.Mz3@world.std.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit In <DG3sG6.Mz3@world.std.com> Monty C Brandenberg wrote: > Not to mention the chutzpah of trying to charge $5+K for this and a > compiler... :-( Don't forget the improved Edit.app interface to gdb also. And remember, the compiler (and debugger) is FREE In fairness, NeXT had been pretty reasonable about UPGRADE pricing UNTIL NS3.3, at which point they totally lost it ($700 for almost no improvements to NS3.3 Developer, and some bad regressions). No comment about non-upgrade pricing though. -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: ccwf@thales.klab.caltech.edu (Charles Fu) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: ... messed up my incoming mail (was Re: New 3.3 patch messed up my outgoing mail...) Date: 17 Oct 1995 09:40:39 GMT Organization: California Institute of Technology, Pasadena Message-ID: <45vtmn$1cf@gap.cco.caltech.edu> References: <eg2gwil4o.fsf@steffi.accessone.com> In article <eg2gwil4o.fsf@steffi.accessone.com>, Robert Nicholson <robert@steffi.accessone.com> wrote: > [useful stuff about how to explicitly flush the queue for outgoing mail] Actually, I'm writing about the related problem of handling incoming mail when the PPP/SLIP connection is not up. As far as I can determine, with all NeXT sendmails since 3.2 (including 3.3 and the 3.3 patch), NeXT attempts to use the resolver library to canonicalize hostnames on incoming mail and punts without delivering if unable to do so (this is normally the situation when dial-up systems are not connected). This is not normally a problem if all non-LAN mails comes in via PPP/SLIP (because the connection is up when delivering the mail, so hostname canonicalization can occur), but is a problem if you (also) get mail in other ways (e.g., UUCP, BITnet, ...). If your PPP/SLIP is configured to do dial-on-demand, what happens is that after getting some mail when the PPP/SLIP connection is not active, your modem will pickup and initiate a PPP/SLIP session and your host exchanges a few packets with the remote nameserver to resolve the senders' hostnames in your incoming mail. (In a non 3.3-patched system, there is a lookupd timeout between receipt of mail by sendmail and initiation of the PPP/SLIP session). This behavior can be especially annoying if you receive a lot of UUCP mail but have to pay for connect time on a PPP/SLIP account. When PPP/SLIP is not configured to do dial-on-demand, mail will sit, queued and locked, until the next time sendmail decides to process the queue while the PPP/SLIP connection is active or the next time the mailserver is rebooted. (The reason a reboot works is that the mail queue is normally processed during system start-up prior to configuration of PPP/SLIP.) Explicitly processing the queue while the connection is down will not help since canonicalization is not possible. Berkeley recognized this particular problem after adding the hostname canonicalization feature and created the `I' option to work around it. (There is a small write-up in the BSD4.4 documentation.) Unfortunately, the 3.2-3.3patch versions of NeXT's sendmail include canonicalization but not the workaround. For the curious, the revised default Berkeley behavior is to go ahead and deliver mail if canonicalization fails. Using the `I' option forces delivery to block until canonicalization succeeds, which is the current NeXT behavior. There are additional variations for advanced sys admins who want finer control over sendmail's use of the resolver library. This problem was not present in the 3.1 sendmail and the 3.1-3.2 patched sendmail available from NeXTanswers. The problems are the same with both Intel and Motorola architectures (unsurprisingly). Note that there have been many versions of sendmail shipped by NeXT since 3.1, but NeXT only occasionally increases the sendmail version number. For example, the 3.1-3.2patch version (no canonicalization) and the 3.3 unpatched version (always canonicalize) both are NeXT version NX5.67e (obviously, the binaries _are_ different). The only good workaround appears to be to compile and use the current Berkeley sendmail daemon instead of NeXT's sendmail daemon. Using the NeXT 3.1-3.2 sendmail patch will work if you are not concerned about the latest security hole. Other than that, it might be possible to get around the problem by periodically killing lookupd, explicitly flushing the mail queue, and restarting lookupd. However, I _really_ do not recommend anyone doing this for many reasons. Finally, keeping information in NetInfo for hosts that send you mail often will reduce the frequency of the problem (because then resolution can be handled locally). Can anyone think of other solutions? -ccwf
From: e9026293 <e9026293@stud1.tuwien.ac.at> Newsgroups: comp.sys.next.bugs Subject: (no subject) Date: 17 Oct 1995 10:46:46 GMT Organization: EDV-Zentrum, TU-Wien Message-ID: <4601im$8ks@news.tuwien.ac.at> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi to everyone I'm running NEXTSTEP 3.3 on an Intel machine under a 486 processor. My problem is, that the system is absolute stable, if I disable the secondary cache in the Setup. Otherwise, some time during working, an error message appears and the system fails. Does anyone have an idea, where the problem may be located. Thank you for your advice Wolfgang
From: e9026293 <e9026293@stud1.tuwien.ac.at> Newsgroups: comp.sys.next.bugs Subject: (no subject) Date: 17 Oct 1995 10:49:50 GMT Organization: EDV-Zentrum, TU-Wien Message-ID: <4601oe$8ks@news.tuwien.ac.at> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi everybody I'v got a problem with my computer. I'm running NS 3.3 on a 486 Intel machine and if I enable the secondary cache in the Setup, my machine is really instable. Does anybody know about this problem or have some advice? Thank you Wolfgang
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: (no subject) Date: 17 Oct 1995 16:29:14 GMT Organization: Information Technology Solutions, Inc. Message-ID: <460lkq$a0r@news.its.com> References: <4601oe$8ks@news.tuwien.ac.at> e9026293 <e9026293@stud1.tuwien.ac.at> wrote: > Hi everybody > I'v got a problem with my computer. I'm running NS 3.3 on a 486 Intel > machine and if I enable the secondary cache in the Setup, my machine is > really instable. Does anybody know about this problem or have some > advice? Well, you may have the cache burst timing wrong, or you may have problems with the SRAM chips that comprise the L2 cache. Try reseating the cache SRAM chips, and examine the bios configuration for cache timing and set it to the default or to the slowest value if the default isn't right. Make sure that you have the motherboard configured correctly for the clock speed. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: (no subject) Date: 17 Oct 1995 20:56:43 GMT Organization: Technical University of Berlin, Germany Message-ID: <4615ab$4qk@news.cs.tu-berlin.de> References: <4601im$8ks@news.tuwien.ac.at> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit e9026293 (e9026293@stud1.tuwien.ac.at) wrote: : Hi to everyone : I'm running NEXTSTEP 3.3 on an Intel machine under a 486 processor. : My problem is, that the system is absolute stable, if I disable the : secondary cache in the Setup. Otherwise, some time during working, an : error message appears and the system fails. Does anyone have an idea, : where the problem may be located. Your motherboard's chipset is buggy. Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: rdl@world.std.com (Robert La Ferla) Subject: Re: ... messed up my incoming mail (was Re: New 3.3 patch messed up my outgoing mail...) In-Reply-To: ccwf@thales.klab.caltech.edu's message of 17 Oct 1995 09:40:39 GMT Message-ID: <RDL.95Oct17224627@world.std.com> Sender: rdl@world.std.com (Robert La Ferla) Organization: The World Public Access UNIX, Brookline, MA References: <eg2gwil4o.fsf@steffi.accessone.com> <45vtmn$1cf@gap.cco.caltech.edu> Date: Wed, 18 Oct 1995 02:46:27 GMT I placed sendmail 8.7.1 quad-fat compiled for NS on ftp.cs.orst.edu. Robert La Ferla Registered OPENSTEP / NEXTSTEP Consultant Boston, MA + 1 (617) 252-0088
From: support@radical.com (Radical) Newsgroups: comp.sys.next.bugs Subject: RadicalNews 0.6 Beta Known Bugs Date: 12 Oct 1995 19:28:19 GMT Organization: Radical System Solutions, Inc. Message-ID: <45jq8j$q4@radical2.radical.com> Thank you to everyone participating in the open beta test of RadicalNews. We appreciate the time and effort you have taken to try the application and to respond. So far, we know about the following bugs: 1) Sometimes RadicalNews installs but does not run at all. The workaround is to do a chmod -R a+r+X RadicalNews.app 2) Preferences don't seem to work properly after turning sound OFF. The workaround is to leave the sound ON in Preferences. 3) Sometimes RadicalNews crashes when reading a certain thread or article. No workaround. We've found a bogus header field value that kills us. 4) Sometimes RadicalNews is a "big, fat, slow hog!" No workaround. We haven't been able to repeat this one yet. As soon as the fixes to these problems are available, a new version of RadicalNews will be submitted to ftp.cs.orst.edu.
From: klar4000@fatmom.zrz.tu-berlin.de (Klaus Rappich) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: ... messed up my incoming mail (was Re: New 3.3 patch messed up my outgoing mail...) Date: 18 Oct 1995 09:51:41 GMT Organization: Technical University Berlin , Germany Message-ID: <462ind$dog@brachio.zrz.TU-Berlin.DE> References: <eg2gwil4o.fsf@steffi.accessone.com> <45vtmn$1cf@gap.cco.caltech.edu> <RDL.95Oct17224627@world.std.com> In article <RDL.95Oct17224627@world.std.com>, Robert La Ferla <rdl@world.std.com> wrote: >I placed sendmail 8.7.1 quad-fat compiled for NS on ftp.cs.orst.edu. > >Robert La Ferla >Registered OPENSTEP / NEXTSTEP Consultant >Boston, MA >+ 1 (617) 252-0088 I use sendmail 8.7 compiled on my NSfIP V 3.3 .
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: ... messed up my incoming mail (was Re: New 3.3 patch messed up my outgoing mail...) Date: 18 Oct 1995 16:09:35 GMT Organization: Information Technology Solutions, Inc. Message-ID: <4638rv$bfd@news.its.com> References: <eg2gwil4o.fsf@steffi.accessone.com> <45vtmn$1cf@gap.cco.caltech.edu> ccwf@thales.klab.caltech.edu (Charles Fu) wrote: [ ... ] > If your PPP/SLIP is configured to do dial-on-demand, what happens is that after > getting some mail when the PPP/SLIP connection is not active, your modem will > pickup and initiate a PPP/SLIP session and your host exchanges a few packets > with the remote nameserver to resolve the senders' hostnames in your incoming > mail. (In a non 3.3-patched system, there is a lookupd timeout between receipt > of mail by sendmail and initiation of the PPP/SLIP session). This behavior can > be especially annoying if you receive a lot of UUCP mail but have to pay for > connect time on a PPP/SLIP account. > > When PPP/SLIP is not configured to do dial-on-demand, mail will sit, queued and > locked, until the next time sendmail decides to process the queue while the > PPP/SLIP connection is active or the next time the mailserver is rebooted. > (The reason a reboot works is that the mail queue is normally processed during > system start-up prior to configuration of PPP/SLIP.) Explicitly processing the > queue while the connection is down will not help since canonicalization is not > possible. What happens if you run a nameserver on the machine in question that is authoritative for the zone of machines that the incoming mail wants to canonicalize? This nameserver shouldn't attempt to bring up the line, even with dial-on-demand, and sendmail should be able to query the resolver routines as it wants to, so UUCP mail should be able to be delivered. > The only good workaround appears to be to compile and use the current Berkeley > sendmail daemon instead of NeXT's sendmail daemon. Using the NeXT 3.1-3.2 > sendmail patch will work if you are not concerned about the latest security > hole. Well, I am of the opinion that you should always upgrade the sendmail that any vendor ships to a recent UCB version. By the way, I've heard that NS 4.0 includes BSD sendmail v8.6.x, which is really cool since I wouldn't feel compelled to upgrade that.... > Other than that, it might be possible to get around the problem by periodically > killing lookupd, explicitly flushing the mail queue, and restarting lookupd. > However, I _really_ do not recommend anyone doing this for many reasons. > Finally, keeping information in NetInfo for hosts that send you mail often will > reduce the frequency of the problem (because then resolution can be handled > locally). Ack! NetInfo is a distributed database akin to the DNS trees, but NetInfo simply does not scale to very large databases of information well (such as a reasonably complete hostname cache). Run a caching nameserver locally, and recompile sendmail with the resolver library from the bind-4.9.x distribution. This will remove the interactions with lookupd and NetInfo-- which is a significant performance improvement, especially for systems with a single-threaded lookupd. > Can anyone think of other solutions? Besides running a nameserver? Can't you change the sendmail.cf file to not perform name canonicalization....? -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
Newsgroups: comp.sys.next.bugs From: brad@instep.bc.ca Subject: Re: Process that cannot be killed Message-ID: <1995Oct18.173823.9360@instep.bc.ca> Sender: usenet@instep.bc.ca Organization: InStep Mobile Communications Inc. References: <461di1$gip@rhino.cis.vutbr.cz> Date: Wed, 18 Oct 1995 17:38:23 GMT Borek Lupomesky writes > I thought it is impossible to have process, that cannot be killed by >'kill -SIGKILL <pid>'. Yet, NS 3.3 for Intel manages event this. The worst >thing is, that when I try to log out, the system locks-up, because >Workspace Manger can't kill the process as well. > > Bye Borek I'm sure you meant 'kill -KILL <pid>' When I encounter stubborn processes, as a last resort, I'll use: NUKE(1L) UNIX Programmer's Manual NUKE(1L) NAME nuke - terminate a process at the mach task level SYNOPSIS nuke processid DESCRIPTION nuke sends a mach task_terminate message to the mach kernel to terminate the specified process. It's use should be res- tricted to terminating stubborn processes that do not ter- minate when a KILL (9) signal is sent. SEE ALSO kill(1), ps(1), sigvec(2) BUGS The use of this should be a last resort since it may leave things in an inconsistent state. Source below: #include <stdio.h> #include <mach.h> main(int argc, char *argv[]){ if(argc != 2){ fprintf(stderr, "Usage: %s pid\n", argv[0]); exit(1); } else{ task_t task; task_by_unix_pid(task_self(), atoi(argv[1]), &task); task_terminate(task); exit(0); } } -- Brad Head <brad@instep.bc.ca> Software Developer, InStep Mobile Communications Inc. Vancouver, British Columbia, Canada
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <476TJL@ptsmail.ptsem.edu> From: 476TJL@ptsmail.ptsem.edu (LUOMA,TIMOTHY) Subject: login problems persist Message-ID: <6317853001A23A7C@-SMF-> Date: 18 Oct 95 15:04:00 EDT This is really beginning to be a problem. I am logging out successfully (no noticeable problems) but can't log back in without rebooting. I can enter the username (either my own or 'root') and the loginwindow disappears and then comes back a few seconds later. Through a 'console' login, I ran a program called 'psloop' (which just does 'ps -auxwww | sort' every 10 seconds) and tried to login. Here's a bit of the output from when I put in a login name (root, I believe) and password: ################################################## root 2825 1.1 0.5 8.65M 176K co S 0:00 sh ./psloop root 2853 13.9 8.0 20.0M 2.55M ? R 0:04 - console (WindowServer) root 2854 21.8 5.9 4.02M 1.89M ? S 0:03 - console (loginwindow) root 2855 0.0 0.7 2.66M 224K ? S 0:00 - console (WindowServer) root 2864 1.0 0.7 5.60M 224K co S 0:00 sort root 2865 0.4 1.6 1.85M 520K co R 0:00 /bin/ps -auxwww ################################################## You can see there are two WindowServer entries. This seems normal, in that when I login as 'luomat' I see two entries for the WindowServer (one on each side of the loginwindow entry). But then look at the next time through: ################################################## root 2825 0.0 0.5 8.65M 176K co S 0:00 sh ./psloop root 2868 1.6 0.6 1.59M 208K ? S 0:00 - std.9600 console (getty) root 2869 3.3 0.7 5.60M 224K co S 0:00 sort root 2870 8.0 1.5 1.84M 504K co R 0:00 /bin/ps -auxwww ################################################## There is a 'getty' line there, but no window server or loginwindow. Any help appreciated. Mach kernel version: NeXT Mach 3.2: Mon Oct 18 21:57:41 PDT 1993; root(rcbuilder):mk-149.30.15.obj~2/RC_m68k/RELEASE_M68K Kernel configured for a single processor only. 1 processor is physically available. Processor type: MC680x0 (68040) Processor speed: 25 MHz Processor active: 0 System type: 1 Board revision: 0x0 Primary memory available: 32.00 megabytes. Default processor set: 58 tasks, 96 threads, 1 processors Load average: 1.32, Mach factor: 0.56 TjL -- Timothy J. Luoma <476tjl@ptsmail.ptsem.edu> ASCII Email only! No NeXTMail or MIME please NOTE: My stupid DOS mail reader strips off REPLY-TO lines. Please make sure to include your email address within the text of your message.
From: kbradley@kbradley.us.oracle.com (Kirk Bradley) Newsgroups: comp.sys.next.bugs,comp.sys.next.advocacy Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Date: 18 Oct 1995 22:16:18 GMT Organization: Oracle Corporation. Redwood Shores, CA Message-ID: <463ubi$sjm@inet-nntp-gw-1.us.oracle.com> where did you get the leaks command? looks quite handy (I'm on 3.2) On 10/05/95, Jonathan W. Hendry wrote: >Garance A Drosehn (gad@eclipse.its.rpi.edu) wrote: > >: > To be sure, I tried with a fresh reboot and started close >: > monitoring of Preferences.app. Sure enough, it slowly but >: > steadily grows over real time, gaining about 1MB every 3 hours. > >: You're measuring via 'ps', right? In my case I have been logged >: in for 12 days, and the process for Preferences.app was at 7.99 >: meg. I killed and restarted it, and it's at 7.64 meg. > >: I wonder if there's something else which would cause the difference >: besides the setting of the clock icon. You're on NeXT hardware, >: right? My NS-3.3 machine is NS/Intel, so my results might be >: different because of that. > >: Or is it that you're letting Preferences run, but you have to >: watch the swapfile (instead of 'ps' output) to see the growth? > >I just used leaks on Preferences.app. > >First, without having used it. Then after opening it, and 'walking' through >the various modules. I also switched to each of the clock styles: > >uptime: >5:49pm up 2:02, 1 user, load average: 0.16, 0.38, 0.44 > >Process 239: 323 nodes malloced >Process 239: 1 leaks >Leak: 0x0004d0c0 size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x43800000 0x43080000 > 0x00000060 0x00000000 0x0004d68c 0x00000000 > 0x00000000 0x00000000 0x00000430 > >Then after using Preferences.app: > >> leaks 239 >Process 239: 3503 nodes malloced >Process 239: 26 leaks >Leak: 0x0009bab8 size=32 instance of 'NXStringTable' > 0x04018190 0x0000000a 0x0509f824 0x0509f820 > 0x0000000f 0x000a2bec 0x00000002 0x00000140 >Leak: 0x0009ccd8 size=12 string 'English' >Leak: 0x0009d0fc size=12 string 'German' >Leak: 0x0009d1bc size=12 string 'Spanish' >Leak: 0x00097f5c size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x00000000 0x00000000 > 0x00000160 0x00000000 0x00097f88 0x00000000 > 0x00000000 0x00000000 0x00000090 >Leak: 0x0009d1d4 size=12 string 'Swedish' >Leak: 0x00098174 size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x00000000 0x00000000 > 0x00000160 0x00000000 0x000981a0 0x00000000 > 0x00000000 0x00000000 0x00000090 >Leak: 0x0009d114 size=12 string 'Italian' >Leak: 0x000a1f90 size=304 > 0x69570000 0x00000120 0x00000120 0x0005059c > 0x00000005 0x0005bb84 0x00000000 0xffffffff > 0x00000000 0x00000000 0x00000000 0x00050378 > 0x00000010 0x0005bb84 0x00000000 0x00000000 > 0x00000000 0x00000000 0x00000000 0x0005059c > 0x00000005 0x0005bb84 0x00000000 0x00000000 > 0x00000000 0x00000000 0x00000000 0x00050378 > 0x0000000a 0x0005bb84 0x00000000 0x00000000 > 0x00000000 0x00000000 0x00000000 0x0005059c > 0x00000003 0x0005bb84 0x00000000 0x00000000 > 0x00000000 0x00000000 0x00000000 0x00050378 > 0x00000004 0x0005bb84 0x00000000 0x00000000 > 0x00000000 0x00000000 0x00000000 0x0005059c > 0x00000008 0x0005bb84 0x00000000 0x00000000 > 0x00000000 0x00000000 0x00000000 0x00050378 > 0x00000011 0x0005bb84 0x00000000 0x00000000 > 0x00000000 0x00000000 0x00000000 0x00050378 > 0x00000001 0x000a142c 0x00000000 0xffffffff > 0x00000000 0x00000000 0x00000000 0x000001f0 >Leak: 0x00053870 size=36 string 'Objects/Menus/Main/Info.rtfd' >Leak: 0x0009d0f0 size=12 string 'French' >Leak: 0x00098b30 size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x00000000 0x00000000 > 0x00000160 0x00000000 0x00098b5c 0x00000000 > 0x00000000 0x00000000 0x00000090 >Leak: 0x0005ad10 size=28 instance of 'NXStringTable' > 0x04018190 0x00000000 0x0004864e 0x00048650 > 0x00000001 0x00050798 0x00000040 >Leak: 0x00098c90 size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x00000000 0x00000000 > 0x00000160 0x00000000 0x00098cbc 0x00000000 > 0x00000000 0x00000000 0x00000090 >Leak: 0x00098010 size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x00000000 0x00000000 > 0x00000160 0x00000000 0x0009803c 0x00000000 > 0x00000000 0x00000000 0x00000090 >Leak: 0x00180330 size=56 instance of 'NXSoundOut' > 0x061879f0 0x06126cdc 0x00000021 0x00000025 > 0x00000000 0x00000000 0x00000000 0x00000000 > 0x00000000 0x00000000 0x001803e8 0x00000000 > 0x00000000 0x00000030 >Leak: 0x00097eac size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x00000000 0x00000000 > 0x00000160 0x00000000 0x00097ed8 0x00000000 > 0x00000000 0x00000000 0x00000090 >Leak: 0x0009d1c8 size=12 string 'Italian' >Leak: 0x0009d0e4 size=12 string 'English' >Leak: 0x0004d0c0 size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x43800000 0x43080000 > 0x00000060 0x00000000 0x0004d68c 0x00000000 > 0x00000000 0x00000000 0x000001a0 >Leak: 0x0009d184 size=44 string '/NextApps/Preferences.app/English.lproj' >Leak: 0x0009d120 size=12 string 'Swedish' >Leak: 0x0009d108 size=12 string 'Spanish' >Leak: 0x0009cce4 size=16 string 'French' >Leak: 0x00098be0 size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x00000000 0x00000000 > 0x00000160 0x00000000 0x00098c0c 0x00000000 > 0x00000000 0x00000000 0x00000090 >Leak: 0x000980c0 size=44 instance of 'NXImage' > 0x06186c80 0x00000000 0x00000000 0x00000000 > 0x00000160 0x00000000 0x000980ec 0x00000000 > 0x00000000 0x00000000 0x00000090 > > > -- Kirk Bradley -- Oracle Corp. kbradley@oracle.com (415) 506-2738
Newsgroups: comp.sys.next.bugs From: Gerben_Wierda@RnA.NL Subject: Re: AArgH! 3.3 window server is eating all my CPU time Message-ID: <DGnzLD.1KG@RnA.NL> Sender: gerben@RnA.NL (Gerben Wierda) Organization: G.R.O.S.S. References: <45tb7m$7rh@newsbf02.news.aol.com> Date: Wed, 18 Oct 1995 21:56:01 GMT In article <45tb7m$7rh@newsbf02.news.aol.com> mpaque@aol.com (Mpaque) writes: > In article <45o6e0$gsu@bright.ecs.soton.ac.uk>, > root@I_should_put_my_domain_in_etc_NNTP_INEWS_DOMAIN (Eren Kotan) writes: > > >Hi, after recently "upgrading" to 3.3 from 3.2 I seem to notice that the > >computer (A Cube with an ND board) sometimes slows to an almost > standstill. > >If > >I do a ps -aux from a shell window I see that the -console (Window > Server) > >process is using about 45% of CPU time. Is this a bug? This did not use > to > >happen with 3.2. > > The odds are very goods that some application is stuck in a loop, pumping > PostScript at the Window Server. After your system has been in this state > for several minutes, from a remote login (to avoid getting a false reading > from Terminal.app) do a 'ps aux'. See which applications are running with > a fairly high CPU usage. One of these is likely to be the culprit. Quit > or kill these apps one at a time until the system load drops. > > I'd be interested in hearing what you find the culprit to be. Me too. I only noticed this after upgrading from 3.3 to 3.3-patch... My problem might be something else, however, since when it happens I even cannot switch apps (let alone use a ps aux tofind out what is wrong). Events are not handled. What I do see is that the load avarage jumps to mucho-mucho (Monitor.app) and the system starts working again after +/- 5 seconds. I wondered if the new netinfo might have something to do with it. -- Gerben_Wierda@RnA.nl (Gerben Wierda) NEXTSTEP RD242 "If you don't know where you're going, any road will take you there" Paraphrased in Alice in Wonderland, originally from the Talmud.
From: root@darkside.aladdin.co.uk (Eren Kotan) Newsgroups: comp.sys.next.bugs Subject: Re: AArgH! 3.3 window server is eating all my CPU time Date: 18 Oct 1995 23:47:33 GMT Organization: Fluffy Barbarians Software Message-ID: <4643ml$fqf@alpha.aladdin.co.uk> References: <45o6e0$gsu@bright.ecs.soton.ac.uk> <45tb7m$7rh@newsbf02.news.aol.com> Mpaque (mpaque@aol.com) wrote: : In article <45o6e0$gsu@bright.ecs.soton.ac.uk>, : root@I_should_put_my_domain_in_etc_NNTP_INEWS_DOMAIN (Eren Kotan) writes: : : I'd be interested in hearing what you find the culprit to be. Hi and thanks for the suggestions. I have now changed my SLIP software from Louis Mamakos' most excellent SLIP to TransSYS PNI 1.13. This seems to have cured the problems I was experiencing with the slow computer. I hardly ever see the spinning disk now :) However, I miss the free SLIP package's ability to dial on demand, it is a minor hassle to bring the connection up manually. But a very minor hassle ;-} The reason I thought I'd change is that I had read somewhere the free SLIP may break or function slowly with the new kernel in 3.3. Mind you, TransSYS PNI 1.13 isn't really supposed to work with 3.3 either, but I have no problems with it. Thanks, Eren Kotan -- ***************************************************************************** * You can tune a car engine, but you can't tune a fish. * *---------------------------------------------------------------------------* *Eren Kotan, the best friend money can buy. ASCIImail:eek93@ecs.soton.ac.uk * *Southampton University,Computer ScienceNeXTMail:root@darkside.aladdin.co.uk* *World Wide Web page: http://whirligig.ecs.soton.ac.uk/~eek93/cv.html* *****************************************************************************
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: nwc@orc-ny.com (Nick Christopher) Subject: PPD print options ignored on non-NSXTSTEP remote printers Message-ID: <DGp9nJ.86@orc-ny.com> Sender: nwc@orc-ny.com (Nick Christopher) Organization: Organization Resources Counselors, Inc. Date: Thu, 19 Oct 1995 14:30:54 GMT We have a problem with controlling a remote printer here. The current setup is analogous to the NEXTSTEP boxes printing to a printer hanging off a non-NEXTSTEP machine (I'll describe the exact setup at the posts end if details are what you want). We have a PPD file for the printer and it allows us to select cool things like double sided printing etc. Setting these options and printing does nothing. I would guess the PPD was bogus but if you use the Print Panel's save "PostScript For Chosen Printer" option and then print that file all the options work. In the NEXTSTEP 3.2 release notes there is a comment that addresses this: Reference 30117 Problem Process PostScript comments at host Description Device-specific PostScript for setting resolution, page size, etc. is included in a job at print time based on information contained in a PPD file. This inclusion is normally handled by the NeXT machine to which the printer is directly connected (the print server). In Release 3.1, if you put a property called "_finalform" in the printer entry, the PostScript inclusion process is handled by the machine submitting the print job for spooling. Reading this I added the "_finalform" property but this made no difference. Has this option gone away since 3.1? Any ideas? Anybody doing this? EXACT SETUP The printer is actually a rather spastic printer and doesn't support any UNIX style of queuing. You can FTP jobs to it however and query its ready status so the vendor provided a program that submits a job and waits for its completion to return. So we use this as the input filter in a local printcap entry and remote entries refer to that machine. So the remote printer entry is identical to any normal remote entry. Here are the two printcap entries: # Remote entry in NetInfo's / domain Lionheart: \ :sd=/usr/spool/NeXT/Lionheart:rm=seaside:_finalform: \ :rp=Local_Lionheart:lp=:ty=KODAK 1392 Model 44: \ :note=KODAK lionheart in Printer Room:lo=lock: # Local entry on machine seaside in its Netinfo Local_Lionheart: \ :lp=/dev/null:if=/usr/local/etc/iprsend.if: \ :sd=/usr/spool/NeXT/Local_Lionheart:lo=lock: \ :ty=KODAK 1392 Model 44: \ :note=Postscript Printer in Printing Room: \ :mx#0:_finalform: I put the _finalform in both places in desperation it really based on NeXT's note should only be in the remote entry. Note the local entry uses the input filter to fire off "iprsend.if" which is a script calling a program that actually ships off the PostScript and waits on completion. \n -- Nicholas Christopher nwc@gun.com,nwc@orc-ny.com
Newsgroups: comp.sys.next.bugs From: nwc@orc-ny.com (Nick Christopher) Subject: PPD parsing bug in NEXTSTEP Message-ID: <DGpALH.A8@orc-ny.com> Sender: nwc@orc-ny.com (Nick Christopher) Organization: Organization Resources Counselors, Inc. Date: Thu, 19 Oct 1995 14:51:17 GMT Got a legitimate ppd file for a printer the other day that would crash any app when the "Options..." panel from the "Print..." panel was completed with the "OK" button. I found that if your PPD contains no resolution option handler (even for printers that support only one resolution) this crashes your app with a segmentation fault. _The Saga That Lead to the Above Conclusion_ I attached gdb to an app and crashed it to see what was happening and saw that strlen was being passed a (char *)0 from the PrintInfo object's setJobFeature:toValue: method. Since all the offending code was stripped I wrote a do-nothing app with a subclass of PrintInfo that posed as PrintInfo. My subclass just overrode setJobFeature:toValue: and printed out the arguments (while watching for nil) and I found that setJobFeature:toValue: was being called with "Resolution",(char *)0 as arguements. I reviewed the PPD file and sure enough since the printer supports only one resolution it didn't specify any options handler for it. I poked around in NeXT's ppds and found that NeXT had a default snippet of code for printers that support only one resolution. Put it in and fixed the crash. But I wonder if NEXTSTEP is right to require the resolution stuff of not? I don't have access the th PPD 4.0 spec. Regardless it should have handled the problem better than crashing. \n
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin,comp.sys.next.programmer From: nwc@orc-ny.com (Nick Christopher) Subject: SOLUTION: PPD And remote printer control problem fixed Message-ID: <DGpBGr.Bo@orc-ny.com> Sender: nwc@orc-ny.com (Nick Christopher) Organization: Organization Resources Counselors, Inc. Date: Thu, 19 Oct 1995 15:10:02 GMT Call of the hounds...if there were any. The documented "_finalform" property is elsewhere (*one* other reference) documented as "_nxfinalform"! Add those two little letters and it all works! \n -- Nick "Document the Arcana" Christopher
Newsgroups: comp.sys.next.bugs From: Jacques Garbi Subject: Re: cause of infinitely-growing swapfile in 3.3 found!!!!!!! Message-ID: <DGoto4.A0@touga.vd.alphanet.ch> Sender: jacques@touga.vd.alphanet.ch (Jacques Garbi) Organization: Touga Management References: <45dvmi$fta@digdug.pencom.com> Date: Thu, 19 Oct 1995 08:45:40 GMT In article <45dvmi$fta@digdug.pencom.com> robin@pencom.com writes: > felix.rauch@harka.limmat.net.ch (Felix Rauch) wrote: > : Strange, but on my system it seems to be the other way around. Shortly > : after booting and logging in, it looks like this (ps output): > > : USER PID %CPU %MEM VSIZE RSIZE TT STAT TIME COMMAND > : xilef 215 0.0 5.0 5.94M 1.59M ? SW 0:01 > /NextApps/Preferences.app/P > > : After some hours: > : xilef 215 0.0 4.2 5.94M 1.34M ? SW 0:04 > /NextApps/Preferences.app/P > > : Before I shut the machine down again (uptime: 7:57): > : xilef 215 0.0 4.0 5.94M 1.29M ? SW 0:04 > /NextApps/Preferences.app/P > > : The RSIZE gets smaller over the time! I wonder whether I could reduze > : its memory-usage to 0 if I let the machine on forever ;-) > : This is on a black colorstation with NS3.3. > > The RSIZE will eventually drop to 'just enough' to run the clock (so long as you > don't open the app and play with the preferences...). On my NeXT Turbo (mono), > that is 1024Kbytes. The reason you are seeing the RSIZE value drop is because > portions of the app are being paged out over time. If you started a _huge_ app, > it would probably page _all_ of the Preferences app out -- with the exception of > the clock. The real interesting number is the VSIZE. Why is this different on > different machines? Mine is 6.15MB (not enough to get worked up about -- but > still ~160Kb different than yours)... > Effectively, this is weird. Mine is 6.04 MB in VSize and 2.01MB in RSize just after a clean bootup If anybody finds the answer to that, I'd be happy to hear it. --- Jacques GARBI TOUGA MANAGEMENT Inc. Av. de France 33 1004 Lausanne Switzerland Phone : 011 41 21 661 16 76 Fax : 011 41 21 661 16 73 NeXTMail or MIME : jgarbi@touga.vd.alphanet.ch
Newsgroups: comp.sys.next.bugs From: tom@hukatronic.cz (Tomas Hurka) Subject: Re: CD:s generate "unexpected kernel page fault" Message-ID: <DGqLEM.G8@hurka.UUCP> Sender: tom@hurka.UUCP (Tomas Hurka) Organization: Hukatronic (H.C.C.) References: <MJANTTI.95Oct14144757@aton.abo.fi> Date: Fri, 20 Oct 1995 07:42:22 GMT Hi Marcus, In article <MJANTTI.95Oct14144757@aton.abo.fi> mjantti@aton.abo.fi (Markus J{ntti STA) writes: > When I try to read a CD other than the NeXT User and Developer my > system crashes, i.e., it throws plenty of garbage in a small monitor > on the screen and will only reboot from the large monitor. [..] It looks like you have some corrupted files in /usr/filesystems/CDROM.fs directory. Try to replace them from the NeXT User CD and see if it helps. Bye, -- Tomas Hurka tom@hukatronic.cz NeXTMAIL and MIME OK (international mail <50 KB accepted)
From: wilkie@narya. (Alexander Wilkie) Newsgroups: comp.sys.next.bugs Subject: Mac CDs not readable Date: 20 Oct 1995 12:44:15 GMT Organization: Vienna University of Technology, Austria Message-ID: <4685iv$3c9@news.tuwien.ac.at> The Problem: On my P90/3.3 system I found that I can't properly read CDs with a Mac filesystem. Was this ever intended to work? Are there some new MacCD.CDfilesystemTypes around NeXTStep 3.3 can't cope with? Symptoms range from .mov and .mov.rsrc files appearing under the same name (causing considerable confusion in WorkspaceManager) to total rejection of the CD. The CD-ROM drive is a Toshiba 3401A and works fine with DOS and NeXT CDs. Any ideas, folks? ys Alexander Wilkie wilkie@cg.tuwien.ac.at
From: andrew.abernathy@mccaw.com (Andrew Abernathy) Newsgroups: comp.sys.next.bugs Subject: Re: Mac CDs not readable Date: 20 Oct 1995 19:18:14 GMT Organization: AT&T Wireless Services, Inc. Message-ID: <468slm$gu6@nwestmail.nwest.mccaw.com> References: <4685iv$3c9@news.tuwien.ac.at> In article <4685iv$3c9@news.tuwien.ac.at> wilkie@narya. (Alexander Wilkie) writes: > The Problem: > On my P90/3.3 system I found that I can't properly read CDs with a Mac > filesystem. Was this ever intended to work? Are there some new > MacCD.CDfilesystemTypes around NeXTStep 3.3 can't cope with? > Symptoms range from .mov and .mov.rsrc files appearing under the same name > (causing considerable confusion in WorkspaceManager) to total rejection of > the CD. The CD-ROM drive is a Toshiba 3401A and works fine with DOS and > NeXT CDs. Ah - good thing you gave the symptoms. I was initially going to say "works for me!" However, I think I know your problem. You're probably using cdroms that have both a Mac filesystem and an ISO whatever filesystem on them. Since the basic CDROM filesystem gets loaded first, you really get strange behavior. Try this: go to /usr/filesystems. Rename CDROM.fs to CDRROM.fs.old (or whatever). Then plop in the cdrom - the Mac filesystem will load, since the basic CDROM filesystem is no longer in the way. When you're done, rename the CDROM filesystem back to normal. (If you want the Mac filesystem to always be checked first, you can do a bunch of renaming - change mac.fs to Apple.fs, and rename stuff inside the .fs directory to match. Not everything should be renamed, and I don't remember just what should, so you may have to experiment some.) -- Any similarities between my views and those of AT&T are purely coincidental. andrew.abernathy@attws.com, AT&T Wireless Services NeXTmail, MIME, MSMail, and SunMail in a pinch http://www.serv.net/~wire/andrew.abernathy/ 14335 Northeast 24th Street, Bldg B, Bellevue, WA 98007, U.S.A.
From: samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) Newsgroups: comp.sys.next.bugs Subject: HP 1200C/PS support in NS totally broken. Date: 21 Oct 1995 18:02:19 GMT Organization: School of Computer Science, McGill Univ. Distribution: world Message-ID: <SAMURAI.95Oct21140219@maggie.cs.mcgill.ca> This problem was submitted to Adboe and NeXT. Synopsis: If you buy an HP1200C/PS printer, and attempt to print or save PS images including font information, you will generate a corrupt, unprintable, unviewable document. Problem: THe HP1200C/PS PPD is invalid. Workaround: Use the HP LaserJet IISi Postscript 2 PPD when attempting to install an HP1200C/PS printer. Comment: This really pisses me off, as the 1200 C is a great, cost effective color printer. In order to get it to work right, I have to use a PPD file that doesn't give me control over being able to print transparencies or glossy paper. NeXT nor Adobe has published a fix yet. The ultimate solution is to just buy a Mac and an apple printer. - db -- "Investing in tech stocks is a high risk." -- Bill Gates, Microsoft Chairman, after incurring a two day paper loss of 2 billion dollars on Microsoft stock.
From: ccwf@acquine.klab.caltech.edu (Charles Fu) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: ... messed up my incoming mail (was Re: New 3.3 patch messed up my outgoing mail...) Date: 21 Oct 1995 22:31:02 GMT Organization: California Institute of Technology, Pasadena Message-ID: <46bsb6$5fc@gap.cco.caltech.edu> References: <45vtmn$1cf@gap.cco.caltech.edu> <4638rv$bfd@news.its.com> In article <4638rv$bfd@news.its.com>, Chuck Swiger <chuck@its.com> wrote: >What happens if you run a nameserver on the machine in question that is >authoritative for the zone of machines that the incoming mail wants to >canonicalize? I would think that would prevent a connection from being started. But it doesn't seem like a good idea to run a local nameserver which claims to be authoritative for all zones which might send you mail. >Ack! NetInfo is a distributed database akin to the DNS trees, but NetInfo >simply does not scale to very large databases of information well (such as a >reasonably complete hostname cache). Agreed. I rather wish that NeXT would abandon NetInfo entirely (my personal opinion). Not that that would solve this particular problem, of course. >Besides running a nameserver? Can't you change the sendmail.cf file to not >perform name canonicalization....? I already do run a caching nameserver on my sites. That, by itself, is not enough. And, no, I do not believe it is possible to turn off canonicalization. Quoting form the v8.37 "Sendmail Installation and Operation Guide" (which covers sendmail v8.6): If you systems supports the name server, then the probability is that sendmail will be using it regardless of how you configure sendmail. In particular, the system routine gethostbyname(3) is used to look up host names, and most vendor versions try some combination of DNS, NIS, and file lookup in /etc/hosts. [The next paragraph then explains the UUCP problem and how to work around it using the `I' option, which is not available in NeXT's sendmail.] Adding NetInfo to the mix doesn't change the essential problem. I agree with you: abandoning NeXT sendmail and using the Berkeley distribution seems like the best solution. -ccwf
From: Robert Worne <rworne@primenet.com> Newsgroups: comp.sys.next.bugs Subject: Re: tip crashes NeXT Date: 18 Oct 1995 01:12:04 GMT Organization: A Big, Black Box. Message-ID: <461k94$3q5@nnrp1.news.primenet.com> References: <45lfqk$8rn@neptune.ethz.ch> <45p3tm$98i@news.it.gvsu.edu> berezaw@river.it.gvsu.edu (Bill Bereza) wrote: > In article <45lfqk$8rn@neptune.ethz.ch>, > Roman Puttkammer <rputtkam@iiic.ethz.ch> wrote: > >Does anybody know, what this is? Some kind of race condition bug??? > > > >I know, Intel-based computer are some sort of crap, but I thought NeXT > >to run stable on them!?!?! > Get the latest serial drivers from the Next web site. > I used to have this same problem, but it's gone away since I upgraded to the 3.33 drivers. I did the same on an Intel Box (Upgraded to 3.33) but it "broke" the freeware PPP on that system, (it worked before the upgrade, didn't work after, and worked again after the "downgrade")... -- //--------------------------------------------------------------------- // Robert Worne rworne@primenet.com OS/2-NeXT -=Starving CS Undergrad=- //--------------------------------------------------------------------- // Microsoft Network is prohibited from redistributing this work in // any form, in whole or in part. Copyright, Robert Worne, 1995. // License to distribute this post is available to Microsoft for $1000. // Posting without permission constitutes an agreement to these terms. // Please send notices of violation to rworne@primenet.com and // postmaster@microsoft.com
From: marcel@cs.tu-berlin.de (Marcel Weiher) Newsgroups: comp.sys.next.bugs Subject: Re: HP 1200C/PS support in NS totally broken. Date: 22 Oct 1995 11:44:18 GMT Organization: Technical University of Berlin, Germany Message-ID: <46daqi$f6u@news.cs.tu-berlin.de> References: <SAMURAI.95Oct21140219@maggie.cs.mcgill.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) writes: >If you buy an HP1200C/PS printer, and attempt to print or save PS images >including font information, you will generate a corrupt, unprintable, >unviewable document. Don't have that printer but previewing works just fine! >Problem: >THe HP1200C/PS PPD is invalid. I couldn't find a HP1200C/PS PPD on my NeXT, so I downloaded the DOS PPD from Adobe. The Mac PPD does _not_ work (Mac is typically a little strange, so no wonder...) >Workaround: >Use the HP LaserJet IISi Postscript 2 PPD when attempting to install an >HP1200C/PS printer. Hm... just use the DOS PPD! >Comment: >This really pisses me off, as the 1200 C is a great, cost effective >color printer. In order to get it to work right, I have to use a PPD >file that doesn't give me control over being able to print >transparencies or glossy paper. NeXT nor Adobe has published a fix yet. Maybe because there is no fix for customers... (When is Darcy going to get out of Alpha? :-) >The ultimate solution is to just buy a Mac and an apple printer. You're kidding, right? Of course, the ultimate solution on NS is to just get a NS printing solution, like eXTRAPRINT, with one of the many fine, low-cost ink jets around. Faster, much cheaper, no hassle. Regards, Marcel P.S.: The only way I could generate an 'unviewable' PS file with that PPD was by saving the PS with the 'PostScript for specified printer' option, but that is _supposed_ to be device specific. Nevertheless, the PPD can be modified to even have those display in Preview.
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <flexus!rfschtkt@maze.ruca.ua.ac.be> Date: Sun, 22 Oct 95 10:27:26 +0100 From: Raf Schietekat <flexus!rfschtkt@maze.ruca.ua.ac.be> Message-ID: <9510220927.AA00431@flexus> Subject: New KBNS (``Known Bugs in NeXT Software'') edition (33.1). (Same message as submitted to comp.sys.next.announce.) Dear NeXT software user, This is to let you know about the new edition 33.1 of KBNS (``Known Bugs in NeXT Software''), the *only* public and aspiring-to-be-comprehensive database of problems and bugs (with workarounds, where known!) in all of NeXT's software products (not just NEXTSTEP itself), to be used together with NeXT's NeXTanswers. It's the second edition mentioning NEXTSTEP 3.3 at all, and there are a few things about OpenStep and EOF as well. ftp://ftp.cs.orst.edu/software/NeXT/documents/KBNS.33.1.README.rtf.(gz/Z) ftp://ftp.cs.orst.edu/software/NeXT/documents/KBNS.33.1.rtf.(gz/Z) (Some verification code in same location, KBNS* something.) So fetch the document, use it, and send me neat&concise reports about whatever is still missing (a new problem, a good workaround), about changes to applicability ranges, about a verification you did, if you volunteer to referee items in a particular category, about any mistakes, or a note about what you think of KBNS (if you think it is no good, I won't know unless you tell me, but please be specific; if you've been able to benefit from it, I'll be very happy to hear from you!). *** If an existing KBNS entry describes a problem you are having, do send a copy to Bug_NeXT! Perhaps effectiveness is in the numbers... *** KBNS is totally dependent on your cooperation. I see my role as an administrator and moderator for compiling the collective knowledge of the NeXT software community (about bugs, of course). It's up to YOU! (The KBNS file itself is rather hefty (do an ls and/or fetch the README first). If anyone knows a free program to distribute patches to RTF files (GNU diff/patch or any other line-based algorithm won't do), please tell me.) Raf Schietekat, RfSchtkt@maze.ruca.ua.ac.be (NeXTmail), Flanders, Belgium If I don't answer: my mail relay can't handle !, % or .uucp, I think *** The year 2000 will be the last year of the 20th century. ***
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin,de.comp.sys.next Subject: Doing multibackup on tape impossible? Date: Sun, 22 Oct 1995 22:54:05 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951022224730.1088B-100000@hphalle0.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, I own a TEAC tape streamer (600MB uncompressed, 12MB/sec, no DAT!, SCSI I). I'm currently trying to put different little backups on one tape. This seems to be impossible! I've tried 'mt -f /dev/nrst0 fsf 1' and afterwards doing the backup with either dump or gnutar. Nothing works. The tape _has_to_be_ rewinded to get things done. It even is impossible to do a 'mt -f /dev/nrst0 <any forward command>; mt -f /dev/nrst0 eof <number>' Anybody experienced the same problems? I believe this is a driver problem, because my Amiga is able to append to a tape as needed. I'm using NS 3.3 FIP and SCSI-Tape V3.31. Greetings, Boerny. -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/ scholz@ve1.rm.op.dlr.de http://www.leo.org/~scholz/
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin,de.comp.sys.next Subject: Re: Doing multibackup on tape impossible? Followup-To: comp.sys.next.bugs,comp.sys.next.sysadmin,de.comp.sys.next Date: 23 Oct 1995 08:15:41 GMT Organization: Technical University of Berlin, Germany Distribution: world Message-ID: <46fivd$5qb@news.cs.tu-berlin.de> References: <Pine.HPP.3.91.951022224730.1088B-100000@hphalle0.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Bernhard Scholz (scholz@informatik.tu-muenchen.de) wrote: : Hello, : I own a TEAC tape streamer (600MB uncompressed, 12MB/sec, no DAT!, SCSI : I). I'm currently trying to put different little backups on one tape. : This seems to be impossible! : I've tried 'mt -f /dev/nrst0 fsf 1' and afterwards doing the backup with : either dump or gnutar. Nothing works. The tape _has_to_be_ rewinded to : get things done. It even is impossible to do a 'mt -f /dev/nrst0 <any : forward command>; mt -f /dev/nrst0 eof <number>' : Anybody experienced the same problems? I believe this is a driver : problem, because my Amiga is able to append to a tape as needed. First of all: There's a new SCSITape Driver out there. Have you checked this out? Second: On black, I have never been able to use the fsf/bsf commands in a reasonable manner. When I want to append to a tape, I forward the tape by simply reading it up to the end (multiple cat /dev/nrst0 >/dev/null until IO error) and then appending to the tape. This has always worked for me. Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 23 Oct 1995 04:15:11 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <46f4sf$8uj@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) Newsgroups: comp.sys.next.bugs Subject: Re: HP 1200C/PS support in NS totally broken. Date: 23 Oct 1995 15:13:21 GMT Organization: School of Computer Science, McGill Univ. Message-ID: <SAMURAI.95Oct23111321@maggie.cs.mcgill.ca> References: <SAMURAI.95Oct21140219@maggie.cs.mcgill.ca> <46daqi$f6u@news.cs.tu-berlin.de> In-reply-to: marcel@cs.tu-berlin.de's message of 22 Oct 1995 11:44:18 GMT <marcel@cs.tu-berlin.de> writes: >samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) writes: >>If you buy an HP1200C/PS printer, and attempt to print or save PS images >>including font information, you will generate a corrupt, unprintable, >>unviewable document. >>THe HP1200C/PS PPD is invalid. >I couldn't find a HP1200C/PS PPD on my NeXT, so I downloaded the DOS >PPD from Adobe. The Mac PPD does _not_ work (Mac is typically >a little strange, so no wonder...) >>Workaround: >>Use the HP LaserJet IISi Postscript 2 PPD when attempting to install an >>HP1200C/PS printer. >Hm... just use the DOS PPD! The DOS PPD is the same as the one shipped by NeXT. I've received email from other people claiming it works fine for them.... however, I have replicated this problem on NS3.3 and NS3.2 on both HPPA and Intel based machines. If you're trying to replicate, please note that you must have fonts in your document *and* you must save "Postscript for Chosen Printer / Include Fonts". Believe me, if this works for everyone else, I'll be happy but confused. >>Comment: >>This really pisses me off, as the 1200 C is a great, cost effective >>color printer. In order to get it to work right, I have to use a PPD >>file that doesn't give me control over being able to print >>transparencies or glossy paper. NeXT nor Adobe has published a fix yet. >Maybe because there is no fix for customers... (When is Darcy going >to get out of Alpha? :-) Huh? I've not received Alpha anything from anyone. >>The ultimate solution is to just buy a Mac and an apple printer. >You're kidding, right? Nope. My recent forays into getting third party software (there still isn't a single decent piece of image editing software) and scanner software (don't even get me started) make me want to get a real system with real tools that have real support. It's next to impossible to do image work under NEXTSTEP. Don't tell me WetPaint... it's weak and less-than-ergonomic. >Of course, the ultimate solution on NS is to just get a NS >printing solution, like eXTRAPRINT, with one of the many >fine, low-cost ink jets around. Faster, much cheaper, no hassle. When an operating system comes bundled with PostScript, which you're paying money for, I think that you should be able to assume that you'll be able to print on PostScript printers without buying more software. I'm not shooting down eXTRAPRINT, as I know nothing about it, but I just think that NEXTSTEP should work right. >P.S.: The only way I could generate an 'unviewable' PS file > with that PPD was by saving the PS with the 'PostScript > for specified printer' option, but that is _supposed_ > to be device specific. Nevertheless, the PPD can be > modified to even have those display in Preview. Unfortunately, the printer likes that file just as much as the WindowServer does. If you care to post your PPD modifications, I'll gladly attempt to use it with my printer and post back the results. - db -- "Investing in tech stocks is a high risk." -- Bill Gates, Microsoft Chairman, after incurring a two day paper loss of 2 billion dollars on Microsoft stock.
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: ... messed up my incoming mail (was Re: New 3.3 patch messed up my outgoing mail...) Date: 23 Oct 1995 16:42:46 GMT Organization: Information Technology Solutions, Inc. Message-ID: <46ggm6$6h6@news.its.com> References: <45vtmn$1cf@gap.cco.caltech.edu> <4638rv$bfd@news.its.com> <46bsb6$5fc@gap.cco.caltech.edu> ccwf@acquine.klab.caltech.edu (Charles Fu) wrote: > In article <4638rv$bfd@news.its.com>, Chuck Swiger <chuck@its.com> wrote: >> What happens if you run a nameserver on the machine in question that is >> authoritative for the zone of machines that the incoming mail wants to >> canonicalize? > > I would think that would prevent a connection from being started. But it > doesn't seem like a good idea to run a local nameserver which claims to > be authoritative for all zones which might send you mail. Not for all zones which send you mail, for all zones that are receiving mail locally. Which type of address is being canonicalized (or are both)? But if you get a lot of mail from a small set of places, you could try having your nameserver be a secondary for those zones as well. It would probably help, and certainly can't hurt anything (because (a) as a secondary, your responses should be correct, and (b) no one in the outside world is likely to ask your nameserver to respond for those zones anyway; they'll go up to the root nameservers and down to the registered nameservers). Perhaps you should try it. > I already do run a caching nameserver on my sites. That, by itself, is > not enough. And, no, I do not believe it is possible to turn off > canonicalization. Sure it is. You can use the [host.do.main] syntax. For example, we've got a set of customers who are on the other end of ISDN dialups, and we are a secondary MX site for them to spool their mail while their links are down. The following rule will attempt direct delivery to the A record associated with the name matching the F class, instead of doing an MX lookup. This is in ruleset 0 before the local delivery mailers. # Deliver directly to machines in the CF class without doing an MX lookup -CWS # Note that the rewriting rules for the headers may not be right. If so, # then we can define a new mailer "smptf" which uses different rulesets. R$+ < @ $=F . > $* $#smtp $@ [$2] $: $1 < @ $2 > $3 This works fine because our nameservers are authoritative for the address records of all of our ISDN customers (because we have to handle their DNS info because their ISDN links will not always be up). -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: vm_stat math overflow error, BugNeXT Log #75719 Date: 23 Oct 1995 23:50:47 GMT Organization: Information Technology Solutions, Inc. Message-ID: <46h9oo$6h6@news.its.com> The vm_stat command has a math overflow error when computing the "Object cache" percentage when the system has been up a long time: % uptime 7:16pm up 23 days, 16:55, 9 users, load average: 1.45, 1.25, 1.14 % vm_stat Mach Virtual Memory Statistics: (page size of 8192 bytes) Pages free: 35. Pages active: 2238. Pages inactive: 1224. Pages wired down: 440. "Translation faults": 404764298. Pages copy-on-write: 101900465. Pages zero filled: 38194538. Pages reactivated: 191956004. Pageins: 76081. Pageouts: 26776. Object cache: 32802966 hits of 32830371 lookups (-30% hit rate) Note the last line is wrong. This bug was found under 3.3 (+ patch) on black hardware, but it's likely to happen on all platforms, although it may be difficult to reproduce. (It requires the system to be up for a long time under fairly heavy usage.) Here is source code to a fixed version of vm_stat, although this doesn't implement the "vm_stat <<interval>>" functionality. It does, however, return a more pleasant format with additional information..... /* * vm_stats replacement. * * Written 9/23/95 by Chuck Swiger. * Last update: 10/20/95 to fix "Object cache" error. */ #import <mach/mach.h> int main() { vm_statistics_data_t vm_stats; int result, total_pages; result=vm_statistics(task_self(), &vm_stats); if (result != KERN_SUCCESS) mach_error("An error calling vm_statistics()!", result); else { total_pages = vm_stats.free_count + vm_stats.active_count + vm_stats.inactive_count + vm_stats.wire_count; printf("Mach Virtual Memory Statistics: (page size of %d bytes)\n", vm_stats.pagesize); printf("Total pages: %20d\n", total_pages); printf("Pages free: %20d (%5.2f %%)\n", vm_stats.free_count, (double)((100.0 * vm_stats.free_count) / total_pages)); printf("Pages active: %20d (%5.2f %%)\n", vm_stats.active_count, (double)((100.0 * vm_stats.active_count) / total_pages)); printf("Pages inactive: %20d (%5.2f %%)\n", vm_stats.inactive_count, (double)((100.0 * vm_stats.inactive_count) / total_pages)); printf("Pages wired down: %20d (%5.2f %%)\n", vm_stats.wire_count, (double)((100.0 * vm_stats.wire_count) / total_pages)); printf("Translation faults: %20d\n", vm_stats.faults); printf("Pages copy-on-write:%20d\n", vm_stats.cow_faults); printf("Pages zero filled: %20d\n", vm_stats.zero_fill_count); printf("Pages reactivated: %20d\n", vm_stats.reactivations); printf("Pageins: %20d\n", vm_stats.pageins); printf("Pageouts: %20d\n", vm_stats.pageouts); if (vm_stats.pageouts) printf("Efficiency ratio: %20d\n", (vm_stats.faults/vm_stats.pageouts)); // Use (long long) to avoid math overflow printf("Object cache: %d hits of %d lookups (%d%% hit rate)\n", vm_stats.hits, vm_stats.lookups, (int) (((long long) 100 * vm_stats.hits) / vm_stats.lookups)); } } -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: marcel@cs.tu-berlin.de (Marcel Weiher) Newsgroups: comp.sys.next.bugs Subject: Re: HP 1200C/PS support in NS totally broken. Date: 24 Oct 1995 00:53:12 GMT Organization: Technical University of Berlin, Germany Message-ID: <46hddo$m7p@news.cs.tu-berlin.de> References: <SAMURAI.95Oct21140219@maggie.cs.mcgill.ca> <46daqi$f6u@news.cs.tu-berlin.de> <SAMURAI.95Oct23111321@maggie.cs.mcgill.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) writes: >If you're trying to replicate, please note that you must have fonts in >your document *and* you must save "Postscript for Chosen Printer / >Include Fonts". >Believe me, if this works for everyone else, I'll be happy but confused. The "PostScript for Chosen Printer" delivers device-_dependent_ PostScript. It is not intended to work on generic output devices other than the printer specified, because it includes special code, from the _PPD_ that is likely to work only on that printer. So it is absolutely within spec. for Preview not to be able to display that file. (That's what the Preview button is for, the print system after all, loads the fonts and device specific info just before it ships the job off to the printer. That was the first part. Second part: The information included with the file is taken from the PPD. Except for that, NS doesn't do anything differently, so the problem is fairly definitely with the PPD file, not with NS. So this group is definitely the wrong forum. >When an operating system comes bundled with PostScript, which you're >paying money for, I think that you should be able to assume that you'll >be able to print on PostScript printers without buying more software. >I'm not shooting down eXTRAPRINT, as I know nothing about it, but I just >think that NEXTSTEP should work right. As I explained above, NEXTSTEP does work right, you have an apparently broken PPD. eXTRAPRINT, by the way, makes _non_-PostScript printers work with NS. Much more cost effective than having a PS-interpreter in the printer (and no font downloading, faster printing, etc.) The modified PPD is in the mail. Regards, Marcel
From: samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) Newsgroups: comp.sys.next.bugs Subject: Re: HP 1200C/PS support in NS totally broken. Date: 24 Oct 1995 14:06:35 GMT Organization: School of Computer Science, McGill Univ. Message-ID: <SAMURAI.95Oct24100635@maggie.cs.mcgill.ca> References: <SAMURAI.95Oct21140219@maggie.cs.mcgill.ca> <46daqi$f6u@news.cs.tu-berlin.de> <SAMURAI.95Oct23111321@maggie.cs.mcgill.ca> <46hddo$m7p@news.cs.tu-berlin.de> In-reply-to: marcel@cs.tu-berlin.de's message of 24 Oct 1995 00:53:12 GMT <marcel@cs.tu-berlin.de> writes: >samurai@maggie.cs.mcgill.ca (Darcy BROCKBANK) writes: >>If you're trying to replicate, please note that you must have fonts in >>your document *and* you must save "Postscript for Chosen Printer / >>Include Fonts". >>Believe me, if this works for everyone else, I'll be happy but confused. >The "PostScript for Chosen Printer" delivers device-_dependent_ >PostScript. It is not intended to work on generic output >devices other than the printer specified, because it includes >special code, from the _PPD_ that is likely to work only >on that printer. >So it is absolutely within spec. for Preview not to be >able to display that file. (That's what the Preview >button is for, the print system after all, loads the >fonts and device specific info just before it ships >the job off to the printer. I understand what you wrote... *However* (1) I've never had a Postscript for Chosen Printer generate PS files that the WindowServer couldn't open... I've tried many types and all were openable by Preview.app. (2) The only one that I've tried that generated a file not handled by the WindowServer was the HP1200C/PS PPD. (3) The only printer that I've tried that hasn't worked properly is the HP1200C/PS PPD with the HP1200C/PS. So, let's look at this graphically... Postscript for Chosen printer gives correct results on the given device with the given PPD file. ---------------------- Window Server ------- HP 1200C/PS HP1200C/PS ppd NO NO HPLaserJet III ppd YES YES Apple LaserWriter ppd YES n/a ... THe HP1200C/PS ppd does not work with the HP1200C/PS printer, when one prints specific PS. The file it generates does not print, even though the file is MEANT for this printer. Nor, is the file previewable. Switch the PPD to another HP printer, and magically, it's both previewable and printable. Coincidence? Yes, I understand that "it's within spec." However, if most of the time Preview can handle it, when Preview does NOT handle it it could indicate that the file is corrupt as a possibility, while not excluding the other possibility that the file was falling within spec. The proof is that the printer chokes on the output. The fact that Preview chokes on it adds to the confidence level. Now, should everyone else's printers seem to work just fine with PS for chosen printer as an option, including font information, then my printer must be damaged or newer than theirs. This is a possibility, but I'd lean towards a software bug, hence the bug report. - db -- "Investing in tech stocks is a high risk." -- Bill Gates, Microsoft Chairman, after incurring a two day paper loss of 2 billion dollars on Microsoft stock.
From: trey@hsv.tybrin.com (Trey McClendon) Newsgroups: comp.sys.next.bugs Subject: Problem with FaxReader.app Date: 24 Oct 1995 08:58:32 -0500 Organization: TYBRIN Corporation Message-ID: <46ire8$b5t@tybrin4.hsv.tybrin.com> When going through and cleaning out old faxes, FaxReader will always 'hang' after opening, printing, and deleting a number of faxes. It is usually after about 6 or so, but I haven't counted to see if it is the same number each time. Has anyone else experienced this? It is with NS 3.3 on NeXT slabs, where both the fax server and client are NeXT machines. Thanks, Trey -- Trey McClendon TYBRIN Corporation trey@hsv.tybrin.com Huntsville, AL NeXT / MIME Mail Accepted Fax: 205-837-3472
From: support@radical.com (Radical) Newsgroups: comp.sys.next.bugs Subject: Re: Problem with FaxReader.app Date: 24 Oct 1995 16:08:51 GMT Organization: Radical System Solutions, Inc. Message-ID: <46j32j$a1q@radical2.radical.com> References: <46ire8$b5t@tybrin4.hsv.tybrin.com> In <46ire8$b5t@tybrin4.hsv.tybrin.com> Trey McClendon wrote: > When going through and cleaning out old faxes, FaxReader will always > 'hang' after opening, printing, and deleting a number of faxes. It is > usually after about 6 or so, but I haven't counted to see if it is the > same number each time. Has anyone else experienced this? It is with NS > 3.3 on NeXT slabs, where both the fax server and client are NeXT machines. I have problems with FaxReader printing faxes from a NS 3.3 Intel machine. After the first fax is printed, I get errors on the printer and I have to restart the app to print more. Sometimes I have to reboot the system to print faxes. There are also some faxes that just downright refuse to print at a certain page! Ralph -- ____________________________________ / Radical System Solutions, Inc. /\ / Software Development, Consulting \/ support@radical.com (NeXTMail, MIME)
From: J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.software Subject: (some) NeXT apps can't print on non-next printer Followup-To: comp.sys.next.bugs Date: 25 Oct 1995 20:56:40 GMT Organization: Theory Group, Physics, QMW College Distribution: world Message-ID: <J.M.FIGUEROA.95Oct25205640@strings7.ph.qmw.ac.uk> We recently connected an HP LaserJet 4MP (600DPI) Postscript printer to our next network. The machines seem to recognise the printer type because it shows a printer definition file for it. There are no problems printing PS files generated from TeX, or text files. However printing from some other NeXT apps doesn't work. Instead of the expected output, one gets a page with the following message: ERROR: configurationerror OFFENDING COMMAND: setpagedevice STACK: Any ideas? Many thanks in advance. Cheers, Jose -- +----------------------------------+----------------------------------+ | Dr Jose M Figueroa-O'Farrill | Vox: +44.171.975-5055 | | Department of Physics (Rm. 227) | Fax: +44.181.981-7465 | | Queen Mary and Westfield College | mailto:J.M.Figueroa@qmw.ac.uk | | Mile End Road, London E1 4NS, UK | | | and | | http://stringswww.ph.qmw.ac.uk/~jmf/jmf.html | +----------------------------------+----------------------------------+
From: Paul Lynch <Paul_Lynch@plsys.com> Newsgroups: comp.sys.next.bugs Subject: Re: (some) NeXT apps can't print on non-next printer Date: Wed, 25 Oct 1995 23:09:12 GMT Organization: P & L Systems Sender: news@seer.demon.co.uk Message-ID: <1995Oct25.230912.7477@seer.demon.co.uk> References: <J.M.FIGUEROA.95Oct25205640@strings7.ph.qmw.ac.uk> In article <J.M.FIGUEROA.95Oct25205640@strings7.ph.qmw.ac.uk> J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) writes: > > We recently connected an HP LaserJet 4MP (600DPI) Postscript printer > to our next network. The machines seem to recognise the printer type > because it shows a printer definition file for it. There are no > problems printing PS files generated from TeX, or text files. However > printing from some other NeXT apps doesn't work. Instead of the > expected output, one gets a page with the following message: > > ERROR: configurationerror > OFFENDING COMMAND: setpagedevice > > STACK: > > > Any ideas? For not very obvious reasons, LaserJets get upset if you try to print a Letter sized document to them when the printer is set to A4. Make sure the document paper size and the printer setting match. Paul -- Paul Lynch (NeXTmail) http://www.plsys.co.uk/~paul (under construction)
From: ccwf@acquine.klab.caltech.edu (Charles Fu) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: ... messed up my incoming mail (was Re: New 3.3 patch messed up my outgoing mail...) Date: 26 Oct 1995 01:13:42 GMT Organization: California Institute of Technology, Pasadena Message-ID: <46mnc6$mdm@gap.cco.caltech.edu> References: <45vtmn$1cf@gap.cco.caltech.edu> <4638rv$bfd@news.its.com> <46bsb6$5fc@gap.cco.caltech.edu> <46ggm6$6h6@news.its.com> In article <46ggm6$6h6@news.its.com>, Chuck Swiger <chuck@its.com> wrote: >ccwf@acquine.klab.caltech.edu (Charles Fu) wrote: >> In article <4638rv$bfd@news.its.com>, Chuck Swiger <chuck@its.com> wrote: >>> [asks about a local authoritative nameserver] >> ... But it doesn't seem like a good idea to run a local nameserver which >> claims to be authoritative for all zones which might send you mail. >Not for all zones which send you mail, for all zones that are receiving mail >locally. Which type of address is being canonicalized (or are both)? Exactly. BOTH the from and to host are canonicalized. That's why the problem is especially severe. >> I already do run a caching nameserver on my sites. That, by itself, is >> not enough. And, no, I do not believe it is possible to turn off >> canonicalization. >Sure it is. You can use the [host.do.main] syntax [and run an authoritative >local nameserver for some zones]. You're right. This can be useful for some situations. It still doesn't solve the general problem, however, since it's hard to predict who might decide to send you mail. -ccwf Al final, quedara' solamente uno. -- Juan Sanchez Villa-Lobos Ramirez (In the end, there can be only Juan.)
From: JOB: Software Engineers for animation software company in Los Angeles Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.software Subject: Re: (some) NeXT apps can't print on non-next printer Date: 26 Oct 1995 08:34:43 GMT Organization: Delft University of Technology Message-ID: <46nh73$m0c@mo6.rc.tudelft.nl> References: <J.M.FIGUEROA.95Oct25205640@strings7.ph.qmw.ac.uk> The postscript error messages you listed are typically messages that indicate that the applications try to use an unknown page format. So, change the page format of the applications to something the printer does recognize. Greetings, Abraham G.
From: J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) Newsgroups: comp.sys.next.bugs Subject: Re: (some) NeXT apps can't print on non-next printer Date: 26 Oct 1995 17:29:16 GMT Organization: Theory Group, Physics, QMW College Message-ID: <J.M.FIGUEROA.95Oct26172916@strings7.ph.qmw.ac.uk> References: <J.M.FIGUEROA.95Oct25205640@strings7.ph.qmw.ac.uk> <1995Oct25.230912.7477@seer.demon.co.uk> In-reply-to: Paul Lynch's message of Wed, 25 Oct 1995 23:09:12 GMT >>>>> "Paul" == Paul Lynch <Paul_Lynch@plsys.com> writes: Paul> In article <J.M.FIGUEROA.95Oct25205640@strings7.ph.qmw.ac.uk> Paul> J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) writes: >> >> We recently connected an HP LaserJet 4MP (600DPI) Postscript printer >> to our next network. The machines seem to recognise the printer type >> because it shows a printer definition file for it. There are no >> problems printing PS files generated from TeX, or text files. However >> printing from some other NeXT apps doesn't work. Instead of the >> expected output, one gets a page with the following message: >> >> ERROR: configurationerror >> OFFENDING COMMAND: setpagedevice >> >> STACK: >> >> >> Any ideas? Paul> For not very obvious reasons, LaserJets get upset if you try to print a Paul> Letter sized document to them when the printer is set to A4. Make sure Paul> the document paper size and the printer setting match. Paul> Paul Paul> -- Paul> Paul Lynch (NeXTmail) Paul> http://www.plsys.co.uk/~paul Paul> (under construction) Thanks! This works. Now, is there a way to set the paper size to A4 by default? All apps seem to be in Letter size and it's a pain to have to go through them all to change them. Cheers, Jose -- +----------------------------------+----------------------------------+ | Dr Jose M Figueroa-O'Farrill | Vox: +44.171.975-5055 | | Department of Physics (Rm. 227) | Fax: +44.181.981-7465 | | Queen Mary and Westfield College | mailto:J.M.Figueroa@qmw.ac.uk | | Mile End Road, London E1 4NS, UK | | | and | | http://stringswww.ph.qmw.ac.uk/~jmf/jmf.html | +----------------------------------+----------------------------------+
From: Paul Lynch <Paul_Lynch@plsys.com> Newsgroups: comp.sys.next.bugs Subject: Re: (some) NeXT apps can't print on non-next printer Date: Thu, 26 Oct 1995 23:51:03 GMT Organization: P & L Systems Sender: news@seer.demon.co.uk Message-ID: <1995Oct26.235103.10615@seer.demon.co.uk> References: <J.M.FIGUEROA.95Oct26172916@strings7.ph.qmw.ac.uk> In article <J.M.FIGUEROA.95Oct26172916@strings7.ph.qmw.ac.uk> J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) writes: > Thanks! This works. Now, is there a way to set the paper size to A4 > by default? All apps seem to be in Letter size and it's a pain to > have to go through them all to change them. If you have 3.3 or better, there is an option in Preferences.app to set default page size. Not all apps pay attention to this, and may need setting independently. Older OS releases can use the dwrite GLOBAL NXPaperType A4. Paul -- Paul Lynch (NeXTmail) http://www.plsys.co.uk/~paul (under construction)
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <476TJL@ptsmail.ptsem.edu> From: 476TJL@ptsmail.ptsem.edu (LUOMA,TIMOTHY) Cc: J.M.Figueroa@qmw.ac.uk Subject: Re: (some) NeXT apps can't print on non-next printer Message-ID: <30A0903001A23A7C@-SMF-> Date: 27 Oct 95 09:02:00 EDT ( J.M.Figueroa@qmw.ac.uk wrote ) > Thanks! This works. Now, is there a way to set the > paper size to A4 > by default? All apps seem to be in Letter size and it's a pain > to have to go through them all to change them. (Paul_Lynch@plsys.com replied) If you have 3.3 or better, there is an option in Preferences.app to set default page size. Not all apps pay attention to this, and may need setting independently. Older OS releases can use the dwrite GLOBAL NXPaperType A4. (now I add:) if you have the OS less than 3.3, there is a preference module out there to do this too: ftp://peanuts.leo.org/pub/comp/platforms/next/Tools/preferences/PaperType.1 .1.NI.b.tar.gz (size: 41115) FTP this and drop in /LocalLibrary/Preferences Or get this: ftp://peanuts.leo.org/pub/comp/platforms/next/Tools/preferences/GKMouseScal er.1.1.NI.b.tar.gz (size: 19195) Which sets the dwrite for the paper size and sets preferences for the mouse scaling (how the mouse moves on the screen according to how you move it on the mousepad) and the dwrite which lets you move your dock over to the left side rather than the right. TjL -- Timothy J. Luoma <476tjl@ptsmail.ptsem.edu> Email: ASCII preferrred / NeXTMail for necessary attachments / NO MIME! YOUR 'REPLY-TO' LINE WILL BE REMOVED BY MY STUPID Microsoft MAILREADER!! PLEASE TELL ME IF YOUR "FROM" ADDRESS IS NOT YOUR PREFERRED EMAIL ADDRESS.
Newsgroups: comp.sys.next.bugs Subject: Re: (some) NeXT apps can't print on non-next prin Message-ID: <1995Oct27.110801.45567@yogi.urz.unibas.ch> From: frank@ifi.unibas.ch Date: 27 Oct 95 11:08:00 MET References: <J.M.FIGUEROA.95Oct25205640@strings7.ph.qmw.ac.uk> <1995Oct25.2309 <J.M.FIGUEROA.95Oct26172916@strings7.ph.qmw.ac.uk> J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) wrote: > >>>>> "Paul" == Paul Lynch <Paul_Lynch@plsys.com> writes: > Paul> In article <J.M.FIGUEROA.95Oct25205640@strings7.ph.qmw.ac.uk> [stuff about postscript errors] > Paul> For not very obvious reasons, LaserJets get upset if you try to print a Paul> Letter sized document to them when the printer is set to A4. Make sure Paul> the document paper size and the printer setting match. > Paul> Paul > Paul> -- > Paul> Paul Lynch (NeXTmail) > Paul> http://www.plsys.co.uk/~paul > Paul> (under construction) > Thanks! This works. Now, is there a way to set the paper size to A4 by default? All apps seem to be in Letter size and it's a pain to have to go through them all to change them. > Cheers, > Jose You bet it is. But NS version 3.3 seems to be a little bit better. You can set the paper size in preferences. In most cases, however, it seems to be a problem of the applications not properly querying the papersize from the printer panel. For pre 3.3 versions, setting 'dwrite GLOBAL NXPaperType A4' as root helps in many, but alas not all, cases. It took NeXT software engineers a too damm ('scuse me) long time to realize that USletter, inches etc. are more or less restricted to the US ... -- it will take even longer for the other software engineers to fully aknowledge this. Sigh, I just had to say this again ... -Robert -- Institut fuer Informatik tel +41 (0)61 321 99 67 Universitaet Basel fax. +41 (0)61 321 99 15 Robert Frank Mittlere Strasse 142 rfc822: frank@ifi.unibas.ch (NeXT,MIME mail ok) CH-4056 Basel X400: S=frank;OU=ifi;O=unibas;P=switch;A=arcom;C=ch Switzerland
From: J.M.Figueroa@QMW.ac.uk (Jose Miguel Figueroa-O'Farrill) Newsgroups: comp.sys.next.bugs Subject: Re: [Fixed] (some) NeXT apps can't print on non-next printer Date: 27 Oct 1995 22:43:37 GMT Organization: Theory Group, Physics, QMW College Message-ID: <J.M.FIGUEROA.95Oct27224337@strings7.ph.qmw.ac.uk> References: <J.M.FIGUEROA.95Oct25205640@strings7.ph.qmw.ac.uk> <1995Oct25.230912.7477@seer.demon.co.uk> <J.M.FIGUEROA.95Oct26172916@strings7.ph.qmw.ac.uk> In-reply-to: J.M.Figueroa@QMW.ac.uk's message of 26 Oct 1995 17:29:16 GMT >>>>> "Me" == Jose Miguel Figueroa-O'Farrill <J.M.Figueroa@QMW.ac.uk> >>>>> "Paul" == Paul Lynch <Paul_Lynch@plsys.com> Me> We recently connected an HP LaserJet 4MP (600DPI) Postscript printer Me> to our next network. The machines seem to recognise the printer type Me> because it shows a printer definition file for it. There are no Me> problems printing PS files generated from TeX, or text files. However Me> printing from some other NeXT apps doesn't work. Instead of the Me> expected output, one gets a page with the following message: Me> Me> ERROR: configurationerror Me> OFFENDING COMMAND: setpagedevice Me> Me> STACK: Me> Me> Me> Any ideas? Paul> For not very obvious reasons, LaserJets get upset if you try to print a Paul> Letter sized document to them when the printer is set to A4. Make sure Paul> the document paper size and the printer setting match. Me> Thanks! This works. Now, is there a way to set the paper size to A4 Me> by default? All apps seem to be in Letter size and it's a pain to Me> have to go through them all to change them. The problem has been fixed. The printer's paper size is indeed A4, but NeXT apps default to Letter size. So all I need to do was dwrite GLOBAL NXPaperType A4 and it all sems to work now. However I still don't understand why the PPD file for the printer contains "setpagedevice", yet the printer doesn't seem to understand this command. Is there a newer PPD file for the HP LaserJet 4M+ than /NextLibrary/PrinterTypes/English.lproj/HP_LaserJet_4_PostScript_600DPI.ppd??? We're all very grateful here to those who answered. Thanks to you all, Jose -- +----------------------------------+----------------------------------+ | Dr Jose M Figueroa-O'Farrill | Vox: +44.171.975-5055 | | Department of Physics (Rm. 227) | Fax: +44.181.981-7465 | | Queen Mary and Westfield College | mailto:J.M.Figueroa@qmw.ac.uk | | Mile End Road, London E1 4NS, UK | | | and | | http://stringswww.ph.qmw.ac.uk/~jmf/jmf.html | +----------------------------------+----------------------------------+
From: joel (Joel Stalder) Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Trouble with HP Drive on Intel Hardware - Thoughts? Date: 29 Oct 1995 17:42:23 GMT Organization: Psyberware Internet Access Message-ID: <470edv$3km@warez.psyber.com> Sirs and Madams, I am appparantely having a bit of trouble configuring a Hewlett Packard C3724S, that I have recently purchased. I have now come to the point that I believe that the problem lies with NS. The problem: The drive proports to be 1149 mb, on the packaging box and according to the SCSI Interface. However, NS believes it to be 1023 mb and no matter what I seem to do, it wont give me my 126 mb! I am running N.S. 3.2 for Intel on a 486DX4-100 with an adaptec 1540/1542 SCSI adapter card. I there a bug in NS 3.2 that doesn't "deal with" more than 1 gig per drive, or something? Regardless, I would apreciate any thoughts that you might have. thanks, joel
From: trail@ix.netcom.com (Jeff Trestrail ) Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Re: Trouble with HP Drive on Intel Hardware - Thoughts? Date: 29 Oct 1995 22:36:13 GMT Organization: Netcom Message-ID: <470vkt$ql3@ixnews7.ix.netcom.com> References: <470edv$3km@warez.psyber.com> In <470edv$3km@warez.psyber.com> joel (Joel Stalder) writes: > >I am appparantely having a bit of trouble configuring a Hewlett >Packard C3724S, that I have recently purchased. I have now come >to the point that I believe that the problem lies with NS. >The problem: The drive proports to be 1149 mb, on the packaging box >and according to the SCSI Interface. However, NS believes it to be >1023 mb and no matter what I seem to do, it wont give me my 126 mb! > >I am running N.S. 3.2 for Intel on a 486DX4-100 with an adaptec >1540/1542 SCSI adapter card. > >I there a bug in NS 3.2 that doesn't "deal with" more than 1 gig per >drive, or something? You don't mention whether you're using the drive to boot off, but if you're not, then using the -useAllSectors flag with fdisk will permit access to all of your disk. Regards, Jeff Trestrail trail@ix.netcom.com
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 30 Oct 1995 05:15:11 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <471n0v$6i4@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: gfin@psych.ualberta.ca (Gary Finley) Newsgroups: comp.sys.next.bugs Subject: changing SCSI drivers Date: 31 Oct 1995 04:33:52 GMT Organization: University of Alberta, Edmonton, Canada Message-ID: <4748vg$12pe@pulp.ucs.ualberta.ca> I could use some expert advice on how to handle a problem I have with changing the SCSI driver on my home machine, a 486/66 with NCR 53C810 SCSI controller on the ASUS motherboard. I've been using the Talus driver at both home and work (where I have a P90 system with the same NCR controller on a PCI card). Both machines are running NS 3.2. To solve some problems with recognition of CD drives and tape drives, I upgraded the work machine's SCSI driver to the SYMbios logic one (version 3.0). The swap of drivers at work was no problem. At home it hasn't been so easy. After I install the new driver (either from a hard disk copy, or from one in /usr/Devices on a floppy). The computer won't boot without giving me the "Insert the floppy disk containing the new driver..." message. I know about the 3.2 Configure.app bug that requires you to save, exit, and re-run Configure after you've removed an old driver that you want to replace. I did all that. When I was done, the System.config/Instance0.table's Boot Drivers line contained "SYM53c8" as it should. The Instance table was gone from NCR53C810.config, and SYM53c8.config had an Instance0.table as it should. I thought I'd done it all right, but the machine *still* insists that I insert the floppy containing the SYM53c8 driver before it would boot. The only way that I can get a floppy-less boot it to revert to the Talus driver, which won't recognize the tape drive that I need to use to make backups. Does anyone know why this happens, and what I can do about it? Any advice would be very welcome. -- -------------------------------------------- Gary Finley, Univ. of Alberta Psychology Dept. gfin@psych.ualberta.ca (NeXTmail welcome!)
From: xdtp@python.viper.net (DTP Products) Newsgroups: control,comp.sys.next.bugs Subject: FREE $$$ MAKING SOFTWARE !!! Date: 31 Oct 1995 15:21:56 -0600 Organization: ViperNet Message-ID: <47641k$ktv@python.viper.net> The NEWEST and most INNOVATIVE SERVER on the Internet. The FIRST & ONLY full Internet Access Provider that gives you the opportunity to make BIG $$$ simply by being a subscriber. YOU KNOW: How incredibly large servers like Prodigy and AOL have become and how much bigger they continue to grow everyday. JUST IMAGINE: If you had been given the opportunity to be in on the ground floor of one of those servies. You would now have tens of thousands, if not hundreds of thousands of users signed under you. Which simply means YOU would now be VERY, VERY WEALTHY!!! And you get 5- HOURS OR 3-DAYS **FREE** TO TRY IT OUT at NO COST TO YOU !!! It Has Everything that the BIG 3 have and MUCH MORE, Plus full 28,800 Access. And The Best Graphics On The Net. If you're going to use a provider, use one that MAKES YOU MONEY And gives you 5-FREE ONLINE HOURS EVERY MONTH. And if you're already in business on the net this is the easiest way to add your income, by simply enclosing a free Demo Disc in every deliver you make to your customers. ------------------------------------------------------------------------------------------- Respond To: dtp@mindspring.com Movieman5@aol.com Or Call 1-334-973-9721 ------------------------------------------------------------------------------------------- For ABSOLUTELY FREE GRAPHICS & SOUND DEMO-DISC Send Your Mailing Address. No Cost. No Obligation. No Bull. Your address will not be used for any other purpose. If you're going to fall.....fall reaching
From: robin@clangunn. (Robin D. Wilson) Newsgroups: comp.sys.next.bugs Subject: Anyone using a Pentium "Overdrive"? Date: 31 Oct 1995 04:57:26 GMT Organization: Pencom Software Message-ID: <474abm$jct@digdug.pencom.com> I have an 83Mhz Pentium "Overdrive" CPU that I stuck in my (former) 486-DX33 system. It _does_ run much faster, but NEXTSTEP is virtually unusable since it also crashes frequently. I get a combination of "hangs", panics ("unknown kernel trap d" is the message), and (most frequent) Workspace manager errors. The frequency is on the order of a few minutes per occurance (one of the above will occur). Swapping the system back to the old DX33 clears the problem right up. (Also, "Preferences.app" won't stay up when I select the "Mouse" icon if I'm running with the Overdrive processor -- but it runs fine on the DX33?!) Has _ANYONE_ else experienced similar problems (perhaps I just have a bad Overdrive?)? Please let me know (so I can decide if this is a processor problem, or a NEXTSTEP problem). (As an aside, Windows also has been somewhat strange as well... If I leave it sit for about 2 hours -- I can't re-access it (it is hung). Even the three-finger salute doesn't work. But with the DX33, it is fine. This is the _only_ unusual symptom in Windows tho'...) If it helps, this is my system config: VLB motherboard. ZIF socket (supposedly Overdrive "ready") 20MB RAM Tseng Labs ET4000w32p Video (the only VESA card in the box) Adaptec 1542c SCSI (ISA bus) IDE drive (DOS only - I boot from floppy) SCSI drives (NEXTSTEP, Linux and DOS on separate drives) Soundblaster Pro (not used in NEXTSTEP) (ISA bus) Don't know what else might matter... The diskette that was supposed to verify the Overdrive once installed said that it passes _all_ tests (I ran it through all of the available tests provided). Please help! -- ----------------------------------------------------------------------------- *** These are my opinions... Mine! All Mine! Minemineminemineminemine! *** ----------------------------------------------------------------------------- Robin D. Wilson robin@pencom.com Pencom Software 701 Canyon Bend Dr. 9050 Capital of Texas Hwy Pflugerville, TX 78660 Austin, TX 78759 (512) 251-1737 (512) 343-6666
From: "John K. Troxell" <jktroxe@erenj.com> Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Re: Trouble with HP Drive on Intel Hardware - Thoughts? Date: Tue, 31 Oct 95 15:48:58 PST Organization: Exxon R&E, NJ; Opinions not those of the Company. Distribution: inet Message-ID: <4762ic$24g@maverick1.erenj.com> To: Joel Stalder <joel> In-Reply-To: Your message of 29 Oct 1995 17:42:23 GMT.<470edv$3km@warez.psyber.com> Encoding: 39 TEXT I think I know what is causing this. If so, it is easy to solve. This is not a NEXTSTEP problem. It is a feature of the card and it is related to a feature of DOS. You need to enable "extended bios translation" on the Adaptec in order to see more than 1 gig. You don't mention what version of the card you have. If it is a 1542C or 1542CF then you can do this in the card's setup menu. If it is a 1542B then you can set this via a jumper on the card, provided you have the correct Bios & Microcode EPROMS on the card (I think its BIOS version 3.2 or greater). If you have an earlier BIOS (check the message on screen when you boot), you can call Adaptec and buy (or beg) new EPROMS. It is easy to pop the old ones out and plug in the new ones. Or you can reburn the existing ones if you have an EPROM burner(!). The code for this is downloadable from the Adaptec bbs. I have used extended bios translation with NEXTSTEP 3.2 and 3.3. >From: joel (Joel Stalder)> >Sirs and Madams, > >I am appparantely having a bit of trouble configuring a Hewlett Packard >C3724S, that I have recently purchased. I have now come to the point that I >believe that the problem lies with NS. > >The problem: The drive proports to be 1149 mb, on the packaging box and >according to the SCSI Interface. However, NS believes it to be 1023 mb and >no matter what I seem to do, it wont give me my 126 mb! > >I am running N.S. 3.2 for Intel on a 486DX4-100 with an adaptec 1540/1542 >SCSI adapter card. > > >I there a bug in NS 3.2 that doesn't "deal with" more than 1 gig per drive, >or something? >
Date: 1 Nov 1995 00:58:15 GMT From: xdtp@python.viper.net (DTP Products) Message-ID: <cancel.47641k$ktv@python.viper.net> Newsgroups: control,comp.sys.next.bugs Subject: cmsg cancel <47641k$ktv@python.viper.net> Control: cancel <47641k$ktv@python.viper.net> Spam cancelled by clewis@ferret.ocunix.on.ca
From: jq@phcs.phcs.com (James E. Quick) Newsgroups: comp.sys.next.bugs Subject: Re: Anyone using a Pentium "Overdrive"? Date: 1 Nov 1995 07:45:22 -0500 Organization: PHCS Message-ID: <477q52$84d@papoose.quick.com> References: <474abm$jct@digdug.pencom.com> In article <474abm$jct@digdug.pencom.com>, Robin D. Wilson <robin@clangunn.> wrote: >I have an 83Mhz Pentium "Overdrive" CPU that I stuck in my (former) >486-DX33 system. It _does_ run much faster, but NEXTSTEP is >virtually unusable since it also crashes frequently. I get a >combination of "hangs", panics ("unknown kernel trap d" is the >message), and (most frequent) Workspace manager errors. The >frequency is on the order of a few minutes per occurance (one of >the above will occur). Swapping the system back to the old DX33 >clears the problem right I upgraded a system from DX2-66 to 83Mhz Pentium overdrive almost 2 weeks ago. I initially had similar problems. I was able to boot NS and it would operate normally for about 1 day at a time. When it would fail, the window server would die, and my PNI slip code would be killed and not restarted. After experimenting with the CMOS settings, I believe I have solved the problem, but I need another few days of burn in time before I know for sure. My problem had to do with bus timing. My bus was previously clocked at CPUCLOCK/4 by setting the bus clock to 7.75 Mhz, the problem went away. I lost about 2% of the potential speedup that I saw when I was crashing, but still gained about 29% raw integer increase and about 38% overall improvement as reported by NXBench.app. My system is EISA VLBUS. I don't know whether or not you can tune the BUSCLOCK on an ISA system or whether that is just for EISA systems. Another thing you might try is the CAS pulse width and cache settings in CMOS. It could be that something in CMOS is assuming either 33 or 66 Mhz operation of the CPU, and is thus bombing because the chip is clocked 2.5% the system clock. -- ___ ___ | James E. Quick jq@phcs.com / / / | Private HealthCare Systems NeXTMail O.K. \_/ (_\/ | Systems Integration Group (617) 895-3343 ) | - My other car has a mouse.
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: changing SCSI drivers Date: 1 Nov 1995 11:08:32 GMT Organization: Technical University of Berlin, Germany Message-ID: <477kfg$m1d@news.cs.tu-berlin.de> References: <4748vg$12pe@pulp.ucs.ualberta.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Gary Finley (gfin@psych.ualberta.ca) wrote: : I could use some expert advice on how to handle a problem I have : with changing the SCSI driver on my home machine, a 486/66 with : NCR 53C810 SCSI controller on the ASUS motherboard. [...] : I thought I'd done it all right, but the machine *still* : insists that I insert the floppy containing the SYM53c8 driver : before it would boot. The only way that I can get a floppy-less : boot it to revert to the Talus driver, which won't recognize the : tape drive that I need to use to make backups. I remember coming across a bug in the scan for drivers which would not find a driver if there are too many drivers in /usr/Devices. The solution was to remove all unnecessary drivers, move /usr/Devices to another name and rebuild it via cp -rp (to shorten the directory). Hope that helps. Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
Message-ID: <46urn9$fdh@emerald.oz.net> Organization: Sense Networking Seattle (www.oz.net) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit From: art@cubicsol.com (Art Isbell) Subject: gnutar 1.11.2 (NS 3.3) bug Date: 29 Oct 1995 02:16:57 +0000 Newsgroups: comp.sys.next.bugs Distribution: world I was dismayed to find that upon restoring a bunch of nibs from a backup created using gnutar, many of the nibs were empty (i.e., no data.nib, data.classes, or data.dependencies). I had used the following to restore the nibs: gnutar -xzpf backup.tar.gz Users.external/art/Trego/Development/CaseServ/CaseServ.debug/English.lproj Note that by default, gnutar strips the leading '/' off of paths. When I examined the contents of backup.tar.gz, many entries similar to the following were printed: Users.external/art/Trego/Development/CaseServ/CaseServ.debug/English.lproj /AddressType.nib/ /Users.external/art/Trego/Development/CaseServ/CaseServ.debug/English.lpro j/AddressType.nib/data.classes /Users.external/art/Trego/Development/CaseServ/CaseServ.debug/English.lpro j/AddressType.nib/data.nib Note that the data.classes and data.nib entries haven't had the leading '/' stripped!! Those paths which haven't had the leading '/' stripped are those whose length exceeds 100 characters so are stored specially by gnutar. Although this is a gnutar, not NEXTSTEP, bug, does anyone know whether a later version of gnutar fixes this bug? --- Art Isbell NeXTmail: art@cubicsol.com NeXT Registered Consultant Voice/Fax: +1 408 335 2515 Trego Systems Voice Mail: +1 408 335 1154 CaseServ: NEXTSTEP managed care US Mail: Felton, CA 95018-9442 contract and case management solutions
From: robert@steffi.accessone.com (Robert Nicholson) Newsgroups: comp.sys.next.bugs Subject: sort bug... Date: 02 Nov 1995 01:51:54 -0800 Organization: me organized? That's a joke! Sender: robert@steffi.accessone.com Message-ID: <en3afwdhh.fsf@steffi.accessone.com> Anybody explain this? (3.3) root:/>sort -T /LocalLibrary 1:50 root:/>sort -T /External1/nextstep/LocalLibrary 1:50 zsh: segmentation fault sort -T /External1/nextstep/LocalLibrary root:/> 1:50 lrwxrwxrwx 1 root 32 Apr 2 1995 LocalLibrary -> /External1/nextstep/LocalLibrary Guess it's time to fetch gnu sort. -- "Mary ate a little lamb and punk rock isn't dead" (PGP key: send email with Subject: request pgp key)
From: gfin@psych.ualberta.ca (Gary Finley) Newsgroups: comp.sys.next.bugs Subject: Changing SCSI drivers: Fixed! Date: 3 Nov 1995 03:34:24 GMT Organization: University of Alberta, Edmonton, Canada Message-ID: <47c2k0$omc@pulp.ucs.ualberta.ca> I posted a couple of days ago asking for help with a problem in changing SCSI drivers. NS 3.2 kept asking me to insert the floppy containing the new driver that I'd just installed. Well I got the expert advice I was looking for: Axel Habermann suggested that there might be a problem if there were too many drivers in the /usr/Devices directory. He suggested that I try this: (1) Delete any drivers that I didn't need. (2) Rename /usr/Devices to Devices.old (3) Make a new dir /usr/Devices (4) cp -rp everything from /usr/Devices.old to /usr/Devices Apparently it's not sufficient to just delete drivers, because that wouldn't make the directory any shorter (I wouldn't have known this myself). I did these things, and re-installed my SYM SCSI driver. The machine booted up on the new driver right away, and no mention was made of any (ugh) floppy disks. Thanks a bunch, Axel! -- -------------------------------------------- Gary Finley, Univ. of Alberta Psychology Dept. gfin@psych.ualberta.ca (NeXTmail welcome!)
From: shill@iphysiol.unil.ch (Sean Hill) Newsgroups: comp.sys.next.bugs,comp.sys.next.misc Subject: NEXTSTEP 3.3 patch (Is it ready?) Date: 3 Nov 1995 17:03:41 GMT Organization: University of Lausanne CH (Switzerland) Message-ID: <47di1d$hhd@cisun2000.unil.ch> I was told that the NEXTSTEP 3.3 patch (fixing lookupd) would be released to the public on November 1st. Can we expect it on the ftp site soon? Thank you! Sean Hill --- Sean L. Hill Research in Computational Neuroscience Institut de Physiologie E-mail: Sean.Hill@iphysiol.unil.ch Rue du Bugnon, 7 Work: ++41 021 692.5516 CH-1005 Lausanne SWITZERLAND Fax: ++41 021 692.5505
From: spotter@netcom.com (Steve Potter) Sender: spotter@netcom22.netcom.com Newsgroups: comp.sys.next.bugs Date: 03 Nov 1995 18:24:29 EST Control: cancel <spotterDHGq2o.6yC@netcom.com> Subject: cmsg cancel <spotterDHGq2o.6yC@netcom.com> Message-ID: <cancel.spotterDHGq2o.6yC@netcom.com> Spam cancelled by dsr@lns598.lns.cornell.edu
From: kkudielk@cacofonix.nt.tuwien.ac.at (Klaus Kudielka) Newsgroups: comp.sys.next.bugs,comp.unix.solaris Subject: HELP: Solaris 2.4 & NextStep 3.1/3.3 & NFS Date: 4 Nov 1995 11:03:05 GMT Organization: Vienna University of Technology, Austria Distribution: inet Message-ID: <47fh99$klb@news.tuwien.ac.at> Keywords: Nextstep Solaris NFS Hello, I'm posting this both to the NeXT group and the Solaris group, since I don't know who's responsible for this. When mounting a Nextstep/Intel 3.1 (or 3.3) partition from Solaris/SPARC 2.4 and just doing a 'ls', the command hangs and must be aborted with Ctrl-C. The Nextstep server logs the following error messages: Nov 4 11:31:04 europa mach: xdr_bytes: bad size FAILED Nov 4 11:31:04 europa mach: svckudp_send: xdr_replymsg failed Nov 4 11:31:04 europa mach: nfs_server: bad sendreply from 128.130.67.30 [ These three messages are logged for each NFS request ] The Solaris machine has the 2.4_Recommended patches applied. I've searched sun-fixes for any additional NFS patch --- no success. The problem does not exist with earlier versions of Solaris. QUESTIONS: * Does anybody else have this problem? * Any hint which OS might be responsible for this? * Is there a fix available? Thanks in advance Klaus -- Dipl.-Ing. Klaus Kudielka Institut fuer Nachrichtentechnik und HF-Technik Phone: (+43 1) 58801-3556 Technische Universitaet Wien Fax: (+43 1) 5870583 Gusshausstr. 25/389, A-1040 Wien, AUSTRIA kkudielk@cacofonix.nt.tuwien.ac.at http://cacofonix.nt.tuwien.ac.at/~kkudielk/
From: stabl@informatik.uni-muenchen.de (Robert Stabl) Newsgroups: comp.sys.next.bugs,comp.sys.next.misc Subject: Re: NEXTSTEP 3.3 patch (Is it ready?) Date: 4 Nov 1995 16:08:51 GMT Organization: Institut fuer Informatik der Universitaet Muenchen Message-ID: <47g36j$1a6@arcadia.informatik.uni-muenchen.de> References: <47di1d$hhd@cisun2000.unil.ch> shill@iphysiol.unil.ch (Sean Hill) writes: >I was told that the NEXTSTEP 3.3 patch (fixing lookupd) would be released to >the public on November 1st. Can we expect it on the ftp site soon? Yes, you can. Just look at ftp.next.com or at peanuts.leo.org. Robert. -- Robert Stabl email: stabl@informatik.uni-muenchen.de Dept. of Computer Science stabl@leo.org University of Munich http://www.pst.informatik.uni-muenchen.de/~stabl/ Leopoldstr. 11B D-80802 Muenchen Tel: +(49) 89 2180 6316 Germany FAX: +(49) 89 2180 6310
From: jq@phcs.phcs.com (James E. Quick) Newsgroups: comp.sys.next.bugs,comp.sys.next.misc Subject: Re: NEXTSTEP 3.3 patch (Is it ready?) Date: 4 Nov 1995 12:18:29 -0500 Organization: PHCS Message-ID: <47g795$ln@papoose.quick.com> References: <47di1d$hhd@cisun2000.unil.ch> In article <47di1d$hhd@cisun2000.unil.ch>, Sean Hill <Sean.Hill@iphysiol.unil.ch> wrote: >I was told that the NEXTSTEP 3.3 patch (fixing lookupd) would be released to >the public on November 1st. Can we expect it on the ftp site soon? The patches were placed on the server yesterday. ftp://ftp.next.com/pub/Patches/NEXTSTEP3.3Patch1/ By the number of ftp connections going at the moment, I think 10 people are downloading it at this moment. -- ___ ___ | James E. Quick jq@phcs.com / / / | Private HealthCare Systems NeXTMail O.K. \_/ (_\/ | Systems Integration Group (617) 895-3343 ) | - My other car has a mouse.
From: Robert Worne <rworne@primenet.com> Newsgroups: comp.sys.next.bugs,comp.sys.next.misc Subject: Re: NEXTSTEP 3.3 patch (Is it ready?) Date: 4 Nov 1995 17:39:39 GMT Organization: A Big Black Cube Message-ID: <47g8gr$onc@nnrp1.news.primenet.com> References: <47di1d$hhd@cisun2000.unil.ch> shill@iphysiol.unil.ch (Sean Hill) wrote: > > > I was told that the NEXTSTEP 3.3 patch (fixing lookupd) would be released to > the public on November 1st. Can we expect it on the ftp site soon? It IS out... ftp://ftp.next.com/pub/Patches/NEXTSTEP3.3Patch1/ -- //--------------------------------------------------------------------- // Robert Worne rworne@primenet.com OS/2-NeXT -=Starving CS Undergrad=- //--------------------------------------------------------------------- // Microsoft Network is prohibited from redistributing this work in // any form, in whole or in part. Copyright, Robert Worne, 1995. // License to distribute this post is available to Microsoft for $1000. // Posting without permission constitutes an agreement to these terms. // Please send notices of violation to rworne@primenet.com and // postmaster@microsoft.com
From: spotter@netcom.com (Steve Potter) Sender: spotter@netcom11.netcom.com Newsgroups: comp.sys.next.bugs Date: 04 Nov 1995 12:47:13 EST Control: cancel <spotterDHH89r.Esz@netcom.com> Subject: cmsg cancel <spotterDHH89r.Esz@netcom.com> Message-ID: <cancel.spotterDHH89r.Esz@netcom.com> Spam cancelled by dsr@lns598.lns.cornell.edu
From: Paul Lynch <Paul_Lynch@plsys.com> Newsgroups: comp.sys.next.bugs Subject: Re: HELP: Solaris 2.4 & NextStep 3.1/3.3 & NFS Date: Sun, 5 Nov 1995 00:01:22 GMT Organization: P & L Systems Sender: news@seer.demon.co.uk Distribution: inet Message-ID: <1995Nov5.000122.5518@seer.demon.co.uk> References: <47fh99$klb@news.tuwien.ac.at> In article <47fh99$klb@news.tuwien.ac.at> kkudielk@cacofonix.nt.tuwien.ac.at (Klaus Kudielka) writes: > When mounting a Nextstep/Intel 3.1 (or 3.3) partition from Solaris/SPARC 2.4 > and just doing a 'ls', the command hangs and must be aborted with Ctrl-C. > The Nextstep server logs the following error messages: > > Nov 4 11:31:04 europa mach: xdr_bytes: bad size FAILED > Nov 4 11:31:04 europa mach: svckudp_send: xdr_replymsg failed > Nov 4 11:31:04 europa mach: nfs_server: bad sendreply from 128.130.67.30 This is a faq of some kind. The fix (I won't attempt to place blame) is to set block sizes to 1K. This should be done on the exporting server. Paul -- Paul Lynch (NeXTmail) http://www.plsys.co.uk/~paul (under construction)
From: kkudielk@cacofonix.nt.tuwien.ac.at (Klaus Kudielka) Newsgroups: comp.sys.next.bugs,comp.unix.solaris Subject: Re: HELP: Solaris 2.4 & NextStep 3.1/3.3 & NFS Followup-To: comp.sys.next.bugs,comp.unix.solaris Date: 5 Nov 1995 11:06:28 GMT Organization: Vienna University of Technology, Austria Distribution: inet Message-ID: <47i5rk$n9g@news.tuwien.ac.at> References: <47fh99$klb@news.tuwien.ac.at> Klaus Kudielka (kkudielk@cacofonix.nt.tuwien.ac.at) wrote: : When mounting a Nextstep/Intel 3.1 (or 3.3) partition from Solaris/SPARC 2.4 : and just doing a 'ls', the command hangs and must be aborted with Ctrl-C. Uhmm... Sorry... This is a Solaris FAQ (Section 5.11, to be exact). The point is that Solaris 2.4 NFS is not "bug-compatible" with older NFS versions. Thanks to all who replied. -Klaus -- Dipl.-Ing. Klaus Kudielka Institut fuer Nachrichtentechnik und HF-Technik Phone: (+43 1) 58801-3556 Technische Universitaet Wien Fax: (+43 1) 5870583 Gusshausstr. 25/389, A-1040 Wien, AUSTRIA kkudielk@cacofonix.nt.tuwien.ac.at http://cacofonix.nt.tuwien.ac.at/~kkudielk/
From: wilkie@narya. (Alexander Wilkie) Newsgroups: comp.sys.next.bugs Subject: nslookup broken in 3.3 patch? Date: 5 Nov 1995 13:41:27 GMT Organization: Vienna University of Technology, Austria Message-ID: <47ieu7$pju@news.tuwien.ac.at> Hi everyone! I just got the 3.3 patch from www.next.com, and unpacked/installed the lot (JFTR I'm running NS3.3/Intel). To be on the safe side of things, I ran the "posix-app-friendly" post-install script. Something I did NOT do was to download and install the new libposix.a. After rebooting, nslookup stopped working. When called, it just plays dead (i.e. no output at all, and no crash either) until I quit it with Ctrl-C. This is odd, since the DNS service appears to be working for other apps & programs. -> ? Please comment! Alexander Wilkie wilkie@cg.tuwien.ac.at -- e-mail: wilkie@cg.tuwien.ac.at (NeXTMail, MIME preferred) www : http://www.cg.tuwien.ac.at/~wilkie/
Message-ID: <spotterDHH89r.Esz@netcom.com> Organization: NETCOM On-line Communication Services (408 261-4700 guest) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit From: spotter@netcom.com (Steve Potter) Subject: 809-474-7607 ! Local Voice Personals/Live Chat! Code3239 Date: 03 Nov 1995 15:53:51 +0000 Newsgroups: comp.sys.next.bugs Distribution: world You must be 18 years of age or older. As low as $0.23/min Available 24 hours a day! ALL LIFESTYLES welcome comp.sys.next.bugs
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.hardware From: liu@turbo.physics (Wei Liu) Subject: Binary data file IO error Keywords: binary, IO, intel, m68k Message-ID: <DHJzx6.3IE@info.physics.utoronto.ca> Sender: news@info.physics.utoronto.ca (System Administrator) Organization: University of Toronto - Dept. of Physics Date: Sun, 5 Nov 1995 04:46:18 GMT Anyone knows what could explain the following, I found I couldn't read those binary data files produced on other unix platforms by my intel Next workstation. To test this, I wrote one code to produce the data, the other reading the data. #include <stdlib.h> #include <stdio.h> int main() { FILE *fp; int i; float x; double z; i=10; x=1.234567; z=1.23456789101112131415161718; fp=fopen("bytetest.dat", "w"); fwrite(&i, sizeof(int), 1, fp); fwrite(&x, sizeof(float), 1, fp); fwrite(&z, sizeof(double), 1, fp); fclose(fp); return 0; } #include <stdlib.h> #include <stdio.h> int main() { FILE *fp; int i; float x; double z; fp=fopen("bytetest.dat", "r"); fread(&i, sizeof(int), 1, fp); fread(&x, sizeof(float), 1, fp); fread(&z, sizeof(double), 1, fp); printf("i=%d\n", i); printf("x=%f\n", x); printf("z=%g\n", z); fclose(fp); return 0; } if you excecute them both on the same machine or any other unix platform, they work properly. output===>i=10 x=1.234567 z=1.23457 However, if you separate compilation of them and read the data by your intel Next workstation, the results are screwed up. output==>i=167772160 x=8822335.000000 z=-1.04945e+238
From: robert@steffi.accessone.com (Robert Nicholson) Newsgroups: comp.sys.next.bugs Subject: Re: NEXTSTEP 3.3 patch (Is it ready?) Date: 05 Nov 1995 17:22:51 -0800 Organization: me organized? That's a joke! Sender: robert@steffi.accessone.com Message-ID: <ed9b6cz9w.fsf@steffi.accessone.com> References: <47di1d$hhd@cisun2000.unil.ch> <47g36j$1a6@arcadia.informatik.uni-muenchen.de> <stabl@informatik.uni-muenchen.de> writes: >shill@iphysiol.unil.ch (Sean Hill) writes: >>I was told that the NEXTSTEP 3.3 patch (fixing lookupd) would be released to >>the public on November 1st. Can we expect it on the ftp site soon? >Yes, you can. Just look at ftp.next.com or at peanuts.leo.org. Can we have reports from people who install this and find that their problems become smaller. I'd like to hear this before diving in. Also, personally I'd like to see bug's maintained like Sun. ie. rather than one wopping great big patch I'd like to see smaller patches that identify dependant patches and specifically which files address which problems solved by which patch. (I want to be able to de-patch if necessary) I'd like to be able to easily minimize risk of change. I'll conclude by saying that this probably isn't possible now because NeXT don't have the resources to do this. I aplaud NeXT for making the patches publically available. BTW: Is there anything else besides mirror that knows about "REST"? -- "Mary ate a little lamb and punk rock isn't dead" (PGP key: send email with Subject: request pgp key)
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: Re: nslookup broken in 3.3 patch? Date: 6 Nov 1995 02:15:36 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-d-43.usc.edu Message-ID: <47jr48$i08@usc.edu> References: <47ieu7$pju@news.tuwien.ac.at> wilkie@narya. (Alexander Wilkie) wrote: > Hi everyone! > To be on the safe side of things, I ran the "posix-app-friendly" > post-install script. Something I did NOT do was to download and > install the new libposix.a. I run NS 3.3 m68K, ran the DNS post-install script and did NOT d/l libposix.a. > After rebooting, nslookup stopped working. When called, it > just plays dead (i.e. no output at all, and no crash either) > until I quit it with Ctrl-C. This is odd, since the DNS service > appears to be working for other apps & programs. I ran a number of nslookups and it worked quite well. -- Thanks and be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV | finger for PGP key "Brooding's the word. Saw him kick the rose bush, kick the green ferns by the porch, decide against kicking the apple tree. God made it too firm. There, he just jumped on a dandelion."
From: Kevin P. Hannan Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.hardware Subject: Re: Binary data file IO error Date: 6 Nov 1995 03:39:26 GMT Organization: Future Services, Inc. Message-ID: <47k01e$878@news.atlcom.net> References: <DHJzx6.3IE@info.physics.utoronto.ca> In <DHJzx6.3IE@info.physics.utoronto.ca> Wei Liu wrote: > Anyone knows what could explain the following, > > I found I couldn't read those binary data files produced on other unix > platforms by my intel Next workstation. To test this, I wrote one code to > produce the data, the other reading the data. > [code removed] > if you excecute them both on the same machine or any other unix platform, > they work properly. > output===>i=10 > x=1.234567 > z=1.23457 > > However, if you separate compilation of them and read the data by your > intel Next workstation, the results are screwed up. > output==>i=167772160 > x=8822335.000000 > z=-1.04945e+238 May I hazard, that the diffrence is due to the architecture of the machines. One is big endian (Motorola) and the other is little endian (Intel). Forgive me if they are reversed. But that is exactly what happens to the bytes representing the numbers. Take a look at the bit patterns of the numbers, and compare the two. -- Kevin --
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 6 Nov 1995 05:15:09 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <47k5kt$ek3@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
Newsgroups: comp.sys.next.bugs,alt.sex.zoophilia From: vac@indirect.com(Anne Schoofs) Subject: Femdom In Search of Naughty Boys Message-ID: <DHLt9p.8C7@goodnet.com> Sender: vac@indirect.com(Anne Schoofs) Organization: Internet Direct, Inc. Date: Mon, 6 Nov 1995 04:17:49 GMT Seriously, this is a chance of a lifetime. I want all you naughty boys to contact me at once. Do not delay. Location is unimportant. Let me know how you've been naughty and what type of corrective punishment you deem appropriate. It's time we clean up the net. Anne --------------------------------------------------- Anne Schoofs vac@indirect.com Fax: 602 912 8823 Attn: RAS -------------------------------------------------
From: ivo@next.lbs.lon.ac.uk (Ivo Welch) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.hardware Subject: Re: Binary data file IO error Date: 6 Nov 1995 09:20:15 GMT Organization: London Business School Message-ID: <47kk0f$i8m@pluto.lbs.lon.ac.uk> You need a little routine instead (the same applies to int's, etc): static float inline getfloat(FILE *fin) { float f; ((char *)&f)[3]= getc(fin); ((char *)&f)[2]= getc(fin); ((char *)&f)[1]= getc(fin); ((char *)&f)[0]= getc(fin); return f; } /ivo welch -- Ivo Welch ivo.welch@anderson.ucla.edu Assoc Prof of Finance Anderson GSM at UCLA Until Dec 95: London Business School, Finance Dept, iwelch@lbs.lon.ac.uk Sussex Place, London NW1 4SA. England.
From: vac@indirect.com(Anne Schoofs) Newsgroups: alt.rock-n-roll.metal.progressive,comp.sys.next.bugs Subject: cmsg cancel <DHLpnq.Arv@goodnet.com> Control: cancel <DHLpnq.Arv@goodnet.com> Date: 06 Nov 95 07:43:40 GMT Organization: Devilbunnies Spam Cancelling Division Sender: vac@indirect.com(Anne Schoofs) Message-ID: <cancel.DHLpnq.Arv@goodnet.com> Spam Cancellation. For details see news.admin.net-abuse.announce
From: vac@indirect.com(Anne Schoofs) Sender: vac@indirect.com(Anne Schoofs) Newsgroups: comp.sys.next.bugs,alt.sex.zoophilia Date: 06 Nov 1995 09:23:34 EST Control: cancel <DHLt9p.8C7@goodnet.com> Subject: cmsg cancel <DHLt9p.8C7@goodnet.com> Message-ID: <cancel.DHLt9p.8C7@goodnet.com> Spam cancelled by dsr@lns598.lns.cornell.edu
From: borek (Borek Lupomesky) Newsgroups: comp.sys.next.bugs Subject: Problems with DNS resolving Date: 5 Nov 1995 23:31:54 GMT Organization: Czech Technical University Distribution: world Message-ID: <47jhha$j3i@ns.felk.cvut.cz> Hi folks! I am experiencing strange problems with DNS resolving. Sometimes it happens, that some name cannot be resolved. Well, might be, I try again, but successive tries exit immediately with 'host not found' message. The cure is to try to resolve non-existent name, break the resolving process and retry with the name that didn't pass before. This behavior happens also when I am connecting to terminal server (where I am getting the Internet) and try to do something (telnet etc.) before connection is established and negotiated. After the connect it continues to refuse to resolve the name. Bye Borek -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Borek Lupomesky (student, co-admin) University of J.E.Purkyne, Usti nad Labem, Czech Republic lupomesk@sun.ujep.cz, http://www.ujep.cz/~lupomesk/ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
From: art@cubicsol.com (Art Isbell) Newsgroups: comp.sys.next.bugs Subject: Re: nslookup broken in 3.3 patch? Date: 6 Nov 1995 20:54:13 GMT Organization: Sense Networking Seattle (www.oz.net) Message-ID: <47lsll$51i@emerald.oz.net> References: <47ieu7$pju@news.tuwien.ac.at> wilkie@narya. (Alexander Wilkie) wrote: > After rebooting, nslookup stopped working. When called, it just plays > dead (i.e. no output at all, and no crash either) until I quit it with > Ctrl-C. This is odd, since the DNS service appears to be working for other > apps & programs. Seems to work fine on black 3.3 over a PPP link to a remote nameserver. -- Art Isbell NeXT & MIME Mail: art@cubicsol.com NeXT Registered Consultant Voice/Fax: +1 408 335 2515 Trego Systems Voice Mail: +1 408 335 1154 CaseServ: NEXTSTEP managed care US Mail: Felton, CA 95018-9442 contract and case management solutions
From: art@cubicsol.com (Art Isbell) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.hardware Subject: Re: Binary data file IO error Date: 6 Nov 1995 20:24:11 GMT Organization: Sense Networking Seattle (www.oz.net) Message-ID: <47lqtb$51i@emerald.oz.net> References: <47kk0f$i8m@pluto.lbs.lon.ac.uk> ivo@next.lbs.lon.ac.uk (Ivo Welch) wrote: > You need a little routine instead (the same applies to int's, etc): > > static float inline getfloat(FILE *fin) { > float f; > ((char *)&f)[3]= getc(fin); > ((char *)&f)[2]= getc(fin); > ((char *)&f)[1]= getc(fin); > ((char *)&f)[0]= getc(fin); > return f; > } Careful!! This assumes that a float is 4 bytes in size which isn't necessarily true on all systems. And according to the on-line Portability Guide: "Some processors modify the value of a floating-point number if that value is invalid, so the swap functions for floating-point values can't simply return float or double." NeXT has provided all sorts of byte-swapping functions which are discussed in the Portability Guide. -- Art Isbell NeXT & MIME Mail: art@cubicsol.com NeXT Registered Consultant Voice/Fax: +1 408 335 2515 Trego Systems Voice Mail: +1 408 335 1154 CaseServ: NEXTSTEP managed care US Mail: Felton, CA 95018-9442 contract and case management solutions
From: mark_bessey@next.com (Mark Bessey) Newsgroups: comp.sys.next.bugs Subject: Re: sort bug... Date: 6 Nov 1995 22:20:09 GMT Organization: NeXT Computer, Inc. Message-ID: <47m1mp$h0b@news.next.com> References: <en3afwdhh.fsf@steffi.accessone.com> Robert Nicholson writes > Anybody explain this? (3.3) > > root:/>sort -T /LocalLibrary > root:/>sort -T /External1/nextstep/LocalLibrary > zsh: segmentation fault > > lrwxrwxrwx 1 root 32 Apr 2 1995 LocalLibrary -> > /External1/nextstep/LocalLibrary Yep. Note the following declaration, from sort.c: char file1[30]; Anybody want to guess what this variable gets used for? That's right- It holds the path to the temporary files directory! So, any path name over 30 characters causes a memory overwrite, which usually crashes sort. > > Guess it's time to fetch gnu sort. Probably not a bad idea. Gotta love those ancient BSD utilities... > > -- > "Mary ate a little lamb and punk rock isn't dead" > (PGP key: send email with Subject: request pgp key) > Thanks for pointing this out. It's been entered into the NeXT bug tracking database. -- Mark Bessey NeXT Computer, Inc Software Quality Assurance -->I DON'T SPEAK FOR NeXT <--
From: wlau1@panther.uwo.ca (Wayne Lau) Newsgroups: comp.sys.next.bugs Subject: IDE and SCSI Date: 7 Nov 1995 02:44:32 GMT Organization: Univ. Western Ontario, London, ON, Canada Sender: etittley@phobos.astro.uwo.ca (Eric Tittley) Distribution: world Message-ID: <47mh6g$aok@falcon.ccs.uwo.ca> Please help. I have a conflict between my IDE card and my SCSI card attatched to my CD-ROM. I'm using NEXTSTEP 3.3 for Intel-based machines. (486DX4-100) When I put it in the PCI slot 4, after installation the system crashes. I am sure that there is a conflict between the video drivers and the SCSI controller. After installation, removal of the SCSI controller cures the ails. The 'ails' occur with the hard drive which is IDE and the video card (ATI mach64). If I put the SCSI in PCI slot 1, where it is supposed to be acording to my mother board configuration document, booting up the system produces an error message: Missed Interrupt. How can I avoid these problems? Any assistance would be helpful. Please respond directly instead of posting since I am using a colleague's Newsnet reader :). -- Wayne Lau wlau1@panther.uwo.ca
From: robert@steffi.accessone.com (Robert Nicholson) Newsgroups: comp.sys.next.bugs Subject: Re: sort bug... Date: 07 Nov 1995 10:59:53 -0800 Organization: me organized? That's a joke! Sender: robert@steffi.accessone.com Message-ID: <ehh0gqmhi.fsf@steffi.accessone.com> References: <en3afwdhh.fsf@steffi.accessone.com> <47m1mp$h0b@news.next.com> <mark_bessey@next.com> writes: >Robert Nicholson writes >>Anybody explain this? (3.3) >> >>root:/>sort -T /LocalLibrary >>root:/>sort -T /External1/nextstep/LocalLibrary >>zsh: segmentation fault >> >>lrwxrwxrwx 1 root 32 Apr 2 1995 LocalLibrary -> >>/External1/nextstep/LocalLibrary >Yep. Note the following declaration, from sort.c: >char file1[30]; and what about MAXPATHLEN? >Anybody want to guess what this variable gets used for? That's right- It >holds the path to the temporary files directory! So, any path name over >30 characters causes a memory overwrite, which usually crashes sort. >> >>Guess it's time to fetch gnu sort. Got both fileutils and textutils and everything works fine now. I even don't get the line to long problems that NeXT sort was complaining about. >Probably not a bad idea. Gotta love those ancient BSD utilities... >> >>-- >>"Mary ate a little lamb and punk rock isn't dead" >>(PGP key: send email with Subject: request pgp key) >> >Thanks for pointing this out. It's been entered into the NeXT bug >tracking database. You're welcome. -- "Mary ate a little lamb and punk rock isn't dead" (PGP key: send email with Subject: request pgp key)
Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs From: bfbaker@netcom.com (Technom Enterprises Inc) Subject: Help with EOF for PDO Message-ID: <bfbakerDHp5K4.GxJ@netcom.com> Organization: Operations Technologies, LLC. Date: Tue, 7 Nov 1995 23:36:04 GMT Sender: bfbaker@netcom4.netcom.com Greetings, We've in the process of converting our product to PDO, and we're having difficulties getting EOF to function. Any attempt to create an EODatabase object, via initWithAdaptor, initWithModel, or indirectly by creating an EODatabaseDataSource, results in an uncaught exception (100003 or 100005). This error doesn't tell me much about what the problem might be, and the docs are less helpful yet. If anyone has experience with EOF for PDO, or better yet, would be willing to send me an example of code that works (including makefiles to build it correctly), I'd like to hear from you. Thanks in advance, Brian Baker. bbaker@technom.com FYI: Our system looks like: HP 715/64 HP-UX 09.05 PDO 3.0 release EOF Server 1.1 Using the Sybase Adaptor
Newsgroups: comp.sys.next.bugs From: brad@instep.bc.ca Subject: Re: sort bug... Message-ID: <1995Nov8.015711.544@instep.bc.ca> Sender: usenet@instep.bc.ca Organization: InStep Mobile Communications Inc. References: <47m1mp$h0b@news.next.com> Date: Wed, 8 Nov 1995 01:57:11 GMT Mark Bessey writes >Robert Nicholson writes >> Anybody explain this? (3.3) >> >> root:/>sort -T /LocalLibrary >> root:/>sort -T /External1/nextstep/LocalLibrary >> zsh: segmentation fault >> >> lrwxrwxrwx 1 root 32 Apr 2 1995 LocalLibrary -> >> /External1/nextstep/LocalLibrary > >Yep. Note the following declaration, from sort.c: >char file1[30]; > >Anybody want to guess what this variable gets used for? That's right- It >holds the path to the temporary files directory! So, any path name over >30 characters causes a memory overwrite, which usually crashes sort. > >> >> Guess it's time to fetch gnu sort. > No need to fetch gnu sort. You should have it as /usr/gnu/bin/sort and /usr/gnu/man/man1/sort.1 under NS3.3 -- Brad Head <brad@instep.bc.ca> Software Developer, InStep Mobile Communications Inc. Vancouver, British Columbia, Canada
From: temp995@ips.ca (jackie downy) Newsgroups: comp.sys.next.bugs Subject: (((((( Fetish Hotline Call 011-592-586-442 ))))) Date: Fri, 03 Nov 1995 05:02:34 GMT Organization: internet passport services inc. Message-ID: <47p1fr$327@news2.his.com>
From: flight@mathi.uni-heidelberg.de (Gregor Hoffleit) Newsgroups: comp.sys.next.bugs Subject: No alerts for floppy I/O error Date: 8 Nov 1995 11:47:37 GMT Organization: University of Heidelberg, Germany Message-ID: <47q5cp$t0k@sun0.urz.uni-heidelberg.de> This strikes me since a while: I've never got any alerts with NEXTSTEP (3.2/3.3, Intel/HP) when there were I/O error with floppies! You hear the floppy going wild (scratching a while at the same point), but NEXTSTEP doesn't warn you about the suspicious I/O at all. This occurs at writing as well as reading. For floppies with DOS file system, I've often verified that DOS and Linux do report I/O error while reading files, while NEXTSTEP when reading the same file from the same floppy quietly outputs some rubbish without any error messages or alerts even in the console. Especially dangerous is that the alerts are lacking for errors at writing, too. This bug can be found not only for DOS, but also for NEXTSTEP filesystems. Gregor -- | Gregor Hoffleit admin MATHInet / contact HeidelNeXT | | MAIL: Mathematisches Institut PHONE: (49)6221 56-5771 | | INF 288, 69120 Heidelberg / Germany FAX: 56-3812 | | EMAIL: flight@mathi.uni-heidelberg.de (NeXTmail) |
From: temp995@ips.ca (jackie downy) Organization: internet passport services inc. Message-ID: <cancel.47p1fr$327@news2.his.com> Control: cancel <47p1fr$327@news2.his.com> Newsgroups: comp.sys.next.bugs Subject: cmsg cancel <47p1fr$327@news2.his.com> Date: Wed, 08 Nov 1995 21:26:53 KST
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.bugs Subject: Diamond Stealth V3.33 Bug! Date: Wed, 8 Nov 1995 18:05:36 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951108180127.25240A-100000@hphalle0.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Upgrading the Diamond Stealth Driver from V3.30 to V3.33 (and additionally the S3 driver) I noticed the following new introduced bug: in 32 Bit modes the new V3.33 DS and S3 Generic Driver give a much more dimmer picture then their old aequivalents. The bug seemes to be introduced in conjunction with the new S3 driver on which the Diamond Stealth driver _seems_ to be based on code. I'm using an old Diamond Stealth V64 4MB VRAM with S3 964 and Idek liyma monitor. Greetings, Boerny. -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/ scholz@ve1.rm.op.dlr.de http://www.leo.org/~scholz/
From: eric@pisces.USask.Ca (Norum,132L,6308,3737314) Newsgroups: comp.sys.next.bugs Subject: printf %g format still broken in 3.3 Patch Date: 8 Nov 1995 18:22:07 GMT Organization: University of Saskatchewan Message-ID: <47qsgf$1f4@tribune.usask.ca> I just installed the 3.3 User and Developer Patch on the machines here. Just for the heck of it, I checked to see if my favorite bug was still there. ================== printfcheck.c ==================== #include <stdio.h> int main (int argc, char **argv) { printf ("Expect 0.00123: %.3g\n", 0.001234567); printf ("Expect 123: %.3g\n", 123.4567); printf ("Expect 123.5: %.4g\n", 123.4567); printf ("Expect 1e+03: %.3g\n", 999.6); return 0; } ======================================================== eric@pisces 217> cc printfcheck.c eric@pisces 216> a.out Expect 0.00123: 0.001 Expect 123: 123.457 Expect 123.5: 123.4567 Expect 1e+03: 999.6 ========================================================== O.K., maybe the posix libraries are fixed..... eric@pisces 218> cc -posix printfcheck.c eric@pisces 219> a.out Expect 0.00123: 0.001 Expect 123: 123.457 Expect 123.5: 123.4567 Expect 1e+03: 999.6 ============================================================ Nope, I guess not. Sigh. That makes it 6 *years* now since I first reported this (in NeXTstep version 1.0) Here's hoping that this gets fixed in 4.0......
From: eric@pisces.USask.Ca (Norum,132L,6308,3737314) Newsgroups: comp.sys.next.bugs Subject: printf %g -- return address wrong Date: 8 Nov 1995 18:24:51 GMT Organization: University of Saskatchewan Message-ID: <47qslj$1f4@tribune.usask.ca> I just started using Alexandra.app for reading/posting news. I guess I don't have things set up properly, yet. -- Eric Norum eric@skatter.usask.ca Saskatchewan Accelerator Laboratory Phone: (306) 966-6308 University of Saskatchewan FAX: (306) 966-6058 Saskatoon, Canada. NeXTMail accepted.
From: Robert Worne <rworne@primenet.com> Newsgroups: comp.sys.next.bugs Subject: Re: No alerts for floppy I/O error Date: 8 Nov 1995 20:04:42 GMT Organization: A Big Black Cube Message-ID: <47r2gq$e1l@nnrp2.news.primenet.com> References: <47q5cp$t0k@sun0.urz.uni-heidelberg.de> flight@mathi.uni-heidelberg.de (Gregor Hoffleit) wrote: > This strikes me since a while: > > I've never got any alerts with NEXTSTEP (3.2/3.3, Intel/HP) when there > were I/O error with floppies! You hear the floppy going wild > (scratching a while at the same point), but NEXTSTEP doesn't warn you > about the suspicious I/O at all. > > This occurs at writing as well as reading. > > For floppies with DOS file system, I've often verified that DOS and > Linux do report I/O error while reading files, while NEXTSTEP when > reading the same file from the same floppy quietly outputs some > rubbish without any error messages or alerts even in the console. > > Especially dangerous is that the alerts are lacking for errors at > writing, too. > I have put up with this for so long on my black HW it is inexcusable. It never reports any errors writing to the floppy. Hence at least one (sometimes two) out of three disk transfers to a PC result in the "Sector not found" error. This especially happens when the disks are formatted DOS on the NeXT system. The same disk can then be reformatted on a PC (with no media errors) and taken to the NeXT, and it them will properly copy the file (usually) -- //--------------------------------------------------------------------- // Robert Worne rworne@primenet.com OS/2-NeXT -=Starving CS Undergrad=- //--------------------------------------------------------------------- // Microsoft Network is prohibited from redistributing this work in // any form, in whole or in part. Copyright, Robert Worne, 1995. // License to distribute this post is available to Microsoft for $1000. // Posting without permission constitutes an agreement to these terms. // Please send notices of violation to rworne@primenet.com and // postmaster@microsoft.com
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.bugs Subject: Re: Diamond Stealth V3.33 Bug! Date: Thu, 9 Nov 1995 10:09:10 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951109100617.2079A-100000@hphalle8b.informatik.tu-muenchen.de> References: <199511090447.UAA25156@albans.cs.ucla.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <199511090447.UAA25156@albans.cs.ucla.edu> On Wed, 8 Nov 1995, Mark Yarvis wrote: > I've spoken with NeXT about this problem. Apparently it's a gamma correction > problem and the driver has no support for user configuratin of gamma > correction. In addition you'll notice that > 1. The 16 bit modes are quite washed out > 2. There is also no support for brighness control ... not a big deal because > you can just adjust your monitor, but it is broken. No you can't, because for me the monitor is on it's limit to give a reasonable picture quality, so I switched back to the old driver. But it's good to hear the NeXT does know about the problem. I'm only wondering why they introduced the bug with the new driver, while the old one worked flawlessly (with the exception to the new S3 chipsets (S3 968/868) Greetings, Boerny. -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/ scholz@ve1.rm.op.dlr.de http://www.leo.org/~scholz/
Newsgroups: comp.sys.next.bugs Subject: Re: printf %g format still broken in 3.3 Patch Message-ID: <1995Nov9.115514.45659@yogi.urz.unibas.ch> From: frank@ifi.unibas.ch Date: 9 Nov 95 11:55:14 MET References: <47qsgf$1f4@tribune.usask.ca> eric@pisces.USask.Ca (Norum,132L,6308,3737314) wrote: > I just installed the 3.3 User and Developer Patch on the machines here. > Just for the heck of it, I checked to see if my favorite bug was still there. > > ================== printfcheck.c ==================== > #include <stdio.h> > > int main (int argc, char **argv) > { > printf ("Expect 0.00123: %.3g\n", 0.001234567); > printf ("Expect 123: %.3g\n", 123.4567); > printf ("Expect 123.5: %.4g\n", 123.4567); > printf ("Expect 1e+03: %.3g\n", 999.6); > return 0; > } > ======================================================== > eric@pisces 217> cc printfcheck.c > eric@pisces 216> a.out > Expect 0.00123: 0.001 > Expect 123: 123.457 > Expect 123.5: 123.4567 > Expect 1e+03: 999.6 > > ========================================================== > O.K., maybe the posix libraries are fixed..... > eric@pisces 218> cc -posix printfcheck.c > eric@pisces 219> a.out > Expect 0.00123: 0.001 > Expect 123: 123.457 > Expect 123.5: 123.4567 > Expect 1e+03: 999.6 > ============================================================ > Nope, I guess not. > > Sigh. That makes it 6 *years* now since I first reported this (in > NeXTstep version 1.0) Here's hoping that this gets fixed in 4.0...... According to the man pages *and* the posix handbook, the above results are *absolutely* correct. Both the manpages and the posix specs clearly state: f The float or double arg is converted to decimal nota- tion in the style `[-]mmm.ddd' where the number of d's is equal to the precision specification. ^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^ If the preci- sion is missing, 6 digits are given; if the precision is explicitly 0, no d's and no decimal point are printed. e, or E The float or double arg is converted in the style `[-]m.ddde+xx' where there is one digit before the decimal point and the number of d's after is equal to the precision specification for the argument. ^^^^^^^^^^^^^^^^^^^^^^^ When the precision is missing, 6 digits are produced. and further: g, or G The float or double arg is printed in style e if the exponent printed would be less than -4 or greater than or equal to the precision; otherwise, style f is used. Trailing zeros and trailing decimal points are not printed. where the precision specification is given as: + an optional digit string specifying a precision which specifies the number of digits to appear after the ^^^^^ decimal point, for e- and f-conversion; or the maximum number of characters to be printed from a string; or the minimum number of digits to be printed with integer formats (leading zeros will be added to fill to the precision); Thus, specifying '%.3g' will cause three *digits* to be printed *after* the decimal point, provided that there are three valid digits - precisely what the program does. As this does not specify a field width, a default is assumed and trailing zeros are not printed. I see absolutley no bug here! Please enlighten me if I missed something essential. -Robert -- Institut fuer Informatik tel +41 (0)61 321 99 67 Universitaet Basel fax. +41 (0)61 321 99 15 Robert Frank Mittlere Strasse 142 rfc822: frank@ifi.unibas.ch (NeXT,MIME mail ok) CH-4056 Basel X400: S=frank;OU=ifi;O=unibas;P=switch;A=arcom;C=ch Switzerland
From: dave@turbocat.snafu.de (David Wetzel) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: 3.3 Patch:INND does not run anymore! Date: 9 Nov 1995 17:16:00 GMT Organization: Fachhochschule fuer Technik und Wirtschaft Berlin, FB1 Message-ID: <47td0g$a5o@commserv.fb1.fhtw-berlin.de> Hi! After installing the 3.3 Patch on my i586, I noticed that my inn does not work anymore. Nov 9 17:29:53 turbocat inndstart: inndstart cant bind Address already in use What shall I do? I was using inn 1.4. Do the people at NeXT.com not use INN? Thanks in advance. _ _ _(_)(_)_ David Wetzel, Turbocat's Development, (_) __ (_) Buchhorster Strasse, D-16567 Muehlenbeck/Berlin, FRG, _/ \_ Phone +49 33056 82151, Fax +49 33056 82152 (______) dave@turbocat.snafu.de (NeXTMail)
Newsgroups: comp.sys.next.bugs From: Joakim Johansson <jocke@rat.se> Subject: Some first impressions of the NS patch [PA-RISC] Message-ID: <DHqK0K.Mtp@rat.se> Sender: jocke@rat.se (Joakim Johansson) Organization: Research & Trade, AB. Date: Wed, 8 Nov 1995 17:45:53 GMT [speaking only for the PA-RISC part of the patch, as that's the only experience I have of it so far. (1.5 days use) ] I can only applaud the engineers who updated the windowserver + kernel! The system now feels much snappier, anyone running NS on PA-RISC should definitely try it out. (that it fixes other serious bugs (networking stuff / sendmail vulnerabilities) as well as some annoying ones (gdb reverse sorting etc..) is a nice bonus..) It seems that event handling is done faster (hard to measure) and graphics performance is better (10-20% with NXBench, feels like more in real situations)... Well, enough praise - better keep them on their toes.. ;-) I hope we see more patches like this in the future.. I don't think NeXT thinks it a good business move to provide ongoing patches like other unix vendors do, but an between-releases large patch like this is probably a reasonable solution in the long term as well... Joakim -- Joakim Johansson Software Developer, Research & Trade jocke@rat.se <NeXTmail, MIME>
Newsgroups: comp.sys.next.bugs From: martin@rat.se Subject: Re: Some first impressions of the NS patch [PA-RISC] Message-ID: <DHrrCM.4Bn@rat.se> Keywords: foundation patch Sender: martin@rat.se (Martin Wennerberg) Organization: Research & Trade, AB. References: <DHqK0K.Mtp@rat.se> Date: Thu, 9 Nov 1995 09:21:57 GMT In article <DHqK0K.Mtp@rat.se> Joakim Johansson <jocke@rat.se> writes: > I can only applaud the engineers who updated the windowserver + kernel! -I second that. I was a bit disappointed to note the lack of a foundation kit update though. There is a patch on the EOF 1.1 CD, but if you don't have EOF you can't get the patch. I think NeXT should make the foundation patch freely available. Martin Martin Wennerberg ____________________________________________________ Research & Trade AB Phone: + 46 - 8 - 21 17 50 Fax: + 46 - 8 - 24 54 11 Email: martin@rat.se [MIME & NeXTmail supported] Snailmail: Box 7742, S-103 95 Stockholm, SWEDEN Visiting: Kungsgatan 33
From: Christopher_Lane@Med.Stanford.EDU Newsgroups: comp.sys.next.bugs Subject: Re: printf %g format still broken in 3.3 Patch Date: 9 Nov 1995 17:38:08 GMT Organization: Stanford University Message-ID: <47tea0$iq6@nntp.Stanford.EDU> References: <1995Nov9.115514.45659@yogi.urz.unibas.ch> frank@ifi.unibas.ch writes > I see absolutley no bug here! Please enlighten me if I missed > something essential. If I understand the bug correctly, the difference is that you've interpreted 'precision' to mean 'decimal places', where in reality the precision defined by '%.3g' would be independent of how many places you shift a number to the right or left. Consider the numbers 1.234567 & 12.34567 -- if I only trust three digits of precision (%.3g) then they format as 1.23 & 12.3, the rest of the digits are noise regardless of where they appear relative to the decimal point which, for this purpose, is completely arbitrary. If you don't buy that, you can always try the traditional approach of compiling Eric's test code on another Unix. I tried a Sun, HP, SGI & RS6000 and all matched the 'Expect'ed number and not NeXT's results. - Christopher
From: eric@skatter.usask.ca (Eric Norum) Newsgroups: comp.sys.next.bugs Subject: Re: printf %g format still broken in 3.3 Patch Date: 9 Nov 1995 21:43:26 GMT Organization: University of Saskatchewan Message-ID: <47tslu$58d@tribune.usask.ca> References: <47qsgf$1f4@tribune.usask.ca> <1995Nov9.115514.45659@yogi.urz.unibas.ch> frank@ifi.unibas.ch wrote: : According to the man pages *and* the posix handbook, the above : results are *absolutely* correct. Both the manpages and the posix : specs clearly state: : f The float or double arg is converted to decimal nota- : tion in the style `[-]mmm.ddd' where the number of d's : is equal to the precision specification. : ^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^ : If the preci- : sion is missing, 6 digits are given; if the precision : is explicitly 0, no d's and no decimal point are : printed. : e, or E : The float or double arg is converted in the style : `[-]m.ddde+xx' where there is one digit before the : decimal point and the number of d's after is equal to : the precision specification for the argument. : ^^^^^^^^^^^^^^^^^^^^^^^ : When the : precision is missing, 6 digits are produced. : and further: : g, or G : The float or double arg is printed in style e if the : exponent printed would be less than -4 or greater than : or equal to the precision; otherwise, style f is used. : Trailing zeros and trailing decimal points are not : printed. : where the precision specification is given as: : + an optional digit string specifying a precision which : specifies the number of digits to appear after the : ^^^^^ : decimal point, for e- and f-conversion; or the maximum : number of characters to be printed from a string; or : the minimum number of digits to be printed with integer : formats (leading zeros will be added to fill to the : precision); : Thus, specifying '%.3g' will cause three *digits* to be printed : *after* the decimal point, provided that there are three valid : digits - precisely what the program does. As this does not specify : a field width, a default is assumed and trailing zeros are not : printed. : I see absolutley no bug here! Please enlighten me if I missed : something essential. I absolutely *cannot* let this go unchallenged. NeXT's goofy handling of `%g' has been one of my pet peeves since December 1989 when I got my cube! I've reported the bug in *every* software release from 1.0 to 3.0, inclusive, but it never gets fixed. To hear that it is still in 3.1 is most disappointing. The problem isn't just when to switch between `e' and `f'. The problem is more in the way the `f' format is rendered! The ANSI standard (X3.159-1989) says (Section 4.9.6.1, page 134, line 33): g,G The double argument is converted in style f or e (or in style E in the case of a G conversion specifier), with the precision specifying the number of significant digits. If the precision is zero, it is taken as 1. The style used depends on the value converted; style e (or E) will be use only if the exponent resulting from the conversion is less than -4 or greater than or equal to the precision. Trailing zeros are removed from the fractional portion of the result; a decimal-point character appears only if it is followed by a digit. *** NOTE THE WORDS `significant digits'. *** Let's have a look at the examples from the original article: >> >> format "%.3g" .0012341 >> >> ==== Result was: >> 0.001 >> ---- Result should have been: >> 0.00123 The zero's in 0.00123 are *not* significant. Score 1 against NeXT. >> >> format "%.4g" 9999.6 >> >> ==== Result was: >> 9999.6 >> ---- Result should have been: >> 1e+04 >> The format calls for 4 SIGNIFICANT DIGITS. Rounding 9999.6 to four significant digits leaves 10000, which has an exponent equal to the precision and therefore should be printed in e format. Score 2 against NeXT. -- Eric Norum eric@skatter.usask.ca Saskatchewan Accelerator Laboratory Phone: (306) 966-6308 University of Saskatchewan FAX: (306) 966-6058 Saskatoon, Canada. NeXTMail accepted.
From: eric@skatter.usask.ca (Eric Norum) Newsgroups: comp.sys.next.bugs Subject: Re: printf %g format still broken in 3.3 Patch Date: 9 Nov 1995 21:47:10 GMT Organization: University of Saskatchewan Message-ID: <47tssu$58d@tribune.usask.ca> References: <47qsgf$1f4@tribune.usask.ca> <1995Nov9.115514.45659@yogi.urz.unibas.ch> <47tslu$58d@tribune.usask.ca> Following up my own response........ Here's the output from three other machines around the lab here. =================================================================== On a SUN 4/380 (SUNOS 4.1.1): Expect 0.00123: 0.00123 Expect 123: 123 Expect 123.5: 123.5 Expect 1e+03: 1e+03 So it would appear that at least *one* vendor can get it right! On an HP-715 (HP-UX V8 -- I think): Expect 0.00123: 0.00123 Expect 123: 123 Expect 123.5: 123.5 Expect 1e+03: 1e+03 Hmmm...looks like there are at least *two* vendors that can get it right. On a DEC MicroVAX II (Ultrix version 2.2): Expect 0.00123: 0.00123 Expect 123: 123 Expect 123.5: 123.5 Expect 1e+03: 1e+03 Wow, even an antique computer can do this properly! =================================================================== -- Eric Norum eric@skatter.usask.ca Saskatchewan Accelerator Laboratory Phone: (306) 966-6308 University of Saskatchewan FAX: (306) 966-6058 Saskatoon, Canada. NeXTMail accepted.
Newsgroups: comp.sys.next.bugs From: csaldanh@mae.carleton.ca (Chris Saldanha) Subject: Re: nslookup broken in 3.3 patch? Message-ID: <DHsr0C.M8o@cunews.carleton.ca> Sender: news@cunews.carleton.ca (News Administrator) Organization: Carleton University References: <47ieu7$pju@news.tuwien.ac.at> Date: Thu, 9 Nov 1995 22:12:12 GMT Alexander Wilkie (wilkie@narya.) wrote: : To be on the safe side of things, I ran the "posix-app-friendly" : post-install script. Something I did NOT do was to download and install the : new libposix.a. : After rebooting, nslookup stopped working. When called, it just plays : dead (i.e. no output at all, and no crash either) until I quit it with : Ctrl-C. This is odd, since the DNS service appears to be working for other : apps & programs. I ran the "ImprovedDNS" script, and had no trouble with nslookup. Having libposix.a shouldn't affect the nslookup command one way or the other, since it is a STATIC library. --Chris Chris Saldanha, Software Analyst -------------------------------------- computerActive, Inc |"The telephone was not invented by | chris@computerActive.on.ca (NeXTMail) | Alexander Graham Unitel" -Bell Ad | http://www.mae.carleton.ca/~csaldanh --------------------------------------
From: stevem@radium.ca (Stephen MacDougall) Newsgroups: comp.sys.next.bugs Subject: Re: printf %g format still broken in 3.3 Patch Date: 9 Nov 1995 21:57:41 GMT Organization: ftn Internet Message-ID: <47ttgl$d5f@master.ftn.net> References: <47qsgf$1f4@tribune.usask.ca> eric@pisces.USask.Ca (Norum,132L,6308,3737314) wrote: > I just installed the 3.3 User and Developer Patch on the machines here. > Just for the heck of it, I checked to see if my favorite bug was still there. > > ================== printfcheck.c ==================== > #include <stdio.h> > > int main (int argc, char **argv) > { > printf ("Expect 0.00123: %.3g\n", 0.001234567); > printf ("Expect 123: %.3g\n", 123.4567); > printf ("Expect 123.5: %.4g\n", 123.4567); > printf ("Expect 1e+03: %.3g\n", 999.6); > return 0; > } > ======================================================== > eric@pisces 217> cc printfcheck.c > eric@pisces 216> a.out > Expect 0.00123: 0.001 > Expect 123: 123.457 > Expect 123.5: 123.4567 > Expect 1e+03: 999.6 > [Chunk] > Sigh. That makes it 6 *years* now since I first reported this (in > NeXTstep version 1.0) Here's hoping that this gets fixed in 4.0...... From what I can see, things are working just fine. By saying '%.3g", for example, you are asking for a formatted number with 3 digits of precision *AFTER* the decimal point. That is precisely what you are getting, where %.3g is being used. Also, your first printf statement reads as follows: printf ("Expect 0.00123: %.3g\n", 0.001234567); How do you expect to get 5 digits after the decimal point with this format?? -- ============================================================================ | | | | Stephen MacDougall | #include <std_disclaimer.h> | | RADium Technology Centre (Canada) | | | stevem@RADium.ca | (NeXTMail and MIME capable) | | | | ============================================================================
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.bugs Subject: Recognition of Audio-CDs locking the system with 3.3Patch Date: Fri, 10 Nov 1995 10:51:45 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951110104424.22151C-100000@hphalle7a.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I recently installed the 3.3 Patch for Intel and beside the mentioned enhancments (fast 12Bit window handling, lookupd, etc.) I noticed, that if I insert an Audio CD, the system starts to hang for about 1 minute before launching CD-Player. Before I installed the patch, this behaviour only occured if I inserted the CD and accessed the root directory. Greetings, Boerny. -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/ scholz@ve1.rm.op.dlr.de http://www.leo.org/~scholz/
From: JDT@bgsm.edu (John Turner) Newsgroups: comp.sys.next.bugs Subject: IBM Token RIng Driver. Date: Fri, 10 Nov 1995 14:22:45 Organization: North Carolina Baptist Hospital/Bowman Gray Medcenter Message-ID: <JDT.31.000E6189@bgsm.edu> Keywords: token ring Well I am working with the 3.33 Beta driver for the IBM Auto 16/4 Token Ring ISA Adapter under NS 3.3 for Intel. When it works it works fine, but I am having problems when the machine is rebooted that it is unable to talk to the network and no routing information gets added to the route tables. This problem is fixed by repeated reboots and some wishful thinking. I am wondering if others have seen/heard of this problem. As I said if the reboot works then I can talk to everyone, but if it fails I can't see a sole. No messages are printed when the drivers get setup, it is no until the client tries to find the Netinfo server do I see a problem. John. jdturner@bgsm.edu
From: altenber@acpub.duke.edu (Lee Altenberg) Newsgroups: comp.sys.next.bugs Subject: Audio CD plays my speakers w/3.3 patch! Date: 10 Nov 1995 17:38:58 -0500 Organization: Duke University, Durham, NC, USA Message-ID: <480ka2$jlj@news.duke.edu> Hey, after installing the 3.3patch, my CDROM now sends its audio output to my Sound Blaster 16 (you have to have them connected) and out through my speakers! So it appears that the promised reworking of NEXTSTEP sound is happening. Thanks, NeXT. -Lee Altenberg altenber@mhpcc.edu
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <flexus!rfschtkt@maze.ruca.ua.ac.be> Date: Sat, 11 Nov 95 10:40:35 +0100 From: Raf Schietekat <flexus!rfschtkt@maze.ruca.ua.ac.be> Message-ID: <9511110940.AA00263@flexus> Subject: Re: printf %g format still broken in 3.3 Patch Robert Frank <frank@ifi.unibas.ch> suggested that there is no %g bug, on the grounds of the man pages and the posix manual. The crucial information is the definition of precision (the semantics of the number following the period). But I have checked my K&R 2nd. ed. about it, and there is an additional specification (translating back to English; I could only obtain a Dutch translation when I bought this book): ``if the conversion character is g or G, the precision indicates the number of significant digits''. Christopher_Lane@Med.Stanford.EDU suggested this, but didn't add a quote to back it up. So the implementation seems to be internally consistent, but it does not follow the ANSI standard. Unless K&R is not exactly ANSI (perhaps, because it says that it is based on the proposed standard)? That's for others to determine. Also interesting would be to check other BSD 4.3 implementations, and BSD 4.4 (could this be used as a patch?). Raf Schietekat, RfSchtkt@maze.ruca.ua.ac.be (NeXTmail), Flanders, Belgium If I don't answer: my mail relay can't handle !, % or .uucp, I think *** The year 2000 will be the last year of the 20th century. ***
From: jblevins@macs14.uwa.edu.au (Jim Blevins) Newsgroups: comp.sys.next.hardware,comp.sys.next.bugs Subject: Re: Boot Error on Thinkpad 755 CX Date: 12 Nov 1995 03:15:06 GMT Organization: Centre for Linguistics, UWA, Australia Distribution: world Message-ID: <JBLEVINS.95Nov12111506@macs14.uwa.edu.au> References: <JBLEVINS.95Nov10152136@macs14.uwa.edu.au> In-reply-to: jblevins@macs14.uwa.edu.au's message of 10 Nov 1995 07:21:36 GMT >I have a Thinkpad 755CX with a 24MB and aSlimSCSI adaptor that refuses to boot >from the 3.3 boot floppy. Almost immediately, the boot process halts with the >following error message: > > NEXTSTEP boot1 v3.3.3.8 > ..Read error The source of the problem appears to be a 2.88 floppy drive, which NS3.3 does not recognize (on Intel). Can anyone suggest a fix or workaround? Thanks, -Jim -- Jim Blevins jblevins@uniwa.uwa.edu.au Centre for Linguistics phone: +61-9-380-2866 University of Western Australia fax: +61-9-380-1154 Nedlands, W.A. 6009
From: daniel@vis.inf.ethz.ch (Daniel Gerald Kluge) Newsgroups: comp.sys.next.bugs Subject: Swaping on two files impossible ? Date: 12 Nov 1995 23:44:51 GMT Organization: Verein der Informatikstudierenden an der ETH Zuerich Message-ID: <4860tj$nis@neptune.ethz.ch> I just installed my new Disk, which happens to be some bytes to large to be one partition (it's a Quantum 2.3GB Disk), so I decided it is time to make a swap-partition, and stumbled over some weirdities: 1. NeXT doesn't support swap-partitions per se, you have to mount them from fstab, and you have to Edit /etc/rc, since rc.swap gets executed before all partitions are mounted. 2. If the systems tries to turn on the swapping, I get the following error: mount: Device busy mount: giving up on: /private/vm/swapfile.front mach_swapon: mach_swapon failed: Device busy To verify, just try by hand to enable a second swapfile... is this broken by now ? I remember, that we run once the computer I'm writing on with two swapfiles.. -daniel P.S. Where did /usr/man/man5/fstab.5 go to in 3.3 ? -- Daniel G. Kluge - cs-student @ Federal Institute of Technology, Zurich E-Mail: daniel@vis.inf.ethz.ch (NeXT-Mail welcome) [ Pager 152-99-54082 ]
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 13 Nov 1995 05:15:10 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <486k8u$9c6@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: amm@NeXT.COM (Alan M. Marcum) Newsgroups: comp.sys.next.bugs Subject: Re: Swaping on two files impossible ? Date: 13 Nov 1995 05:50:23 GMT Organization: NeXT Computer, Inc. Message-ID: <486mav$fh@news.next.com> References: <4860tj$nis@neptune.ethz.ch> Daniel Gerald Kluge writes > 2. If the systems tries to turn on the swapping, I get the > following error: > mount: Device busy > mount: giving up on: > /private/vm/swapfile.front > mach_swapon: mach_swapon failed: Device busy This happens when you use swapfile compression on a second file. In fact, the second swapfile is enabled, though the compression is enabled only on the first. Just add the nocompress option to your swaptab for all but one of your entries. -- Alan M. Marcum NeXT Tech Support amm@NeXT.COM
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Re: Swaping on two files impossible ? Date: 13 Nov 1995 08:12:21 GMT Organization: Technical University of Berlin, Germany Message-ID: <486ul5$ktd@news.cs.tu-berlin.de> References: <4860tj$nis@neptune.ethz.ch> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Daniel Gerald Kluge (daniel@vis.inf.ethz.ch) wrote: [...two swapfiles?...] It not possible to _compress_ both swapfiles. Thus the error message on the attempt to mount a second compressed swapfile volume. Multiple swapfiles worked just fine last time I tried (on 3.3 Black). : To verify, just try by hand to enable a second swapfile... : is this broken by now ? No (see above) : I remember, that we run once the computer I'm writing on with two swapfiles.. : -daniel : P.S. Where did /usr/man/man5/fstab.5 go to in 3.3 ? It's gone. This _is_ a bug. Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.bugs Subject: Re: Audio CD plays my speakers w/3.3 patch! Date: Mon, 13 Nov 1995 12:28:15 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951113122501.16927C-100000@hphalle10b.informatik.tu-muenchen.de> References: <480ka2$jlj@news.duke.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <480ka2$jlj@news.duke.edu> On 10 Nov 1995, Lee Altenberg wrote: > Hey, after installing the 3.3patch, my CDROM now sends its > audio output to my Sound Blaster 16 (you have to have them > connected) and out through my speakers! So it appears that Hm, this already worked for me before installing the patch. But there seems to be another new/old bug: Try adjusting the balance slider, and now move the volume slider. Result: the balance slider gets misaligned due to the movement of the volume slider. BTW ist there a key combination to adjust the balance? You surely know (on Intel Boxes) Insert/Delete=Volume, Command-Insert/Delete=InputLevel. Best regards, Boerny. -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/ scholz@ve1.rm.op.dlr.de http://www.leo.org/~scholz/
From: wilkie@narya. (Alexander Wilkie) Newsgroups: comp.sys.next.bugs Subject: Re: cognition of Audio-CDs locking the system with 3.3Patch Date: 13 Nov 1995 10:50:38 GMT Organization: Vienna University of Technology, Austria Message-ID: <4877tu$vn@news.tuwien.ac.at> References: <Pine.HPP.3.91.951110104424.22151C-100000@hphalle7a.informatik.tu-muenchen.de> Hi! Same problem with my system.(P90, Adaptec 1542, SoundBlaster16 8 & 16 Bit DMA, NEXTIME Sound, Serial Mouse, old serial drivers) On the console the following appears. Once CDPlayer has been launched, though, an inserted CD comes up as quickly as usual. probing for DOS probing for mac probing for cdaudio Nov 13 11:45:34 Workspace: Mounted scsi disk at /cdaudio trackEntry: Track number out of range cdutil: Catalog failed. ys Alexander Wilkie --- e-mail: wilkie@cg.tuwien.ac.at (NeXTMail, MIME preferred) www : http://www.cg.tuwien.ac.at/~wilkie/ -- e-mail: wilkie@cg.tuwien.ac.at (NeXTMail, MIME preferred) www : http://www.cg.tuwien.ac.at/~wilkie/
From: charlie@why.az.stratus.com (Charles_Spitzer) Newsgroups: comp.sys.next.bugs Subject: Mail.app bug/change in 3.3 Date: 13 Nov 1995 16:48:57 GMT Organization: Stratus Computer Inc, Marlboro MA Message-ID: <487stp$cuv@transfer.stratus.com> I recently installed 3.3 User on my black NeXT. A couple days later I read some mail that came in from an Internet mailing list, and replied to one of the pieces of mail. I have the read receipt requested turned on in my preferences, so of course Mail.app sent my mail off to the mailing list with this flag on. I then read my own post, which sent the read receipt to the mailing list (because it was the sender) instead of to myself. I then read the receipt, which then sent another read receipt mail back to the list. It took me a couple of tries to figure this out, at a result of 1000+ pieces of mail that the remailer had to send out. 3.2 Mail.app didn't do this. I'd consider this a bug and a regression. -- Charles Spitzer charlie@az.stratus.com Customer Assistance Center Stratus Computer, Inc. Phoenix, AZ 85018
Newsgroups: comp.sys.next.bugs,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.software,comp.soft-sys.nextstep,comp.sys.next.bugs From: s2404675@techst02.technion.ac.il (Shkolnik Uri ) Subject: HELP Please Organization: Technion, Israel Institute of Technology Date: Tue, 14 Nov 1995 08:16:36 GMT Message-ID: <DI0xnp.4At@discus.technion.ac.il> Sender: news@wang.com Hello, I'am a computers administrator in a small Israeli company. We have 2 NexTStep which I had never touch before (the other parts of our system are sun's & PC's). Two days ago I tryed to change some E-mail configuration in one of the NexTStep, but its system crashed !!! I found only one User Guide book and I know nobody in Israel that know anything about this operation system. May you help me with: 1) Advice about good NexTStep administration book(s). 2) Internet NexTStep information sites. 3) Help me by E-mail for a while (I have poor English as you see but I'll do my best..). Thanks -- ****************************************************************** * Uri Shkolnik * * ============ * * Harmonic Lightwaves (Israel) * * Phone: 972 4 550180 / Fax ~181 * * E-Mail: root@harmonic.co.il * * [or]: uri@harmonic.co.il [or]: s2404675@t2.technion.ac.il * ****************************************************************** Newsgroups: comp.sys.next.hardware Summary: Keywords: -- ****************************************************************** * Uri Shkolnik * * ============ * * Harmonic Lightwaves (Israel) * * Phone: 972 4 550180 / Fax ~181 * * E-Mail: root@harmonic.co.il * * [or]: uri@harmonic.co.il [or]: s2404675@t2.technion.ac.il * ******************************************************************
From: eh@e_nx_eh.nl.stratus.com (Erik Hoogers) Newsgroups: comp.sys.next.bugs Subject: New NCR/Symbios SCSI driver fails on my system Date: 14 Nov 1995 13:17:49 GMT Organization: Stratus Computer Inc, Marlboro MA Message-ID: <48a4tt$rtj@transfer.stratus.com> Keywords: SCSI PCI The following is what is got on my screen when booting with the new SCSI driver that has been on Synbios' shelf for more than 6 months, and still does not work in my system. PCI bus support enabled Registering: PCI0 sc0: Probing for device Symbios Logic SCSI Adapter instance 0. Board count = 0001 ROMSearchComplete unmap ROM etc... intnum = 9 dmachan = 0 ROM addr 0xc000 ROM len 29696 PATH 0 is a Rev. 3.00x CAMcore (R) at C000 with IMAGE=0022:0BF8 IRQ=09 DMA=0 BusNowOperational path=0 intnum=9 ROM addr 0xc000 PATH0, ID0, LUN0 is MICROP 1588-15MB1036511 PATH0, ID1, LUN0 is DEC DSP3105S PATH0, ID2, LUN0 is ARCHIVE VIPER 150 21247 Registering sc0: ... here it takes about one minute to get to the next messages .... unexpected kernel trap d eip 219b0c frame 37f7f48 called by 21a067 args 58 22053c 0 3819f00 frame 37f7f88 called by 219f2c args 22053c 0 e 3819f00 frame 37f7fa4 called by 219792 args 22053c 22053c 29302ac 29302ac frame 37f7fd0 called by 218cb4 args 0 37f7ffc 15dcf3 29302ac Failed instruction exception (2,d,220000) Waiting for remote debugger connection. (Type 'c' to continue or 'r' to reboot) _ My PC is an FIC/Leo Computer 486-DX2-66 using the ISA/VLB/PCI 'VIO' board. I start Next_3.3 from an IDE disk, the PCI SCSI card is installed in slot # 0, the BusMaster slot, the PCI S3 VideoCard is installed in slot 4. Depending on what other devices are configured, the IMAGE address is changing, I also recorded it to be at 0021:97D0 or 0022:CFC4. The ROM len of 29696 or 0x7400, I believe to be correct. But the actual address in my PC is at 0xE800 and is part of the onboard BIOS, it is not on the PCI card itself. Furthermore, it is not a true ROM image either, I can write and change the code at will as soon as i have enabled it in the setup. Using MS-DOS, I can use this PCI-SCSI adapter without any problems Question : What does 'registering sc0' mean, since it somehow already used the PCI SCSI card to scan for attached devices. It there anyone outthere who use this driver successfully ? - Erik -
From: peter@beast.math.ualberta.ca (Peter Karbaliotis) Newsgroups: comp.sys.next.bugs Subject: Re: Swaping on two files impossible ? Date: 14 Nov 1995 17:41:49 GMT Organization: University of Alberta, Edmonton, Canada Message-ID: <48akct$15lu@pulp.ucs.ualberta.ca> References: <4860tj$nis@neptune.ethz.ch> In article <4860tj$nis@neptune.ethz.ch> daniel@vis.inf.ethz.ch (Daniel Gerald Kluge) writes: .. >P.S. Where did /usr/man/man5/fstab.5 go to in 3.3 ? > See /usr/man/man5/mntent.5 -- Peter Karbaliotis - University of Alberta - NeXTmail & MIME ok
From: Dean_Reece@NeXT.COM(Dean Reece) Newsgroups: comp.sys.next.bugs Subject: Re: IBM Token RIng Driver. Date: 14 Nov 1995 18:16:12 GMT Organization: NeXT Computer, Inc. Distribution: world Message-ID: <48amdc$al7@news.next.com> References: <JDT.31.000E6189@bgsm.edu> There is a known problem with our source routing (SR) code which might be the cause of the behaviour you are seeing. The problem will occur if the NEXTSTEP token ring server and client machines are not on the same ring. In this case, the client's BOOTP request will reach the server fine, but the server's response will not contain any SR information, and therefore will not make it to the client's ring. There is a workaround which you can try to see if it helps the situation. Boot the client machine single-user by typing '-s' at the boot prompt. At the shell prompt type: driverLoader d=IBMTokenRing Respond with 'y' to the next two prompts, the driver should now be loaded and initialized. Then type: ifconfig tr0 <ip address> netmask <netmask> where <ip address> and <netmask> correspond to the appropriate values. Exit the shell by typing ctrl-D. The rest of network startup should proceed normally. We are looking into fixing this problem, and will hopefully have a proper solution in the near future. -Dean John Turner writes | Well I am working with the 3.33 Beta driver for the IBM Auto 16/4 Token | Ring ISA Adapter under NS 3.3 for Intel. When it works it works fine, | but I am having problems when the machine is rebooted that it is unable | to talk to the network and no routing information gets added to the route | tables. | | This problem is fixed by repeated reboots and some wishful thinking. | I am wondering if others have seen/heard of this problem. As I said if | the reboot works then I can talk to everyone, but if it fails I can't see | a sole. No messages are printed when the drivers get setup, it is no | until the client tries to find the Netinfo server do I see a problem. | | John. | | jdturner@bgsm.edu
From: ingo@ipge.toppoint.de (Ingo Prochaska) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: 3.3 Patch:INND does not run anymore! Date: 15 Nov 1995 08:13:11 +0100 Organization: Private Site, Kiel, Germany Message-ID: <48c3u7$2ar@ipge.toppoint.de> References: <47td0g$a5o@commserv.fb1.fhtw-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit dave@turbocat.snafu.de (David Wetzel) writes: >Hi! >After installing the 3.3 Patch on my i586, >I noticed that my inn does not work anymore. It works with installed patch wonderful for me: news@ipge:~$ ctlinnd mode Server running Allowing remote connections Parameters c 10 i 0 (0) l 0 o 243 t 300 normal specified Not reserved Readers separate enabled >Nov 9 17:29:53 turbocat inndstart: inndstart cant bind Address already in >use This sounds like an used 119 port. Used by whom? I could only guess: Perhaps your inetd.conf contains a Line, which tels inetd to listen on 119, so inetd is using 119 and innd can't bind to it. news@ipge:~$ grep 119 /etc/services nntp 119/tcp readnews untp # USENET News Transfer Protocol news@ipge:~$ grep nntp /etc/inetd.conf #nntp stream tcp nowait root /usr/etc/tcpd /usr/local/bin/nntpd >What shall I do? I was using inn 1.4. Take a look at /etc/inetd.conf, and comment out the nntp-line. -- Ingo Prochaska, Olshausenstr. 20, 24118 Kiel, +049 431 84638
From: giddings@menominee.menominee.chem.wisc.edu (Michael Giddings) Newsgroups: comp.sys.next.bugs Subject: Sendmail Queueing - NS3.3/patch Date: 15 Nov 1995 18:03:00 GMT Organization: University of Wisconsin, Madison Distribution: world Message-ID: <48da0k$1ekc@news.doit.wisc.edu> After the 3.3 patch was installed, I am now having problems with mail that is queued when the machine is not connected to the network. Instead of queueing, it immediately bounces mail back, stating "Host unknown (Authoritative answer from name server): No such file or directory". I remember someone else mentioning this. Is there a fix for this new problem? Thanks in advance. -- ---------------------------------------------------------------------------- - Michael Giddings \ Tcl definitions that apply: UW Madison Dept. of Chemistry \ set job "Consultant and Graduate Student" Madison, Wisconsin \ set specialty "Scientific Computation" (608) 692-2851 \ set InRealLife "Whitewater kayaker and\ giddings@whitewater.chem.wisc.edu or\ outdoor enthusiast" giddings@students.wisc.edu \ ---------------------------------------------------------------------------- --
From: daniel@vis.inf.ethz.ch (Daniel Gerald Kluge) Newsgroups: comp.sys.next.bugs Subject: E-Mail lost ! (isn't UNIX mature yet? ) Date: 16 Nov 1995 23:15:15 GMT Organization: Verein der Informatikstudierenden an der ETH Zuerich Message-ID: <48ggm3$agd@elna.ethz.ch> From time to time, my NeXT crashes while executing the uucp queue, which isn't that bad, appart from the 10 minutes it takes, until I can restart my work. But today, I actually lost an e-mail message, the one, on which the mail execution crashed. The UUCP logfile clearly indicates, that there where 2 jobs comming, but in the end I only got one E-mail in Mail.app. It's somehow strange, that such thing can happen, even after the 20+ years of constant improvement UNIX had. BTW I use taylor uucp 1.06.1 and sendmail 8.6.12, what else should I upgrade to avoid this problem in the future ? somebody sugestet once, to use procmail as well... UNIX better die soon... -daniel -- Daniel G. Kluge - cs-student @ Federal Instiute of Technology Zurich E-Mail: daniel@vis.inf.ethz.ch (NeXT-Mail welcome) [ Pager 152-99-54082 ]
From: Calvin Soon-Aun Yeap <yeap@finance.commerce.ubc.ca> Newsgroups: comp.sys.next.bugs Subject: test Date: Thu, 16 Nov 1995 14:22:16 -0800 Organization: Finance Division, UBC Message-ID: <30ABB998.789B@finance.commerce.ubc.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit test
From: Peter Beerli <beerli@genetics.washington.edu> Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: Thu, 16 Nov 1995 18:21:13 -0800 Organization: University of Washington Message-ID: <30ABF199.41C6@genetics.washington.edu> References: <48da0k$1ekc@news.doit.wisc.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I have a similar problem, but I encounter unresolved addresses even if I am connected to the network, under 3.3 (unpatched) this worked fine. -> sendmail failed with established connection, see following excerpt. If this is trivial please forgive, I am NO sendmail guru. Peter zork# /usr/lib/sendmail -v beerli@genetics.washington.edu < /tmp/gaga beerli@genetics.washington.edu... Connecting to mailhost (ether)... beerli@genetics.washington.edu... 550 Host unknown (Authoritative answer from n ame server): No such file or directory Saving message in /Users/xy/dead.letter /Users/xy/dead.letter... 550 Can't create output zork# finger beerli@genetics.washington.edu [genetics.washington.edu] Login name: beerli In real life: Peter Beerli Directory: /usr/users/beerli Shell: /usr/local/bin/tcsh
From: perkins@cps.msu.edu (Stephen J. Perkins) Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: 17 Nov 1995 13:57:26 GMT Organization: Michigan State University Message-ID: <48i4c6$unc@msunews.cl.msu.edu> References: <48da0k$1ekc@news.doit.wisc.edu> <30ABF199.41C6@genetics.washington.edu> > zork# /usr/lib/sendmail -v beerli@genetics.washington.edu < /tmp/gaga > beerli@genetics.washington.edu... Connecting to mailhost (ether)... > beerli@genetics.washington.edu... 550 Host unknown (Authoritative answer > from n > ame server): No such file or directory > Saving message in /Users/xy/dead.letter > /Users/xy/dead.letter... 550 Can't create output > zork# finger beerli@genetics.washington.edu > [genetics.washington.edu] > Login name: beerli In real life: Peter Beerli > Directory: /usr/users/beerli Shell: /usr/local/bin/tcsh > Use nslookup to find out which nameserver is being contacted. If for some reason, your machine is serving names (and its not a name server) then you will run into problems. - Steve -- ============================================================== Stephen J. Perkins | mailto:perkins@cps.msu.edu Dept. of Comp. Science | NeXT, MIME, finger for PGP Michigan State University | NeXT OS 3.3 using PPP-2.3 NeXT PPP-2.2 info at http://www.thoughtport.com:8080/PPP/
From: ken@mutt.com (Ken Craig) Newsgroups: comp.sys.next.bugs,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.software,comp.soft-sys.nextstep,comp.sys.next.bugs Subject: Re: HELP Please Date: Fri, 17 Nov 1995 10:01:24 -0500 Organization: Information Management Inc. Distribution: inet Message-ID: <ken-1711951001250001@witsmac5993.turner.com> References: <DI0xnp.4At@discus.technion.ac.il> In article <DI0xnp.4At@discus.technion.ac.il>, s2404675@techst02.technion.ac.il (Shkolnik Uri ) wrote: >1) Advice about good NexTStep administration book(s). >2) Internet NexTStep information sites. >3) Help me by E-mail for a while (I have poor English as you see but I'll > do my best..). > I'm not 100% sure what comes with what distributions and what's been installed on your machine, so this might not help. You should have a directory called /NextApps which contains an application called Librarian (everybody has this). Run the app. At the top of the window is a bookshelf of on-line manuals. If there's something called NeXTAdmin just click on it, and in the text area below, enter a topic and search. There's also a way to view all titles, then browse through the manual on your own. If there is NOT a NeXTAdmin manual in the bookshelf, you'll need to find it and add it. Leave Librarian running (move the window off to the side, you'll need it visible later) and switch back to your directory listings. Look in the directory /NeXTLibrary/Documentation. You should (hopefully) see a directory named NextAdmin. If you do, highlight this, then drag the icon (just above the list) onto the bookshelf of the Librarian window (you'll see a ghost image appear in the Librarian when you've hit the right spot). Then, continue with the searching/viewing as above. OK - if you don't have NeXTAdmin in your /NeXTLibrary/Documentation directory, then you hopefully have the NeXTStep CD for your machine. Put it in your CD ROM Drive - it will appear in your browser window. On this disk, you should find a /NeXTLibrary/Documentation directory with the Admin manual. You can drag it directly onto the Librarian window as above, it'll just be slower cause it's on CD. As far as the Internet NeXTStep sites, there are a few. Probably the best starting point is http://www.next.com If any of the above advice sounds unreliable it's because I used to own a NextStation (sold it about 18 months ago) and I'm only just now getting back into working with one so I'm relearning most everything about them. I've received enough help from people on USENET that I'm always happy to help. Good luck! Ken
From: kelley@mudpot.ATMOS.ColoState.Edu Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: 17 Nov 1995 14:39:56 GMT Organization: Colorado State University, Fort Collins, CO 80523 Message-ID: <48i6rs$3f33@yuma.ACNS.ColoState.EDU> References: <48da0k$1ekc@news.doit.wisc.edu> <30ABF199.41C6@genetics.washington.edu> <48i4c6$unc@msunews.cl.msu.edu> i've run into similar problems. i installed the 3.3 patch on 4 m68k's and then my ns/hppa server. the black machines are just fine. the hp, however, has no name resolving capabilities anymore. our name resolver, on campus, is still there and I CAN NSLOOKUP from this hp correctly. the resolv.conf is ok. but i cannot send mail from this machine. i can receive but get errors mailing: mudpot> mail -v kelley@tundra.atmos.colostate.edu Subject: dork dork EOT kelley@tundra.atmos.colostate.edu... Connecting to tundra.ATMOS.ColoState.EDU (ether)... kelley@tundra.atmos.colostate.edu... Connecting to yuma.ACNS.ColoState.EDU (ether)... kelley@tundra.atmos.colostate.edu... 550 Host unknown (Valid name but no A or MX record) this hp, mudpot, is not a nameserver. doh! kelley wittmeyer dept of atmospheric science colorado state university kelley@kiwi.atmos.colostate.edu
From: bl03@uwrf.edu (BENJAMIN A LINDSTROM) Newsgroups: comp.sys.next.bugs Subject: Re: E-Mail lost ! (isn't UNIX mature yet? ) Date: 17 Nov 1995 21:06:57 GMT Message-ID: <48ithh$hat@daffy.anetsrvcs.uwrf.edu> References: <48ggm3$agd@elna.ethz.ch> <Pine.NXT.3.91.951117121414.1118E-100000@capitalist.princeton.edu> Timothy J. Luoma (luomat@capitalist.princeton.edu) wrote: : > : > UNIX better die soon... : > : : I hate to disagree with your sign-off, but I can't help it. : agreed. : If you think UNIX handling of email is bad, try doing it in DOS. I had : to spend the last few months on a DOS email system and it is terrible. : The mail system only retained the 'to' 'from' and 'subject' header. : MicroSloth's mail apparently doesn't understand about 'Reply-to' lines. : Often times the 'to' line which appeared was entirely wrong. Mail over a : certain size was converted into an attachment, which meant having to open : it up in a different application. : What's worse is Mac to UNIX pop mail transfers. =) My boss lost a group of e-mail when he locked his machine while retrieving his mail..so he rebooted and tried again...In the end he had 10 popmailer processes out there that had to be killed and he losted his mail. It's not so much that problem of UNIX it's the problem of set-up. UUCP is OLD...And (IHMO) should be destroyed before UNIX is destroyed.=) : I hate when I lose email, even one message, but I think that you : indicated yourself that this was an uncommon occurrence. : Agree...I get a lot of e-mail (fan-mail/hate-mail/techincal bull) and even one of them being lost can cause problems. : DOS might try hard to act like it knows how to handle the internet, but : it is a long way off... : You have to admit if you go up to the next level with Windows and you use Eudura that it works really slick after you take the 10 minutes and set it up...It's not very smart...but it's good for popmail stuff.
From: Dan_Danz <ddan@az.stratus.com> Newsgroups: comp.sys.next.bugs Subject: Re: Mail.app bug/change in 3.3 Date: Fri, 17 Nov 1995 19:52:12 -0800 Organization: Customer Assistance Center, Stratus Computer, Inc. Message-ID: <30AD586C.4CBC@az.stratus.com> References: <487stp$cuv@transfer.stratus.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Charles_Spitzer <charlie@why.az.stratus.com> Charles_Spitzer wrote: > > I recently installed 3.3 User on my black NeXT. A couple days later I read > some mail that came in from an Internet mailing list, and replied to one of > the pieces of mail. I have the read receipt requested turned on in my > preferences, so of course Mail.app sent my mail off to the mailing list with > this flag on. I then read my own post, which sent the read receipt to the > mailing list (because it was the sender) instead of to myself. I then read the > receipt, which then sent another read receipt mail back to the list. It took > me a couple of tries to figure this out, at a result of 1000+ pieces of mail > that the remailer had to send out. > > 3.2 Mail.app didn't do this. I'd consider this a bug and a regression. You might find that the remailer has mucked up your message; in general remailers SHOULD NEVER pass on a receipt header, but most of them are stupid. But the really bad thing is that they set "reply-to" to the list, and therefore when your system had to send a receipt, it MUST obey the reply-to if one is given. This is the usual cause of nonsense like this. I don't know why 3.2 might have done it differently. Was it the same remailer involved on 3.2 and 3.3 ....
From: daniel@vis.inf.ethz.ch (Daniel Gerald Kluge) Newsgroups: comp.sys.next.bugs Subject: Re: E-Mail lost ! (isn't UNIX mature yet? ) Date: 18 Nov 1995 10:45:47 GMT Organization: Verein der Informatikstudierenden an der ETH Zuerich Message-ID: <48kdgr$1a3@elna.ethz.ch> References: <48ggm3$agd@elna.ethz.ch> <Pine.NXT.3.91.951117121414.1118E-100000@capitalist.princeton.edu> luomat@capitalist.princeton.edu (Timothy J. Luoma) wrote in comp.sys.next.bugs: : > UNIX better die soon... : > : I hate to disagree with your sign-off, but I can't help it. : : If you think UNIX handling of email is bad, try doing it in DOS. Sorry, but that's not the point ! I'm working with software (UNIX, UUCP, sendmail) that was developped for a longer time, that I myself work with computers, so I think one can expect, that it makes some progress, in the time it took me to learn to walk, jump, speak in 3 different languages, comprehend how computer and software actually work, etc. If you really want (and can read german), I can forward you my rants about DOS & Mac, because I have the urge to express my frustration every time a system screwed me, and this time it was UNIX. BTW DOS & Mac screw harder, that's true. -daniel -- Daniel G. Kluge - cs-student @ Federal Institute of Technology Zurich E-Mail: daniel@vis.inf.ethz.ch (NeXT-Mail welcome) [ Pager 152-99-54082 ]
From: daniel@vis.inf.ethz.ch (Daniel Gerald Kluge) Newsgroups: comp.sys.next.bugs Subject: E-Mail lost , not really Date: 18 Nov 1995 15:30:32 GMT Organization: Verein der Informatikstudierenden an der ETH Zuerich Message-ID: <48ku6o$7vs@neptune.ethz.ch> It did it again, but as you will see in the logfiles, gracfully. This means, the machine didn't crash, but did a clean reboot, which didn't take that long, but gcc was just about to finish preprocessing all Header-Files, which it does now again... I apologize for my claim that the E-Mail is lost; it isn't really! Well, there are just some 'Bogus queue files' lying around in /usr/spool/mqueue, which can only be retrieved by root, and which will NOT get delivered by a 'sendmail -q' !! Of course the files are owned by uucp, and mode 600, so only the superuser can, if he has the time, see them through, and do whatever he sees fit. Still, a crash every other poll, and mail files locked in /usr/where/the/heck is NOT acceptable, my friends using Macs to read their mail have less problems, well they have others :-) so long, -daniel Here the logfiles, the names have been changed to protect the innocent... /usr/adm/messages: Nov 18 15:49:14 milliways mach: trap: type 0x10 fcode 0 rw 0 faultaddr 0x233a Nov 18 15:49:14 milliways mach: trap: pc 0x40011a4 sp 0x3fff424 sr 0x2208 Nov 18 15:49:14 milliways mach: trap: cpu 0 th 0x101dcff0 proc 0x10188780 pid 5124 pcb 0x101dd300 Nov 18 15:49:14 milliways mach: traceback: fp 0x4001188 Nov 18 15:49:14 milliways mach: called from pc 0x00000009 fp 0x00000008 4-args 0000233a 00000000 00002700 00000000 Nov 18 15:49:14 milliways mach: last fp 0x8 Nov 18 15:49:14 milliways mach: panic: (Cpu 0) illegal instruction Nov 18 15:49:14 milliways mach: NeXT ROM Monitor 2.5 v66 Nov 18 15:49:14 milliways mach: panic: NeXT Mach 3.3: Mon May 22 17:56:06 PDT 1995; root(rcbuilder):mk-171.12.obj~11/RC_m68k/RELEASE_M68K Nov 18 15:49:14 milliways mach: Nov 18 15:49:19 milliways syslogd: going down on signal 15 Nov 18 15:50:02 milliways mach: Killing all processes Nov 18 15:50:02 milliways mach: continuing Nov 18 15:50:02 milliways mach: unmounting Data ... done Nov 18 15:50:02 milliways mach: unmounting swapfile.front ... done Nov 18 15:50:02 milliways mach: unmounting home ... done Nov 18 15:50:02 milliways mach: unmounting swapdisk ... done Nov 18 15:50:02 milliways mach: rebooting Mach... Nov 18 15:50:02 milliways mach: Nov 18 15:50:02 milliways mach: init arg: -xx [...] Nov 18 15:50:14 milliways sendmail[159]: PAA05099: bogus queue file, uid=4, mode=100600 Nov 18 15:50:14 milliways sendmail[159]: PAA05131: bogus queue file, uid=4, mode=100600 /usr/spool/uucp: uucico nessie - (1995-11-18 15:48:38.27 4588) Handshake successful (protocol 'f') uucico nessie daniel (1995-11-18 15:48:38.81 4588) Sending rmail somebody@somwhere.dom (D.milliwaC0080) (643 bytes) uucico nessie daniel (1995-11-18 15:48:41.50 4588) Sending rmail somebody@elsewhre.dom (D.milliwaC0082) (900 bytes) uucico nessie daemon (1995-11-18 15:48:45.82 4588) Receiving D.nessic8949e7 uucico nessie daemon (1995-11-18 15:48:48.32 4588) Receiving X.nessieCc894 uucico nessie daemon (1995-11-18 15:48:49.66 4588) Receiving D.nessic897713 uucico nessie daemon (1995-11-18 15:48:51.59 4588) Receiving X.nessieCc897 uucico nessie daemon (1995-11-18 15:48:52.56 4588) Receiving D.nessic898229 uucico nessie daemon (1995-11-18 15:48:54.17 4588) Receiving X.nessieCc898 uucico nessie - (1995-11-18 15:48:54.85 4588) Protocol 'f': sent 1779 bytes for 1731, received 6100 bytes for 5933 uucico nessie - (1995-11-18 15:48:57.06 4588) Call complete (23 seconds 7664 bytes 333 bps) uuxqt nessie daemon (1995-11-18 15:49:08.69 5088) Executing X.nessieCc894 (rmail daniel) uuxqt nessie daemon (1995-11-18 15:49:13.36 5088) Executing X.nessieCc897 (rmail daniel) uuxqt nessie daemon (1995-11-18 15:49:18.29 5088) ERROR: Got termination signal uuxqt nessie daemon (1995-11-18 15:49:18.30 5088) ERROR: Execution: Got signal 15 uuxqt nessie daemon (1995-11-18 15:49:18.30 5088) Execution failed (X.nessieCc897) uuxqt nessie - (1995-11-18 15:54:22.12 293) ERROR: Stale lock nessie/L.nessieCc897 held by process 5088 created 1995-11-18 15:49:13 uuxqt nessie daemon (1995-11-18 15:54:27.23 293) Executing X.nessieCc897 (rmail daniel) uuxqt nessie daemon (1995-11-18 15:54:30.11 293) Executing X.nessieCc898 (rmail daniel) /usr/spool/mqueue/syslog: Nov 18 15:47:54 milliways sendmail[4540]: PAA04504: to=Wesemann <somebody@elsewhre.dom>, ctladdr=daniel (100/20), delay=00:00:04, mailer=uucp-dom, relay=nessie, stat=Sent Nov 18 15:49:10 milliways sendmail[5099]: /etc/sendmail/aliases: 9 aliases, longest 10 bytes, 150 bytes total Nov 18 15:49:11 milliways sendmail[5099]: PAA05099: from=provider.dom!friend, size=2270, class=0, pri=32270, nrcpts=1, msgid=<9511181149.AA00664@harka>, relay=uucp@localhost Nov 18 15:49:16 milliways sendmail[5131]: /etc/sendmail/aliases: 9 aliases, longest 10 bytes, 150 bytes total Nov 18 15:50:14 milliways sendmail[158]: starting daemon (8.6.12): SMTP+queueing@01:00:00 Nov 18 15:50:14 milliways sendmail[159]: /etc/sendmail/aliases: 9 aliases, longest 10 bytes, 150 bytes total Nov 18 15:50:14 milliways sendmail[159]: PAA05099: bogus queue file, uid=4, mode=100600 Nov 18 15:50:14 milliways sendmail[159]: PAA05131: bogus queue file, uid=4, mode=100600 Nov 18 15:50:22 milliways sendmail[206]: /etc/sendmail/aliases: 9 aliases, longest 10 bytes, 150 bytes total Nov 18 15:50:23 milliways sendmail[206]: PAA00206: from=root, size=101, class=0, pri=30101, nrcpts=1, msgid=<199511181450.PAA00206@milliways.hallus.swill.org>, relay=root@localhost Nov 18 15:54:16 milliways sendmail[218]: PAA00206: to=daniel, ctladdr=root (0/1), delay=00:03:54, mailer=local, stat=Sent Nov 18 15:54:28 milliways sendmail[296]: /etc/sendmail/aliases: 9 aliases, longest 10 bytes, 150 bytes total Nov 18 15:54:28 milliways sendmail[296]: PAA00296: from=hype.comp.com!daemon, size=1656, class=-60, pri=139656, nrcpts=1, msgid=<9511173432.AA009341@not.your.business.com>, relay=uucp@localhost Nov 18 15:54:30 milliways sendmail[298]: PAA00296: to=daniel, ctladdr=hype.comp.com!daemon (4/20), delay=00:00:02, mailer=local, stat=Sent Nov 18 15:54:31 milliways sendmail[303]: /etc/sendmail/aliases: 9 aliases, longest 10 bytes, 150 bytes total Nov 18 15:54:31 milliways sendmail[303]: PAA00303: from=hype.comp.com!daemon, size=1086, class=-60, pri=139086, nrcpts=1, msgid=<9511192343690_23847623@actually.aol.com>, relay=uucp@localhost Nov 18 15:54:32 milliways sendmail[306]: PAA00303: to=daniel, ctladdr=hype.comp.com!daemon (4/20), delay=00:00:01, mailer=local, stat=Sent --- Daniel G. Kluge @ Swiss Federal Institute of Technology Zuerich E-Mail: Internet : daniel@hallus.swill.org (NeXT-Mail welcome) daniel@vis.inf.ethz.ch (PGP-keyID: 7D5C6D) Pager: numeric : 152-99-54082 (inside Switzerland only)
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: Mail.app bug/change in 3.3 Date: 18 Nov 1995 16:52:34 GMT Organization: Information Technology Solutions, Inc. Message-ID: <48l30i$rij@news.its.com> References: <487stp$cuv@transfer.stratus.com> [ reformatted to ~75 columns ] charlie@why.az.stratus.com (Charles_Spitzer) wrote: > I recently installed 3.3 User on my black NeXT. A couple days later I > read some mail that came in from an Internet mailing list, and replied > to one of the pieces of mail. I have the read receipt requested turned > on in my preferences, so of course Mail.app sent my mail off to the > mailing list with this flag on. I then read my own post, which sent the > read receipt to the mailing list (because it was the sender) instead of > to myself. I then read the receipt, which then sent another read receipt > mail back to the list. It took me a couple of tries to figure this out, > at a result of 1000+ pieces of mail that the remailer had to send out. That sounds roughly like the correct behavior. Why did you send mail to a mailing list with read-receipt enabled? That's not a good idea at all. I also think read-receipts in general aren't useful (since they don't really mean much) and are an invasion of privacy. I've got a filter that stops all read-receipts.... > 3.2 Mail.app didn't do this. I'd consider this a bug and a regression. Perhaps so. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: pete@ohm.york.ac.uk (-bat.) Newsgroups: comp.sys.next.bugs Subject: 3.3 patch - what exactly are the problems ? Date: 18 Nov 1995 19:03:33 GMT Organization: The University of York, UK Message-ID: <48lam5$4sk@netty.york.ac.uk> Ummm.... so, what exactly DOES happen if I install the improved DNS and attempt to run posix applications or staticly linked executables that uuse gethostbyname() ? i.e. waht is the failure mode of these things ? also...can anyone tell me what happens if I build Posix applications with the new libraris (or statically linked exevcutables using gethostbyname()) and try and run them on an old unpatched 3.3 ? do this fail too ? sigh.... -bat.
Message-ID: <Pine.NXT.3.91.951117121414.1118E-100000@capitalist.princeton.edu> Organization: Princeton University References: <48ggm3$agd@elna.ethz.ch> <48ggm3$agd@elna.ethz.ch> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit From: luomat@capitalist.princeton.edu ("Timothy J. Luoma") Subject: Re: E-Mail lost ! (isn't UNIX mature yet? ) Date: 17 Nov 1995 11:19:52 +0000 Newsgroups: comp.sys.next.bugs Distribution: world > > UNIX better die soon... > I hate to disagree with your sign-off, but I can't help it. If you think UNIX handling of email is bad, try doing it in DOS. I had to spend the last few months on a DOS email system and it is terrible. The mail system only retained the 'to' 'from' and 'subject' header. MicroSloth's mail apparently doesn't understand about 'Reply-to' lines. Often times the 'to' line which appeared was entirely wrong. Mail over a certain size was converted into an attachment, which meant having to open it up in a different application. I hate when I lose email, even one message, but I think that you indicated yourself that this was an uncommon occurrence. DOS might try hard to act like it knows how to handle the internet, but it is a long way off... TjL
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.sysadmin Subject: Symbol encoding scheme unusable? Date: Sun, 19 Nov 1995 23:33:52 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951119231956.15531C-100000@hphalle0.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, I was just wondering, wether this is a bug, or wether I'm just to ignorant to get this done: I'm using a font which features some of the symbols described in the NEXTSTEP Symbol Encoding Scheme (positioned correctly) and I want to use them, but: NeXT's symbol encoding scheme seems to map everything to the Symbol Font! E.g. If I type a <trademark> character, the Symbol.font is used instead of my custom font. I can' even remap this. The only solution seems to me not type <trademark> but <icircumflex> and remap this to my symbol font. Unusable to me. This all prevents you from using other symbol fonts, doesn't it? Why shouldn't it be possible to use a font, which lacks e.g. the <icircumflex> character but features the definition for <trademark> and use the <trademark> charakter when I type <trademark> ond does nothing when I type <icircumflex>. The point is: I know that the Adobe encoding scheme only features 256 different characters, but why can't these characters be used? If a font lacks one of the characters NeXT wants in its encoding scheme, and features another which NeXT wants in its symbol encoding schema, it isn't used, because NeXT switches automatically to the Symbol.font. (Even if the character is mapped correctly!) Am I doing something wrong with my handcreated Adobe fonts, or is this a standard behaviour? Best regards, Boerny (hoping that the encoding scheme is better in NS4.0). -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/ scholz@ve1.rm.op.dlr.de http://www.leo.org/~scholz/
From: sepp@ppcger.inka.de (Josef Wolf) Newsgroups: comp.sys.next.bugs Subject: Kernel panics on DOS-FS accesses Date: 18 Nov 1995 22:03:28 +0100 Organization: PPCGer mail and news system (Germersheim, Germany) Message-ID: <48lhn0$qlj@ppcger.inka.de> When doing extensive copying from floppy to the DOS-Partition, the kernel sometimes panics. Sometimes the same panic occures when working with SoftPC. Most of the time when pressing CTRL-ALT-DEL. The panic-window say 'zalloc failed'. Hardware Configuration: Pentium 90 on ASUS P/I-P55TP4XE/90 with 16MB Boot Drivers: Adaptec2940SCSIDriver SerialPorts Floppy PS2Keyboard PCIBus EISABus Drivers: Adaptec2940 System Serial at IRQ 4 3; ports 0x3f8-0x3ff 0x2f8-0x2ff Floppy at IRQ 6; DMA 2; ports 0x3f2-0x3f7 PS2Keyboard at IRQ 1; ports 0x60-0x65 PCIBus EISABus at IRQ 2; ports 0x00-0x0f 0x20-0x21 0x40-0x4b 0x70-0x71 0x81-0x8F 0x92-0x92 0xc0-0xcf PS2Mouse at IRQ 12; System Parallel at IRQ 7; ports 0x378-0x37f Beep ATI Mach64 ports 0x02e8-0x02ef 0x03b0-0x03df 0x06ec-0x06ef 0x0aec-0x0aef 0x0eec-0x0eef 0x12ec-0x12ef 0x16ec-0x16ef 0x1aec-0x1aef 0x1eec-0x1eef 0x42ec-0x42ef 0x46ec-0x46ef 0x4aec-0x4aef 0x4eec-0x4eef 0x52ec-0x52ef 0x5eec-0x5eef 0x62ec-0x62ef 0x6aec-0x6aef 0x6eec-0x6eef 0x72ec-0x72ef 0x76ec-0x76ef map 0xfc000000-0xfc1fffff 0xa0000-0xbffff 0xc0000-0xcffff; Greetings -- Josef Wolf -- sepp@ppcger.inka.de Germersheim, Germany
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 20 Nov 1995 05:15:15 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <48p2t3$ch3@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: mow@navigator.de (Markus Wenzel) Newsgroups: comp.sys.next.bugs Subject: Re: E-Mail lost ! (isn't UNIX mature yet? ) Date: 19 Nov 1995 17:54:09 +0100 Organization: Navigator Message-ID: <48nnfh$u7@marsu.navigator.de> References: <48ggm3$agd@elna.ethz.ch> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Daniel Gerald Kluge (daniel@vis.inf.ethz.ch) wrote: > From time to time, my NeXT crashes while executing the uucp queue, which > isn't that bad, appart from the 10 minutes it takes, until I can restart my > work. Never had such strange behavior. > But today, I actually lost an e-mail message, the one, on which the mail > execution crashed. You never lost a letter delivered by snail mail? > It's somehow strange, that such thing can happen, even after the 20+ years > of constant improvement UNIX had. It is more strange that people keep on buying Win 95 after 5+ years of constant crashes with Win 3.x. > UNIX better die soon... Nonsensical conclusion. Your problem has nothing at all to do with Unix itself. -- who? // Dipl.-Phys. Markus Wenzel work? // Navigator - IT Consulting & System administration mail? // mow@navigator.de more? // http://www.pilhuhn.de/~marsu/
Newsgroups: comp.sys.next.bugs From: brad@instep.bc.ca Subject: BUG: NEXTSTEP 3.3 Project Builder Message-ID: <1995Nov20.174739.9536@instep.bc.ca> Sender: usenet@instep.bc.ca Organization: InStep Mobile Communications Inc. Date: Mon, 20 Nov 1995 17:47:39 GMT ProjectBuilder.app does not preserve /Net mount path for compiling on remote host. If I attempt to compile a project that is local to the current host, but is mounted network wide, the path PB uses is my local path, not the network path, so that the remote build can't find the working directory and fails. For example, if I compile a project in /Users/brad/MyProject, PB uses this path instead of /Net/vancouver/Users/brad/MyProject (where my home directory is /Net/vancouver/Users/brad). -- Brad Head <brad@instep.bc.ca> Software Developer, InStep Mobile Communications Inc. Vancouver, British Columbia, Canada
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: Re: Mail.app bug/change in 3.3 Date: Mon, 20 Nov 1995 12:19:07 -0500 Organization: Princeton University Message-ID: <Pine.NXT.3.91.951120121348.10809B-100000@capitalist.princeton.edu> References: <487stp$cuv@transfer.stratus.com> <48l30i$rij@news.its.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <48l30i$rij@news.its.com> On 18 Nov 1995, Chuck Swiger wrote: > [ reformatted to ~75 columns ] > charlie@why.az.stratus.com (Charles_Spitzer) wrote: > > I recently installed 3.3 User on my black NeXT. A couple days later I > > read some mail that came in from an Internet mailing list, and replied > > to one of the pieces of mail. I have the read receipt requested turned > > on in my preferences, so of course Mail.app sent my mail off to the > > mailing list with this flag on. I then read my own post, which sent the > > read receipt to the mailing list (because it was the sender) instead of > > to myself. I then read the receipt, which then sent another read receipt > > mail back to the list. It took me a couple of tries to figure this out, > > at a result of 1000+ pieces of mail that the remailer had to send out. > > That sounds roughly like the correct behavior. Why did you send mail to a > mailing list with read-receipt enabled? That's not a good idea at all. > > I also think read-receipts in general aren't useful (since they don't > really mean much) and are an invasion of privacy. I've got a filter that > stops all read-receipts.... Anyone who has Mail3.3 can stop this via a dwrite... something like dwrite Mail ReadReceiptOption X where X is 'yes' 'no' or 'ask' > > > 3.2 Mail.app didn't do this. I'd consider this a bug and a regression. > > Perhaps so. Mail 3.2 did handle something about read-receipts differently, there was something different, but I can't remember what now... read receipts are no longer a viable way to tell if someone has actually read your message. receiptfilter has been around for a couple of years, and others have hacked ways around this too.. TjL
Newsgroups: comp.sys.next.software,comp.sys.next.bugs From: Mark_Tarbell@Radical.Com (Mark Tarbell) Subject: Anyone running DoomII on a NeXTcube? Message-ID: <1995Nov20.180218.10576@Radical.Com> Sender: news@Radical.Com Organization: Radical System Solutions, Inc. Date: Mon, 20 Nov 1995 18:02:18 GMT DoomII seems to run just about everywhere EXCEPT my mono '040 25MHz NeXTcube. Everything's up-to-date (including v20 of libInterceptor.a.shlib). Here's what I get: % DoomII.app/DoomII -scale 2 Call to host_processor_set_priv() failed: invalid port(-102) Running with standard scheduling policy. ioctl MegaPixel fails: 6 framebuffer encoding: KK bitsPerSample: 2 bitsPerPixel: 2 DOOM System Startup v1.8 V_Init: allocate screens. M_LoadDefaults: Load system defaults. Z_Init: Init zone memory allocation daemon. W_Init: Init WADfiles. adding doom1.wad shareware version. M_Init: Init miscellaneous info. R_Init: Init DOOM refresh daemon - [...................] P_Init: Init Playloop state. I_Init: Setting up machine state. D_CheckNetGame: Checking network game status. startskill 2 deathmatch: 0 startmap: 1 startepisode: 1 player 1 of 1 (1 nodes) S_Init: Setting up sound. Swapping sound samples. Discrete sampling rates: 22050.000000 44100.000000 Supported encodings: 600 Stream encoding: SND_FORMAT_LINEAR_16 Device sampling rate = 22050.000000 Using sampling rate of = 22050.000000 shortZero = 0 buffer = 0x1b1350, bufferBytes = 2048 HU_Init: Setting up heads up display. ST_Init: Init status bar. ... and then an error panel pops up with: W_GetNumForName: STTMINUS not found! Any ideas? DoomI.app runs fine, but it'd be nice to finally get sound! Thanks Mark_Tarbell@Radical.Com
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: Re: Mail.app bug/change in 3.3 Date: 20 Nov 1995 19:48:37 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-h-59.usc.edu Message-ID: <48qm2l$96h@usc.edu> References: <487stp$cuv@transfer.stratus.com> <48l30i$rij@news.its.com> <Pine.NXT.3.91.951120121348.10809B-100000@capitalist.princeton.edu> "Timothy J. Luoma" <luomat@capitalist.princeton.edu> wrote: > Anyone who has Mail3.3 can stop this via a dwrite... something like > > dwrite Mail ReadReceiptOption X > > where X is 'yes' 'no' or 'ask' Does anyone know what the correct dwrite is? I tried this one to no avail. -- Thanks and be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV PGP key - send email to reichman@scf.usc.edu with Subject "PGP"
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: Re: Mail.app bug/change in 3.3 Date: 21 Nov 1995 00:10:11 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-i-66.usc.edu Message-ID: <48r5d3$h6l@usc.edu> References: <487stp$cuv@transfer.stratus.com> <48l30i$rij@news.its.com> <Pine.NXT.3.91.951120121348.10809B-100000@capitalist.princeton.edu> c/o Timothy J. Luoma: > Anyone who has Mail3.3 can stop this via a dwrite... something like > > dwrite Mail ReadReceiptOption X > > where X is 'yes' 'no' or 'ask' > which is to say: dwrite Mail ReadReceiptOption NO dwrite Mail ReadReceiptOption YES dwrite Mail ReadReceiptOption ASK -- Thanks and be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV For my PGP key - send email subject "request_PGP"
From: american@internetMCI.COM (Charles C. Hocker) Newsgroups: comp.sys.next.bugs Subject: Are there NeXTSTEP Viruses? Date: 21 Nov 1995 06:43:56 GMT Organization: InternetMCI Message-ID: <48rsfc$crc@news.internetmci.com> Hello, I am not trying create a scare. I am just wondering if there are any known NS viruses, trojan horses, or worms that one should watch-out for. And if there are, is there anyway to protect against them? Charles -- ‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘ Charles C. Hocker War is Peace american@internetMCI.COM Freedom is Slavery american@aztec.asu.edu Ignorance is Strength ASCII, MIME & NeXTmail (preferred) George Orwell's 1984 ‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘
Newsgroups: comp.sys.next.bugs From: jpmeia@netcom.com Subject: Print problem Message-ID: <nntpuserDIEpoI.8x9@netcom.com> Keywords: print, Laserjet Sender: netnews@mork.netcom.com Organization: NETCOM On-line Communication Services (408 261-4700 guest) Date: Tue, 21 Nov 1995 18:50:42 GMT Hi. I've got a Versa notebook running NS3.3 (unpatched), has worked fine for over a year. Today it would not print...sent the print job to the printer (an HPLJ4M), printer lights blink normally for a second or two, then stop. The NS PrintManager print queue is empty, but nothing has printed. No telltale entries in any of the /usr/adm files. The printer checks out 4.0 and is not a problem. I can't correlate to any specific (stupid) action on my part. Any suggestions? Thanks, please respond to: jpmeia@netcom.com nextmail welcome
From: "Alan C. Horn" <ahorn@dreamworks.com> Newsgroups: comp.sys.next.bugs Subject: Swapfiles under NeXTStep3.3 White Date: 22 Nov 1995 03:33:57 GMT Organization: Earthlink Network, Inc. Message-ID: <48u5n5$2e9@mars.earthlink.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Forgive me if I'm making a blatant error and overlooking something incredibly straightforward here, but I'm something of a newcomer to the NeXT OS. I'm experiencing problems with swafiles expanding until they fill the entire system disk on my workstations (intel.. not cubes). NeXT Support told me that this was a known problem, and that the solution was to reboot the machine every so often. This isn't really a viable option in a situation where we need very low downtime. If anyone has any suggestions or alternatives to the existing situation, or if anyone has discovered a workaround to this problem, I'd be very pleased to hear about it. Mail followups would be best I think. Many thanks Al
From: Andrew_Forkes@arzana.co.uk Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: [Q]: Riddle me this: bundles compile -g -O, but not -O Date: Wed, 22 Nov 1995 01:09:59 GMT Organization: Arzana Limited Sender: news@arzana.co.uk Message-ID: <DIF78n.K2v@arzana.co.uk> Keywords: gcc I know this has been reported before (for app compiles), but I can't find the reference to it, does anyone remember the solution ? Why would a bundle compile with -g -O and work fine, but not -O ? Bemused. Andrew ---- Andrew D. Forkes Arzana Limited email: <Andrew_Forkes@arzana.co.uk> (NeXTmail & MIME are fine)
From: Andrew_Forkes@arzana.co.uk Newsgroups: comp.sys.next.bugs Subject: Re: Are there NeXTSTEP Viruses? Date: Wed, 22 Nov 1995 01:01:46 GMT Organization: Arzana Limited Sender: news@arzana.co.uk Message-ID: <DIF6uy.K27@arzana.co.uk> References: <48rsfc$crc@news.internetmci.com> Keywords: Viruses, DOS, NEXTSTEP, Windows In article <48rsfc$crc@news.internetmci.com> american@internetMCI.COM (Charles C. Hocker) writes: > > Hello, > I am not trying create a scare. I am just wondering if there are any > known NS viruses, trojan horses, or worms that one should watch-out for. > And if there are, is there anyway to protect against them? > One potential loop-hole in NS that I have considered is this: If you are running NS and SoftPC (or NS and Windows on a dual boot), and you inherit a virus via a DOS-floppy, it is feasible that the virus may treat the your hard-disk as if it's a DOS formatted disk. This could prove a problem if the virus likes it lurk in the boot sector of a disk (as many PC viruses do), because your disk's boot sector will not be in a DOS format. Therefore, the virus may unintentionally corrupt your entire hard disk, by simply trying to propagate itself to the boot sector of your NS hard disk, whereby over-writing the inode index (?) of your entire NS system - not a pretty prospect. I've never heard of this *actually* happening, but I think it's the closest possibility of NS "catching" a DOS virus. I'd be more than happy to be proved wrong though. The solution (if the above is possible), never insert a DOS disk into your NS/Windows (virtual or otherwise) machine before virus-checking it on another real DOS system first. Never heard of any specific NS viruses though (a few unix ones). It's a shame really, it might encourage more people to back their systems up more regularly 8-/ later, Andrew ---- Andrew D. Forkes Arzana Limited email: <Andrew_Forkes@arzana.co.uk> (NeXTmail & MIME) "Peace cannot be achieved by force, only by understanding." - Albert Einstein.
Newsgroups: comp.sys.next.bugs From: ian.stephenson@insignia.co.uk Subject: Re: Are there NeXTSTEP Viruses? Message-ID: <DIG7nC.J5q@isltd.insignia.com> Sender: news@isltd.insignia.com Organization: Insignia Solutions plc References: <DIF6uy.K27@arzana.co.uk> Date: Wed, 22 Nov 1995 14:16:24 GMT In article <DIF6uy.K27@arzana.co.uk> Andrew_Forkes@arzana.co.uk writes: > In article <48rsfc$crc@news.internetmci.com> american@internetMCI.COM > (Charles C. Hocker) writes: > > > > Hello, > > I am not trying create a scare. I am just wondering if there are any > > known NS viruses, trojan horses, or worms that one should watch-out for. > > And if there are, is there anyway to protect against them? > > > > One potential loop-hole in NS that I have considered is this: > > If you are running NS and SoftPC (or NS and Windows on a dual boot), and > you inherit a virus via a DOS-floppy, it is feasible that the virus may > treat the your hard-disk as if it's a DOS formatted disk. > This was a consideration when we recently revamped some of the support for DOS partitions in SoftPC. SoftPC can no longer access the real boot sector of the hard disc, and certainly can't do anything with the NeXT partition(s). If can only access the actuall DOS bit of the disc. Any attempt to access either the Boot sector or outside the partition are stopped, and an alert box warning of a potential virus is displayed. I won't say that it can't happen, but I think it's unlikely. If you're worried about this then don't mount a real partition as the D drive, and your NeXT OS is completly safe - access to the host file system is done through ordinary OS calls, so there's no way anything can happen. Simlarly hdf's are self contained, and can't corrupt your NeXT system (though an hdf could HOLD a virus it would only affect SoftPC - this has happened on other Unix platforms). Ian (the above comements are not official comments of insignia - just my opinions),
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: Wed, 22 Nov 1995 02:07:00 -0500 Organization: Princeton University Message-ID: <Pine.NXT.3.91.951122020058.2771A-100000@capitalist.princeton.edu> References: <48u5n5$2e9@mars.earthlink.net> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII To: ahorn@dreamworks.com In-Reply-To: <48u5n5$2e9@mars.earthlink.net> the swapfile will grow until you reboot. There is no way around it. You can set the highwater mark in /etc/rc.swap which will keep your system from freezing up because the disk space filled, but you will stil have to reboot. Your options are thus: 1) program a reboot at some point during the day when usage is low 2) set the hiwat and reboot only when needed 3) buy another HD and use that just for swapping... it will fill up, but less often. How big depends on your usage. 4) Turn your computer off and read a book. There is a recurring call for freeing swapspace without rebooting... so far, no one has found a way. TjL -- Timothy J. Luoma luomat@capitalist.princeton.edu (MIME fine! NeXTMail if necessary) 476tjl@ptsmail.ptsem.edu (UGH! DOS! Yuck.... Don't use this) "If I understood the man page, I wouldn't have asked the question."
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: 22 Nov 1995 18:51:23 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-d-22.usc.edu Message-ID: <48vrfb$6rd@usc.edu> References: <48u5n5$2e9@mars.earthlink.net> <Pine.NXT.3.91.951122020058.2771A-100000@capitalist.princeton.edu> "Timothy J. Luoma" <luomat@capitalist.princeton.edu> wrote: > 1) program a reboot at some point during the day when usage is low How do yo do this? -- Thanks and be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV For my PGP key - send email subject "request_PGP"
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <flexus!rfschtkt@maze.ruca.ua.ac.be> Date: Wed, 22 Nov 95 19:33:31 +0100 From: Raf Schietekat <flexus!rfschtkt@maze.ruca.ua.ac.be> Message-ID: <9511221833.AA01151@flexus> Subject: Re: E-Mail lost , not really Daniel Gerald Kluge <daniel@vis.inf.ethz.ch> writes: > I apologize for my claim that the E-Mail is lost; it > isn't really! Perhaps not in this case, but for a more complete view, you might want to check out KBNS.32.2.018: UUCP will lose some messages predictably (except perhaps in the event of a crash :-), but then the message will be lost later on). Includes a description of a workaround. ftp://ftp.cs.orst.edu/software/NeXT/documents/KBNS.33.1.README.rtf.gz ftp://ftp.cs.orst.edu/software/NeXT/documents/KBNS.33.1.rtf.gz Raf Schietekat, RfSchtkt@maze.ruca.ua.ac.be (NeXTmail), Flanders, Belgium If I don't answer: my mail relay can't handle !, % or .uucp, I think *** The year 2000 will be the last year of the 20th century. ***
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: A little buggy humor (LONG) Date: 22 Nov 1995 23:26:14 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-d-14.usc.edu Message-ID: <490bim$du7@usc.edu> From comp.risks Date: Fri, 27 Oct 1995 10:31:43 +0100 From: Klaus Brunnstein <brunnstein@rz.informatik.uni-hamburg.d400.de> Subject: Mr.Bill Gates: MS software essentially bug-free In an interview for German weekly magazine FOCUS (nr.43, October 23,1995, pages 206-212), Microsoft`s Mr. Bill Gates has made some statements about software quality of MS products. After lengthy inquiries about how PCs should and could be used (including some angry comments on some questions which Mr. Gates evidently did not like), the interviewer comes to storage requirements of MS products; it ends with the following dispute (translated by submitter; at some interesting points, I added the German phrase): Focus: But it is a fact: if you buy a new version of a program to overcome faults of an old one, you unavoidably get more features and need more storage. Gates: We have strong competitors and produce only products which we believe to be able to sell. New versions are not offered to cure faults. I have never heard of a less relevant reason to bring a new version on the market. Focus: There are always bugs in programs. Gates: No. There are no essential bugs ("keine bedeutenden Fehler") in our software which a significant number fo users might wish to be removed. Focus: Hey? I get always crazy when my Macintosh Word 5.1 hides page numbers under my text. Gates: Maybe you make errors, have you ever thought about that? It often appears that machine addicts ("Maschinenstuermer") cannot use software properly. We install new features because we were asked to. Nobody would buy a new software because of bugs in an old one. Focus: If I call a hotline or a dealer and complain about a problem, I have to hear: `Get the update to version 6`. Everybody has such experiences. This is how the system works. Gates: We pay 500 million $ a year for telephone advice. Less than 1% of calls which we get has to do with software bugs. Most callers wish advice. You are kindly invited to listen to the millions of calls. You must wait for weeks until one complains about a bug. Focus: But where does this feeling of frustration come from which unites PC users? Everybody is confronted every day that programs do not work as they should? Gates: That is talking, following the motto: `yes, I also know about this bug`. I understand this as sociological phenomenon, not as technical. The RISK? While there is NO risk that experienced users believe Mr. Gates, there are 2 serious ones: first, that politicians (who rarely experience the lowlands of PCs but develop their "political visions" from their unexperience) may believe him. Second and worst: that Mr. Gates and his enterprise believe what he is saying, and act accordingly :-) Maybe someone can inform Mr. Gates that it was HIS enterprise which recently distributed the first Macro virus WordMacro.Concept on a CD-ROM to OEM customers, in July, and to participants of a Windows 95 seminar in Germany, in September); but indeed, this is NOT a BUG BUT an ATTACK on unaware users :-) According to a German saying those whose reputation is corrupted may live free and easy ("Ist der Ruf erst ruiniert, lebt sich's doppelt ungeniert!") Klaus Brunnstein (October 27,1995) [... und nicht ingeniert (which is not pronounced engineert!) PGN] -- Thanks and be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV For my PGP key - send email subject "request_PGP"
From: magnan@mathcn.umontreal.ca (Francois Magnan) Newsgroups: comp.sys.next.software,comp.sys.next.bugs Subject: Re: Anyone running DoomII on a NeXTcube? Date: Wed, 22 Nov 1995 19:39:16 -0500 Organization: University of Montreal Distribution: world Message-ID: <magnan-2211951939200001@10.0.2.15> References: <1995Nov20.180218.10576@Radical.Com> In article <1995Nov20.180218.10576@Radical.Com>, Mark_Tarbell@Radical.Com (Mark Tarbell) wrote: > DoomII seems to run just about everywhere EXCEPT my mono '040 25MHz > NeXTcube. Everything's up-to-date (including v20 of libInterceptor.a.shlib). > Here's what I get: > > % DoomII.app/DoomII -scale 2 > Call to host_processor_set_priv() failed: invalid port(-102) > Running with standard scheduling policy. > ioctl MegaPixel fails: 6 > framebuffer encoding: KK > bitsPerSample: 2 bitsPerPixel: 2 > DOOM System Startup v1.8 > V_Init: allocate screens. > M_LoadDefaults: Load system defaults. > Z_Init: Init zone memory allocation daemon. > W_Init: Init WADfiles. > adding doom1.wad > shareware version. > M_Init: Init miscellaneous info. > R_Init: Init DOOM refresh daemon - [...................] > P_Init: Init Playloop state. > I_Init: Setting up machine state. > D_CheckNetGame: Checking network game status. > startskill 2 deathmatch: 0 startmap: 1 startepisode: 1 > player 1 of 1 (1 nodes) > S_Init: Setting up sound. > Swapping sound samples. > Discrete sampling rates: > 22050.000000 > 44100.000000 > Supported encodings: > 600 > Stream encoding: SND_FORMAT_LINEAR_16 > Device sampling rate = 22050.000000 > Using sampling rate of = 22050.000000 > shortZero = 0 > buffer = 0x1b1350, bufferBytes = 2048 > HU_Init: Setting up heads up display. > ST_Init: Init status bar. > > ... and then an error panel pops up with: > > W_GetNumForName: STTMINUS not found! > > > Any ideas? DoomI.app runs fine, but it'd be nice to finally get sound! > > Thanks > Mark_Tarbell@Radical.Com ItŒs not worth it. Even if you make it work the game is not playable on a NeXT Cube. Much too slow. Francois Magnan -- Francois Magnan magnan@mathcn.umontreal.ca
From: beerli@genetics.washington.edu (Peter Beerli) Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: 23 Nov 1995 04:24:26 GMT Organization: University of Washington Message-ID: <490t1q$g75@nntp5.u.washington.edu> References: <48da0k$1ekc@news.doit.wisc.edu> <30ABF199.41C6@genetics.washington.edu> <48i4c6$unc@msunews.cl.msu.edu> <30ACE370.41C6@genetics.washington.edu> Until now I received no helpful response on the problem that sendmail does not deliver any mail from my Next system (black 3.3 patched, unposixed) [of course everything was working before the patch] Some mentioned the named and nmserver, but this is NOT the problem, if I specify the e-mail address somehow fritz@university.edu it will fail, but knowing on which machine fritz works -> sending to fritz@fritzensmaschine.university.edu will go through, Am I the only person with this problem ? Or is it so obvious that there is no response to this? Peter Beerli beerli@genetics.washington.edu
From: chin@clark.net (Chin) Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: 22 Nov 1995 22:03:55 -0500 Organization: Clark Internet Services, Inc., Ellicott City, MD USA Message-ID: <490oar$4p2@clark.net> References: <48u5n5$2e9@mars.earthlink.net> <Pine.NXT.3.91.951122020058.2771A-100000@capitalist.princeton.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit "Timothy J. Luoma" <luomat@capitalist.princeton.edu> writes: >the swapfile will grow until you reboot. There is no way around it. You >can set the highwater mark in /etc/rc.swap which will keep your system >from freezing up because the disk space filled, but you will stil have to >reboot. Generally, the swapfile will grow to the size of the largest amount of memory you are using at one time. If the memory in use at the end of the swapfile is freed, then the swapfile will be truncated. The problem stems from the fact that holes develop as processes interleave memory allocation/deallocation. if the memory freed is not at the end of the swapfile, then the swapfile will not shrink. NS will reuse the holes in the swapfile (for the most part) however. But you could end up in a scenario where an app chewed up a whole bunch of VM, then say, NetInfo allocates a page after that, and then your big app exits and frees all that memory. Your swapfile doesn't truncate since NetInfo is still holding a page at the end of the swapfile. >Your options are thus: >1) program a reboot at some point during the day when usage is low >2) set the hiwat and reboot only when needed >3) buy another HD and use that just for swapping... it will fill up, but >less often. How big depends on your usage. >4) Turn your computer off and read a book. Also try upgrading to 3.3 Patch. It alters the way the window server allocates memory and has a better chance of freeing memory and shrinking the swapfile. Also, if it is your window server holding up a few pages of memory at the end of your swapfile, then you can try logging out and at the login panel type "exit" and hit return. That should restart the window server, freeing up it's memory and hopefully re-allocating in earlier portions of the swapfile. Recently, I had chewed up 150mb of swapspace doing some tests... I killed the big process that was chewing up all the memory and that brought the swapfile down to 120mb. I killed a few more apps/daemons that had allocated memory recently and it dropped to 90mb. I then logged out and went into console mode (type "console" at the login panel). That kills the window server. The swapfile dropped to 50mb. I had to reboot to drop all the way to 16mb. So... swapfile truncating does work. The answer is, have enough swap space for what you are doing. Use less memory (at one time) and/or buy more swap space. ..Bill
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: 23 Nov 1995 08:50:30 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-f-70.usc.edu Message-ID: <491ckm$1ll@usc.edu> References: <48da0k$1ekc@news.doit.wisc.edu> <30ABF199.41C6@genetics.washington.edu> <48i4c6$unc@msunews.cl.msu.edu> <30ACE370.41C6@genetics.washington.edu> <490t1q$g75@nntp5.u.washington.edu> beerli@genetics.washington.edu (Peter Beerli) wrote: [snip] > if I specify the e-mail address somehow fritz@university.edu it > will fail, but knowing on which machine fritz works -> sending > to fritz@fritzensmaschine.university.edu will go through You might try inserting your the name of your university's mail server, e.g.: fritz@[mailserver].university.edu -- Thanks and be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV For my PGP key - send email subject "request_PGP"
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: Re: [Q]: Riddle me this: bundles compile -g -O, but not -O Date: 23 Nov 1995 15:25:16 GMT Organization: Information Technology Solutions, Inc. Message-ID: <4923os$fnh@news.its.com> References: <DIF78n.K2v@arzana.co.uk> Andrew_Forkes@arzana.co.uk wrote: > I know this has been reported before (for app compiles), but I can't find > the reference to it, does anyone remember the solution ? > > Why would a bundle compile with -g -O and work fine, but not -O ? If the bundle doesn't load (rld_ errors), you've stripped too many symbols from it and/or the main executable, and the dynamic loader can't resolve everything. Make sure you've set APP_STRIP_OPTS to $(DYLD_APP_STRIP_OPTS) in Makefile.postamble. If the bundle loads but crashes, you've likely got a problem with uninitialized variables or something similar. Make sure you get zero warnings when compiling with "-g -O -Wall". -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: 23 Nov 1995 15:34:09 GMT Organization: Information Technology Solutions, Inc. Message-ID: <49249h$fnh@news.its.com> References: <48u5n5$2e9@mars.earthlink.net> "Alan C. Horn" <ahorn@dreamworks.com> wrote: > I'm experiencing problems with swafiles expanding until they fill the entire > system disk on my workstations (intel.. not cubes). NeXT Support told me that > this was a known problem, and that the solution was to reboot the machine every > so often. This isn't really a viable option in a situation where we need very > low downtime. > > If anyone has any suggestions or alternatives to the existing situation, or if > anyone has discovered a workaround to this problem, I'd be very pleased to > hear about it. It's very simple: get more swapspace by purchasing another hard drive for your high-availability machines. At ~ $0.30 per MB, this is by far the most cost-effective solution. You can also add more memory to your systems, which will both improve performance and reduce paging activity, but that's a more expensive route. Also, be sure that you don't have a long-running process which leaks memory. Some performance monitors (early versions of KPerfMon, for example), and even Preferences.app will leak memory slowly and fill up your swapspace. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs From: tom@hukatronic.cz (Tomas Hurka) Subject: Bug: return value of the currentCountForChannel: method is always '1' Message-ID: <DIIAwF.xB@hurka.UUCP> Sender: tom@hurka.UUCP (Tomas Hurka) Organization: Hukatronic (H.C.C.) Date: Thu, 23 Nov 1995 17:21:50 GMT Hi All, While writing the NEXTSTEP driver for the data acquisition ISA card I discovered the bug in the DriverKit IODirectDevice class. The return value of the currentCountForChannel: method was always '1', regardless the actual value of the DMA word count register. This is valid for the 16-bit DMA channels and probably also for 8-bit DMA channels. This bug also affects the method isDMADone:, which does not work too. The only work around is to read the value directly from the DMA word count register. I have NEXTSTEP 3.3/Intel (of course) with the latest 3.3 User patch installed. I sent report about this bug to bug_next@next.com too. Best regards, -- Tomas Hurka tom@hukatronic.cz NeXTMAIL and MIME OK (international mail <50 KB accepted) 0 while the correct transformation is Page=0x0 Addr MSB=0xc0 Addr LSB=0x00 The problem in the NEXTSTEP code is that only lower 16 bits (up to A15) are shifted right. The correct approach is to shift right lower 17 bits (up to A16). Therefore the transformation is done wrong if the bit A16 in the physical address is one. The same error is in reading the DMA address, so if you read and write DMA address via DriverKit methods everything looks good, but the data goes to the different place than your driver expect. The only work around is to set/read the DMA address for 16-bit channels manually using direct access to the DMA registers. I would like to thank to Eric Norum <eric@skatter.USask.Ca> for checking my conclusions. I have NEXTSTEP 3.3/Intel (of course) with the latest 3.3 User patch installed. I sent report about this bug to bug_next@next.com too. Best regards, -- Tomas Hurka tom@hukatronic.cz NeXTMAIL and MIME OK (international mail <50 KB accepted)
Newsgroups: comp.sys.next.bugs From: kjell@oops.se(Kjell Nilsson) Subject: PPP2.2 on HP Message-ID: <DIJHws.KGv@oops.se> Sender: usenet@oops.se Organization: OOPS art, HB Date: Fri, 24 Nov 1995 08:50:51 GMT I tried to run ppp2.2 on my HP 715/100 using NextStep 3.3. It crached the kernel. (Can it be the serial port :?) Is there anybody that knows if the latest 3.3 patch fixes this? -- -- Kjell Nilsson OOPS art NeXTMail, Mime welcome kjell@oops.se
From: perkins@cps.msu.edu (Stephen J. Perkins) Newsgroups: comp.sys.next.bugs Subject: Re: PPP2.2 on HP Date: 24 Nov 1995 17:13:18 GMT Organization: Michigan State University Message-ID: <494ufe$gn9@msunews.cl.msu.edu> References: <DIJHws.KGv@oops.se> In <DIJHws.KGv@oops.se> Kjell Nilsson wrote: > I tried to run ppp2.2 on my HP 715/100 using NextStep 3.3. It crached the > kernel. > (Can it be the serial port :?) > Is there anybody that knows if the latest 3.3 patch fixes this? PPP-2.2 does not run on HPPA running OS 3.3. I don't have a local machine against which I can debug... Sorry. - Steve -- ============================================================== Stephen J. Perkins | mailto:perkins@cps.msu.edu Dept. of Comp. Science | NeXT, MIME, finger for PGP Michigan State University | NeXT OS 3.3 using PPP-2.3 NeXT PPP-2.2 info at http://www.thoughtport.com:8080/PPP/
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: 24 Nov 1995 16:04:09 GMT Organization: Information Technology Solutions, Inc. Message-ID: <494qdp$kop@news.its.com> References: <48u5n5$2e9@mars.earthlink.net> <Pine.NXT.3.91.951122020058.2771A-100000@capitalist.princeton.edu> <48vrfb$6rd@usc.edu> reichman@scf.usc.edu wrote: > "Timothy J. Luoma" <luomat@capitalist.princeton.edu> wrote: >> 1) program a reboot at some point during the day when usage is low > > How do yo do this? Execute /usr/etc/reboot via cron at something like 3 AM. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: Fri, 24 Nov 1995 13:22:49 -0500 Organization: Princeton University Message-ID: <Pine.NXT.3.91.951124131727.6852C-100000@capitalist.princeton.edu> References: <48u5n5$2e9@mars.earthlink.net> <Pine.NXT.3.91.951122020058.2771A-100000@capitalist.princeton.edu> <48vrfb$6rd@usc.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <48vrfb$6rd@usc.edu> > "Timothy J. Luoma" <luomat@capitalist.princeton.edu> wrote: > > 1) program a reboot at some point during the day when usage is low > > How do yo do this? well, strictly speaking I don't do this. I was guessing that it could be done with 'reboot' and/or 'shutdown' in a cron job at around 5am or so (some length of time after your other cron jobs). You'll probably want to check to make sure no one is logged on before running this, in case you have early-bird or night-owl users ;-) I *heard* that someone else was doing this, but can't find the old post... I probably didn't save it. I'd be interested in hearing from anyone who gets this to work. TjL -- Timothy J. Luoma luomat@capitalist.princeton.edu (MIME fine! NeXTMail if necessary) 476tjl@ptsmail.ptsem.edu (UGH! DOS! Yuck.... Don't use this) "If I understood the man page, I wouldn't have asked the question."
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.hardware,comp.sys.next.bugs Subject: BIOS error with Symbios (NCR) driver?! Date: Fri, 24 Nov 1995 23:29:48 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951124232053.4507H-100000@hphalle0.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, I'm using an Intel Plato (Premiere II) board with BIOS revision 12AX1 I'm using an NCR SCSI II controller successfully in conjunction with an Quantum Maverick, and SEAGATE 1GB drive. Now I recently tried to install a new Conner 2 GB (Conner 2105S). The following problems occured: lowlevel format worked nice under DOS utility by NCR. fdisk worked nice under DOS format worked nice under DOS now under NEXTSTEP: mounting the disk (DOS file system) worked nice for the first partitition and failed for the rest, giving failures: 'mount can't get raw size of disk'. fdisk fails: Bogus BIOS information. disk fails: I/O error writing disk label. The drivers all recognize the disk and can read the disk size (reported during startup). This shi* happens with both: the original SymbiosLogic driver and the new beta driver by NeXT. So (of course) why does this happen? Is there really something wrong with my BIOS? And if so, why doesn't this happen to affect DOS? Any help and hints are pretty much appreciated, becausse I'm going to perchuse an even larger driver for this configuration and don't want to loose all the money on a disk which is unusable under NEXTSTEP. Best regards, Boerny. -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/ scholz@ve1.rm.op.dlr.de http://www.leo.org/~scholz/
From: gta@earthlink.net (Benoit Pecqueur) Newsgroups: comp.sys.next.bugs Subject: ###########Free Mac classified ads & tech support############## Date: 25 Nov 1995 01:23:07 GMT Organization: gta Message-ID: <gta-2411951724580001@gta.earthlink.net> ###########Free Mac classified ads & tech support############## http:/www.gta-net.com
From: berezaw@river.it.gvsu.edu (Bill Bereza) Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: 25 Nov 1995 03:45:05 GMT Organization: My Bedroom, Ravenna, MI Message-ID: <4963g1$pl3@news.it.gvsu.edu> References: <48da0k$1ekc@news.doit.wisc.edu> <48i4c6$unc@msunews.cl.msu.edu> <30ACE370.41C6@genetics.washington.edu> <490t1q$g75@nntp5.u.washington.edu> In article <490t1q$g75@nntp5.u.washington.edu>, Peter Beerli <: beerli@genetics.washington.edu> wrote: >Until now I received no helpful response on the problem >that sendmail does not deliver any mail from my Next system (black 3.3 >patched, unposixed) [of course everything was working before the patch] After I installed the 3.3 patch on NSFIP, my mail was also no longer queued. I'm not sure what the real cause of the problem is, but I did find a workaround. I just put an entry for my mail host into the Netinfo hosts database. -- Bill Bereza berezaw@river.it.gvsu.edu <NeXT/MIME> Beware of all enterprises that require new clothes.
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: 25 Nov 1995 10:49:23 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-h-32.usc.edu Message-ID: <496sbj$9ba@usc.edu> References: <48da0k$1ekc@news.doit.wisc.edu> <48i4c6$unc@msunews.cl.msu.edu> <30ACE370.41C6@genetics.washington.edu> <490t1q$g75@nntp5.u.washington.edu> <4963g1$pl3@news.it.gvsu.edu> berezaw@river.it.gvsu.edu (Bill Bereza) wrote: > I just put an entry for my mail host into the Netinfo hosts database. I don't believe this is a workaround. Sendmail 8.7.* is set up to use netinfo and fqdn, and, as I understood it, what you did is part of the normal setup. -- Thanks and be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV For my PGP key - send email subject "request_PGP"
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: Sat, 25 Nov 1995 12:30:25 -0500 Organization: Princeton University Message-ID: <Pine.NXT.3.91.951125122733.1419E-100000@capitalist.princeton.edu> References: <48da0k$1ekc@news.doit.wisc.edu> <30ABF199.41C6@genetics.washington.edu> <48i4c6$unc@msunews.cl.msu.edu> <30ACE370.41C6@genetics.washington.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <30ACE370.41C6@genetics.washington.edu> On Fri, 17 Nov 1995, Peter Beerli wrote: > > What is mail/sendmail really doing before it responds with > "...connecting to mailhost (ether)...", I am worrying especially about > the "ether"? I believe that is set in /etc/sendmail/sendmail.cf (which is probably a link to another file in that directory. Open it up and take a look. TjL -- Timothy J. Luoma luomat@capitalist.princeton.edu (MIME fine! NeXTMail if necessary) 476tjl@ptsmail.ptsem.edu (UGH! DOS! Yuck.... Don't use this) "If I understood the man page, I wouldn't have asked the question."
From: berezaw@river.it.gvsu.edu (Bill Bereza) Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: 25 Nov 1995 22:00:45 GMT Organization: My Bedroom, Ravenna, MI Message-ID: <4983md$7k0@news.it.gvsu.edu> References: <48u5n5$2e9@mars.earthlink.net> <Pine.NXT.3.91.951122020058.2771A-100000@capitalist.princeton.edu> <48vrfb$6rd@usc.edu> <Pine.NXT.3.91.951124131727.6852C-100000@capitalist.princeton.edu> In article <Pine.NXT.3.91.951124131727.6852C-100000@capitalist.princeton.edu>, Timothy J. Luoma <luomat@capitalist.princeton.edu> wrote: > >> "Timothy J. Luoma" <luomat@capitalist.princeton.edu> wrote: >> > 1) program a reboot at some point during the day when usage is low >> >> How do yo do this? > >I *heard* that someone else was doing this, but can't find the old >post... I probably didn't save it. I'd be interested in hearing from >anyone who gets this to work. My university has the NS workstations setup to reboot every day. Here's the /etc/crontab.local entry: 45 3 * * * root /etc/shutdown -r +15 Regularly scheduled shutdown -- Bill Bereza berezaw@river.it.gvsu.edu <NeXT/MIME> Beware of all enterprises that require new clothes.
From: robert@steffi.accessone.com (Robert Nicholson) Newsgroups: comp.sys.next.bugs Subject: make ... infinite recursion? Date: 24 Nov 1995 16:05:38 -0800 Organization: x Sender: robert@steffi.accessone.com Message-ID: <eu43tv9rh.fsf@steffi.accessone.com> Anybody seen make crash like this before? root:/tmp/tpage-2_40>gdb make 16:03 GDB is free software and you are welcome to distribute copies of it under certain conditions; type "show copying" to see the conditions. There is absolutely no warranty for GDB; type "show warranty" for details. GDB 4.7 (NeXT 3.1), Copyright 1992 Free Software Foundation, Inc... Reading symbols from /bin/make...(no debugging symbols found)...done. Reading symbols from /usr/shlib/libsys_s.B.shlib...done. (gdb) r do Starting program: /bin/make do Program generated(1): Memory access exception on address 0x3eff29c (invalid address). 0x940c in ?? () (gdb) where #0 0x940c in ?? () #1 0x93dc in ?? () #2 0x938c in ?? () #3 0x4fc8 in ?? () #4 0x504e in ?? () #5 0x504e in ?? () #6 0x504e in ?? () #7 0x504e in ?? () #8 0x504e in ?? () #9 0x504e in ?? () #10 0x504e in ?? () #11 0x504e in ?? () #12 0x504e in ?? () #13 0x504e in ?? () #14 0x504e in ?? () #15 0x504e in ?? () #16 0x504e in ?? () #17 0x504e in ?? () #18 0x504e in ?? () #19 0x504e in ?? () #20 0x504e in ?? () #21 0x504e in ?? () #22 0x504e in ?? () #23 0x504e in ?? () #24 0x504e in ?? () #25 0x504e in ?? () #26 0x504e in ?? () #27 0x504e in ?? () #28 0x504e in ?? () #29 0x504e in ?? () #30 0x504e in ?? () #31 0x504e in ?? () #32 0x504e in ?? () #33 0x504e in ?? () #34 0x504e in ?? () #35 0x504e in ?? () #36 0x504e in ?? () #37 0x504e in ?? () #38 0x504e in ?? () #39 0x504e in ?? () #40 0x504e in ?? () #41 0x504e in ?? () #42 0x504e in ?? () #43 0x504e in ?? () #44 0x504e in ?? () #45 0x504e in ?? () #46 0x504e in ?? () #47 0x504e in ?? () #48 0x504e in ?? () #49 0x504e in ?? () #50 0x504e in ?? () #51 0x504e in ?? () #52 0x504e in ?? () #53 0x504e in ?? () #54 0x504e in ?? () #55 0x504e in ?? () #56 0x504e in ?? () #57 0x504e in ?? () #58 0x504e in ?? () #59 0x504e in ?? () #60 0x504e in ?? () #61 0x504e in ?? () #62 0x504e in ?? () #63 0x504e in ?? () #64 0x504e in ?? () #65 0x504e in ?? () #66 0x504e in ?? () #67 0x504e in ?? () #68 0x504e in ?? () #69 0x504e in ?? () #70 0x504e in ?? () #71 0x504e in ?? () #72 0x504e in ?? () #73 0x504e in ?? () #74 0x504e in ?? () #75 0x504e in ?? () #76 0x504e in ?? () #77 0x504e in ?? () #78 0x504e in ?? () #79 0x504e in ?? () #80 0x504e in ?? () #81 0x504e in ?? () #82 0x504e in ?? () #83 0x504e in ?? () #84 0x504e in ?? () #85 0x504e in ?? () #86 0x504e in ?? () #87 0x504e in ?? () #88 0x504e in ?? () #89 0x504e in ?? () #90 0x504e in ?? () #91 0x504e in ?? () #92 0x504e in ?? () #93 0x504e in ?? () #94 0x504e in ?? () #95 0x504e in ?? () #96 0x504e in ?? () #97 0x504e in ?? () #98 0x504e in ?? () #99 0x504e in ?? () (More stack frames follow...) (gdb) Even more interesting is... root:/tmp/tpage-2_40>make -d 16:10 Reading internal rules. doname(all, 0) doname(do, 1) unify(do, %): % : do doname(fixpl, 2) unify(fixpl, %): % : fixpl doname(tpage, 3) unify(tpage, %): % : tpage doname(tpage.pl, 4) unify(tpage.pl, %): % : tpage.pl doname(tpage.pl.pl, 5) unify(tpage.pl.pl, %): % : tpage.pl.pl doname(tpage.pl.pl.pl, 6) unify(tpage.pl.pl.pl, %): % : tpage.pl.pl.pl doname(tpage.pl.pl.pl.pl, 7) unify(tpage.pl.pl.pl.pl, %): % : tpage.pl.pl.pl.pl doname(tpage.pl.pl.pl.pl.pl, 8) unify(tpage.pl.pl.pl.pl.pl, %): % : tpage.pl.pl.pl.pl.pl doname(tpage.pl.pl.pl.pl.pl.pl, 9) unify(tpage.pl.pl.pl.pl.pl.pl, %): % : tpage.pl.pl.pl.pl.pl.pl doname(tpage.pl.pl.pl.pl.pl.pl.pl, 10) unify(tpage.pl.pl.pl.pl.pl.pl.pl, %): % : tpage.pl.pl.pl.pl.pl.pl.pl doname(tpage.pl.pl.pl.pl.pl.pl.pl.pl, 11) unify(tpage.pl.pl.pl.pl.pl.pl.pl.pl, %): % : tpage.pl.pl.pl.pl.pl.pl.pl.pl doname(tpage.pl.pl.pl.pl.pl.pl.pl.pl.pl, 12) unify(tpage.pl.pl.pl.pl.pl.pl.pl.pl.pl, %): % : tpage.pl.pl.pl.pl.pl.pl.pl.pl.pl doname(tpage.pl.pl.pl.pl.pl.pl.pl.pl.pl.pl, 13) -- "Mary ate a little lamb and punk rock isn't dead" (PGP key: send email with Subject: request pgp key)
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: 26 Nov 1995 05:05:20 GMT Organization: Information Technology Solutions, Inc. Message-ID: <498sig$g47@news.its.com> References: <48da0k$1ekc@news.doit.wisc.edu> <30ABF199.41C6@genetics.washington.edu> <48i4c6$unc@msunews.cl.msu.edu> <30ACE370.41C6@genetics.washington.edu> <Pine.NXT.3.91.951125122733.1419E-100000@capitalist.princeton.edu> On Fri, 17 Nov 1995, Peter Beerli wrote: > What is mail/sendmail really doing before it responds with > "...connecting to mailhost (ether)...", I am worrying especially about > the "ether"? It's the name of the mailer selected to handle the mail (based on the "$#" directive in the config file); this mailer tries to connect via a SMTP connection over TCP/IP to the remote address. There is nothing significant about the name...you are free to name your mailers anything you like. Sendmail does a number of things before this response, including placing the mail into the spool directory and (normally) forking a subprocess to deliver the mail while the parent continues to listen for incoming connections. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
Newsgroups: comp.sys.next.bugs From: altenber@uhunix.uhcc.Hawaii.Edu (Lee Altenberg) Subject: My swapfile solution Message-ID: <DIn0q9.56u@news.hawaii.edu> Sender: news@news.hawaii.edu Organization: University of Hawaii Date: Sun, 26 Nov 1995 06:30:08 GMT It seems to me that the obvious answer to the swapfile problems is: allocate a fixed amount of space on the disk to the swapfile, using an entry in /etc/swaptab such as: /private/vm/swapfile lowat=134217728,hiwat=134217728 I chose such a high value (134 MB) because Mathematica is a VM hog. But anyway, this allocates a fixed chunk of the disk to the swapfile. When you reboot, the swapfile stays at 134 MB. As I understand it, the memory within this space just keeps getting automatically recycled. I would think this would prevent disk fragmentation caused by repeatedly growing and shrinking the swapfile. Also, you know that the swapfile can't grow to fill up your disk, once it reaches hiwat. Now, am I missing something about swapfile management that would make this a bad idea? -- Lee Altenberg altenber@mhpcc.edu
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: Re: My swapfile solution Date: 26 Nov 1995 08:33:23 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-d-01.usc.edu Message-ID: <4998oj$6te@usc.edu> References: <DIn0q9.56u@news.hawaii.edu> altenber@uhunix.uhcc.Hawaii.Edu (Lee Altenberg) wrote: > It seems to me that the obvious answer to the swapfile problems > is: allocate a fixed amount of space on the > disk to the swapfile, using an entry in /etc/swaptab such as: > > /private/vm/swapfile lowat=134217728,hiwat=134217728 [snip] > Now, am I missing something about swapfile management that would > make this a bad idea? From what I understand, if your vm requirements then exceeded your hiwater mark, you will crash. -- Thanks and be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV For my PGP key - send email subject "request_PGP"
From: Garance A Drosehn <gad@eclipse.its.rpi.edu> Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: 26 Nov 1995 04:41:01 GMT Organization: Rensselaer Polytechnic Institute, Troy NY, USA Message-ID: <498r4t$alq@usenet.rpi.edu> References: <48u5n5$2e9@mars.earthlink.net> "Alan C. Horn" <ahorn@dreamworks.com> wrote: > I'm experiencing problems with swafiles expanding until they fill > the entire system disk on my workstations (intel.. not cubes). > NeXT Support told me that this was a known problem, and that the > solution was to reboot the machine every so often. This isn't > really a viable option in a situation where we need very low > downtime. > > If anyone has any suggestions or alternatives to the existing > situation, or if anyone has discovered a workaround to this > problem, I'd be very pleased to hear about it. The two workarounds I've used the most are: 1) Buy more RAM (which may alleviate more of the problem than you'd initially expect it to). 2) Buy more disk space for swapfiles. There are also: 3) If you're running NeXTSTEP 3.3, apply the just-released patch for 3.3 4) If you're running NeXTSTEP 3.3 (probably with or without the patch), see what happens with the process for Preferences.app. Depending on what setting you have for the clock displayed in the icon of Preferences.app, you may see a fairly significant virtual memory leak. What is the hardware configuration that you have? How much RAM, and how much disk space is available? --- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
From: wilkie@cg.tuwien.ac.at (Alexander Wilkie) Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: 26 Nov 1995 11:15:01 GMT Organization: Vienna University of Technology, Austria Message-ID: <499i7l$87e@news.tuwien.ac.at> References: <48u5n5$2e9@mars.earthlink.net> <498r4t$alq@usenet.rpi.edu> Garance A Drosehn <gad@eclipse.its.rpi.edu> wrote: > 3) If you're running NeXTSTEP 3.3, apply the just-released > patch for 3.3 Does this have any benefits with respect to vm/swapfiles apart from allegedly fixing the memory leak in Preferences.app? ys Alexander Wilkie -- e-mail: wilkie@cg.tuwien.ac.at (NeXTMail preferred, MIME o.k.) www : http://www.cg.tuwien.ac.at/~wilkie/
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: My swapfile solution Date: 26 Nov 1995 16:48:32 GMT Organization: Information Technology Solutions, Inc. Message-ID: <49a5p0$28g@news.its.com> References: <DIn0q9.56u@news.hawaii.edu> <4998oj$6te@usc.edu> reichman@scf.usc.edu wrote: > altenber@uhunix.uhcc.Hawaii.Edu (Lee Altenberg) wrote: > > It seems to me that the obvious answer to the swapfile problems > > is: allocate a fixed amount of space on the > > disk to the swapfile, using an entry in /etc/swaptab such as: > > > > /private/vm/swapfile lowat=134217728,hiwat=134217728 > [snip] > > Now, am I missing something about swapfile management that would > > make this a bad idea? > > From what I understand, if your vm requirements then exceeded your > hiwater mark, you will crash. Sort of. Various system calls, such as fork(), vm_allocate(), and so forth will fail when you run out of memory. Eventually, vm_allocate() failing will cause malloc() to fail, which will cause the Workspace to be unhappy. But the system itself shouldn't fail completely, until the kernel needs to get more memory and cannot, at which time it will probably panic. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: Re: My swapfile solution Date: Sun, 26 Nov 1995 13:39:02 -0500 Organization: Princeton University Message-ID: <Pine.NXT.3.91.951126133135.2006A-100000@capitalist.princeton.edu> References: <DIn0q9.56u@news.hawaii.edu> <4998oj$6te@usc.edu> <49a5p0$28g@news.its.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <49a5p0$28g@news.its.com> On 26 Nov 1995, Chuck Swiger wrote: > reichman@scf.usc.edu wrote: > > altenber@uhunix.uhcc.Hawaii.Edu (Lee Altenberg) wrote: > > > It seems to me that the obvious answer to the swapfile problems > > > is: allocate a fixed amount of space on the > > > disk to the swapfile, using an entry in /etc/swaptab such as: > > > > > > /private/vm/swapfile lowat=134217728,hiwat=134217728 > > [snip] > > > Now, am I missing something about swapfile management that would > > > make this a bad idea? > > > > From what I understand, if your vm requirements then exceeded your > > hiwater mark, you will crash. > > Sort of. Various system calls, such as fork(), vm_allocate(), and so forth > will fail when you run out of memory. Eventually, vm_allocate() failing > will cause malloc() to fail, which will cause the Workspace to be unhappy. > But the system itself shouldn't fail completely, until the kernel needs to > get more memory and cannot, at which time it will probably panic. > I'm still confused then. I thought setting a HIWAT mark meant that the system would be better off than if it fillled up (with the lovely "pageout failed" message), but this makes it sounds like it is worse. I'm using a swapdisk, which means that when the swapfile hits the hiwat on my swapdisk, it starts swapping on /private/vm/swapfile, so I've made a little script that checks the time stamp on '/private/vm/swapfile' (I use 'prefer' on the first swapfile). If it changes, I pop-up a warning panel to tell me to reboot. -- Timothy J. Luoma luomat@capitalist.princeton.edu (MIME fine! NeXTMail if necessary) 476tjl@ptsmail.ptsem.edu (UGH! DOS! Yuck.... Don't use this) "If I understood the man page, I wouldn't have asked the question."
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 27 Nov 1995 05:15:09 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <49bhgt$a07@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: Alexander.Hauer@ifr.luftfahrt.uni-stuttgart.de (Alexander Hauer) Newsgroups: comp.sys.next.bugs Subject: Re: Print problem Date: 27 Nov 1995 08:07:55 GMT Organization: Comp.Center (RUS), U of Stuttgart, FRG Message-ID: <49brkr$1hig@info4.rus.uni-stuttgart.de> References: <nntpuserDIEpoI.8x9@netcom.com> jpmeia@netcom.com writes > Hi. I've got a Versa notebook running NS3.3 (unpatched), has worked > fine for over a year. Today it would not print...sent the print job > to the printer (an HPLJ4M), printer lights blink normally for a second > or two, then stop. The NS PrintManager print queue is empty, but > nothing has printed. No telltale entries in any of the /usr/adm files. > The printer checks out 4.0 and is not a problem. > > I can't correlate to any specific (stupid) action on my part. > Any suggestions? > > Thanks, please respond to: > > jpmeia@netcom.com > nextmail welcome We had a quite strange problem with NeXT Printers on NS32 black which took some time to locate. If you have a paper-casette with a switch to select LETTER and A4 paper make sure that your papersize (Software) and the setting at your printer is the same otherwise you get a postscript error on your console and no printouts. Maybe that helps. Ciao Alex. -- ___________________________________________________________________ Dipl.-Phys. Alexander A. Hauer Institute of Flightmechanics Systemadministration and Flight Control Phone: (+49)711-121-1432 University of Stuttgart, Forststrasse 86 Fax: (+49)711-634856 D-70176 Stuttgart
From: flight@mathi.uni-heidelberg.de (Gregor Hoffleit) Newsgroups: comp.sys.next.bugs Subject: POSIX append bug not gone with new libposix Date: 27 Nov 1995 15:52:43 GMT Organization: University of Heidelberg, Germany Message-ID: <49cmsb$cac@sun0.urz.uni-heidelberg.de> Just to keep this warm: The Bug in append with POSIX has not been corrected in the 3.3 Patch1 libposix.a. What a pity. -- | Gregor Hoffleit admin MATHInet / contact HeidelNeXT | | MAIL: Mathematisches Institut PHONE: (49)6221 56-5771 | | INF 288, 69120 Heidelberg / Germany FAX: 56-3812 | | EMAIL: flight@mathi.uni-heidelberg.de (NeXTmail) |
From: matthias@arkon.amg.de.amg.de (Matthias Schuerhoff) Newsgroups: comp.sys.next.hardware,comp.sys.next.bugs Subject: Re: BIOS error with Symbios (NCR) driver?! Date: 27 Nov 1995 11:39:41 GMT Organization: AMG Industrieconsulting GmbH Message-ID: <49c81t$1pn@hagen.amg.de> References: <Pine.HPP.3.91.951124232053.4507H-100000@hphalle0.informatik.tu-muenchen.de> >fdisk fails: Bogus BIOS information. Boerny, I dont't think that there's something wrong with the BIOS. A friend had the same problem with "Bogus BIOS information" while trying to install Nextstep on two different Quantum hard disks (the system always recognized the 1 gig disk as a 30 MB disk:-) I think his problem was a wrong SCSI id for the hard disk or the CDROM. One of them must have the ID 0, I think. Hope that helps... Matthias ============================= Matthias Schuerhoff AMG Industrie Consulting GmbH Joseph-von-Fraunhoferstr. 27 Germany - 44227 Dortmund Phone: 231 97 53 54 0 EMail: matthias@amg.de =============================
From: eric@skatter.USask.Ca Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs Subject: MallocDebug.app broken in 3.3Patch? Date: 27 Nov 1995 16:28:35 GMT Organization: University of Saskatchewan Message-ID: <49covj$k8b@tribune.usask.ca> I installed the 3.3Patch on all the NEXTSTEP machines here a couple of weeks ago. This morning I tried using MallocDebug on one of my application (which uses the Foundation Kit). Every time I clicked on one of the MallocDebug display buttons (New/All) the MallocDebug application would quit and the MallocDebug window would disapper. To try and see what was going on, I ran: gdb /NextDeveloper/Apps/MallocDebug.app/MallocDebug Here's what I got: Reading symbols from /NextDeveloper/Apps/MallocDebug.app/MallocDebug...(no debugging symbols found)...done. Reading symbols from /usr/shlib/libNeXT_s.C.shlib...done. Reading symbols from /usr/shlib/libsys_s.B.shlib...done. (gdb) run Starting program: /NextDeveloper/Apps/MallocDebug.app/MallocDebug <<<<<<<<<<<<< At this point, I clicked on the `All' >>>>>>>>>>>>>>>>> <<<<<<<<<<<<< button in the MallocDebug window >>>>>>>>>>>>>>>>> Program generated(1): Memory access exception on address 0xc64034 (invalid address). 0x8677 in ?? () (gdb) where #0 0x8677 in ?? () #1 0x3762 in ?? () #2 0x3bf7 in ?? () #3 0x3978 in ?? () #4 0x3d9c in ?? () #5 0x5032fcf in -[Object perform:with:] () #6 0x60623ed in -[Application sendAction:to:from:] () #7 0x6066950 in -[Control sendAction:to:] () #8 0x60717d9 in -[Matrix sendAction:to:] () #9 0x606e814 in -[Cell _sendActionFrom:] () #10 0x606994f in -[Cell trackMouse:inRect:ofView:] () #11 0x606ff1c in -[ButtonCell trackMouse:inRect:ofView:] () #12 0x606b84e in -[Matrix _mouseDownNonListmode:] () #13 0x6071842 in -[Matrix mouseDown:] () #14 0x60636d5 in -[Window sendEvent:] () #15 0x6029ea1 in -[Application sendEvent:] () #16 0x603088f in -[Application run] () #17 0x43e9 in ?? () #18 0x2e97 in ?? () So, what's happening here? Does MallocDebug not work after the 3.3Patch is installed? Does MallocDebug not work with applications which use the foundation kit? -- Eric Norum eric@skatter.usask.ca Saskatchewan Accelerator Laboratory Phone: (306) 966-6308 University of Saskatchewan FAX: (306) 966-6058 Saskatoon, Canada. NeXTMail accepted.
From: eric@skatter.USask.Ca Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: DPSAddTimedEntry doesn't trigger autorelease Date: 27 Nov 1995 19:38:30 GMT Organization: University of Saskatchewan Message-ID: <49d43m$1i0@tribune.usask.ca> For the past day or so I've being trying to track down what I thought was a memory leak in one of my applications. The application displays, in real time, some data acquired via distributed objects from a server talking to an IEEE-488 digital oscilloscope. The objects returned from the server have NSObject as their ancestor, and use autorelease. The heart of the application is a method called from a DPSTimedEntryProc function. The method gets a couple of autoreleased waveforms from the digital oscilloscope server and displays them. The problem I was having was that the application would grow, and grow, and grow, then occasionally shrink again (as shown by the VSIZE column of ps alx). After some poking around, I found that the problem is that the DPS time entry events don't seem to trigger the autorelease cleanup -- the cleanup would occur only when I pressed some button, or selected some menu item in my application. Maybe this is a well-known problem, but it was new to me! The fix is simple -- just add an autorelease subpool to the timed-entry method: =================================================================== - tickHandler { NSAutoreleasePool *subpool = [[NSAutoreleasePool alloc] init]; if (xyMonitorViewX != nil) [xyMonitorViewX acquireDataFrom:scope channels:@"C1":@"C2"]; if (xyMonitorViewY != nil) [xyMonitorViewY acquireDataFrom:scope channels:@"C3":@"C4"]; [scope armTrigger]; if ((xyMonitorViewX != nil) && [[xyMonitorViewX window] isVisible]) [xyMonitorViewX display]; if ((xyMonitorViewY != nil) && [[xyMonitorViewY window] isVisible]) [xyMonitorViewY display]; [subpool release]; return self; } static void tickHandler (DPSTimedEntry tag, double now, char *userData) { [(ScopeController *)userData tickHandler]; } ===================================================================== -- Eric Norum eric@skatter.usask.ca Saskatchewan Accelerator Laboratory Phone: (306) 966-6308 University of Saskatchewan FAX: (306) 966-6058 Saskatoon, Canada. NeXTMail accepted.
From: cam@xedoc.com.au (Cameron Bromley) Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs Subject: Re: MallocDebug.app broken in 3.3Patch? Followup-To: comp.sys.next.programmer,comp.sys.next.bugs Date: 27 Nov 1995 23:50:42 GMT Organization: Access One Pty Ltd., Melbourne, Australia Message-ID: <49disi$ovv@news.mel.aone.net.au> References: <49covj$k8b@tribune.usask.ca> eric@skatter.USask.Ca wrote: : So, what's happening here? Does MallocDebug not work after the 3.3Patch : is installed? Does MallocDebug not work with applications which use the : foundation kit? MallocDebug works fine on my Intel 3.3 patched system with DBKit-based apps. I have not tested it with Foundation and/or EOF. -- ---------------------------------------------------------------- Cameron Bromley Email : cam@xedoc.com.au Xedoc Software Development Pty. Ltd. Fax : +61-3-9214-0102 Unit 11, 663 Victoria St, Phone : +61-3-9214-0199 Abbotsford, VIC, 3067, Australia http://www.xedoc.com.au/
From: bmw@leia.eric.on.ca (Bruce Walker) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Q: Workspace won't detect CDROM on secondary IDE ...? Date: 28 Nov 1995 09:02:39 -0500 Organization: Visible Genetics Inc. Message-ID: <49f4pv$l5@leia.eric.on.ca> This is a variation on the "IDE-only Intel system and CD-ROMs" theme. I'm using an HP Vectra (VL/3) with an HP/Sony CDU76E-S [1.0c], and believe it or not, I was able to get NeXTSTEP installed just fine from the CD-ROM. I connected it as the slave to the harddisk on the primary controller and used the new and improved install disk (NeXTAnswers #1921 & #1984) with the Drivers disk (NA #1921 & #1923), although I modified the Drivers disk by adding the Beta EIDE driver to it. This worked just fine, and I didn't even have to install a SCSI driver to fool the install process; it just worked. OK, so NeXTSTEP comes up, I can login and other than an occaisional bogus message about "SCSI disk is not initialized: format or eject?" when I login with a CD in the drive, all's well. But the doc's suggest installing the CD-ROM as the master device on the secondary IDE controller, so I did that. Now, when I login, the CD-ROM is effectively ignored. That is, the Workspace never "sees" a CD in the device, so no icon appears in the FileViewer. The CD-ROM device is seen by the kernel as it boots, I get lines in /usr/adm/messages to that effect. And I can manually mount the CD with % mount /mnt /dev/sd0a and then it's browseable and all. But automagic recognition by the Workspace is gone. Any ideas anyone? Thanks! -- Bruce M. Walker | Visible Genetics Inc. | bmw@visgen.com
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: My swapfile solution Date: 28 Nov 1995 22:07:27 GMT Organization: Information Technology Solutions, Inc. Message-ID: <49g16v$ba4@news.its.com> References: <DIn0q9.56u@news.hawaii.edu> <4998oj$6te@usc.edu> <49a5p0$28g@news.its.com> <Pine.NXT.3.91.951126133135.2006A-100000@capitalist.princeton.edu> "Timothy J. Luoma" <luomat@capitalist.princeton.edu> wrote: > On 26 Nov 1995, Chuck Swiger wrote: >> Sort of. Various system calls, such as fork(), vm_allocate(), and so >> forth will fail when you run out of memory. Eventually, vm_allocate() >> failing will cause malloc() to fail, which will cause the Workspace to >> be unhappy. But the system itself shouldn't fail completely, until the >> kernel needs to get more memory and cannot, at which time it will >> probably panic. > > I'm still confused then. I thought setting a HIWAT mark meant that the > system would be better off than if it fillled up (with the lovely > "pageout failed" message), but this makes it sounds like it is worse. Setting a hiwatermark on a swapfile so that it only grows to a certain point and stops does not mean the system has run out of swapspace, because you can swap to multiple swapfiles. What I've described above is the sequence of events that occur when you've run out of memory (ie, no swapfile can expand, either because you're out of drive space or because you've hit the hiwatermarks on all swapfiles). > I'm using a swapdisk, which means that when the swapfile hits the hiwat > on my swapdisk, it starts swapping on /private/vm/swapfile, so I've made > a little script that checks the time stamp on '/private/vm/swapfile' (I > use 'prefer' on the first swapfile). If it changes, I pop-up a warning > panel to tell me to reboot. And that setup should work just fine. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: pete@ohm.york.ac.uk (-bat.) Newsgroups: comp.sys.next.bugs Subject: Re: PPP2.2 on HP Date: 29 Nov 1995 11:55:17 GMT Organization: The University of York, UK Message-ID: <49hhn5$1pv@netty.york.ac.uk> References: <DIJHws.KGv@oops.se> kjell@oops.se(Kjell Nilsson) writes: > I tried to run ppp2.2 on my HP 715/100 using NextStep 3.3. It crached the > kernel. > (Can it be the serial port :?) > Is there anybody that knows if the latest 3.3 patch fixes this? Which version of PPP did you try ? Earlier ones had a problem with alignment of kalloc blocks which I fixed in the latest (0.4.5) version. -pete.
From: kelley@mudpot.ATMOS.ColoState.Edu Newsgroups: comp.sys.next.bugs Subject: SUMMARY: nameserving, incl mail, bombing after 3.3 patch Date: 29 Nov 1995 15:30:03 GMT Organization: Colorado State University, Fort Collins, CO 80523 Message-ID: <49hu9r$3k1k@yuma.ACNS.ColoState.EDU> i posted a couple weeks ago that i had successfully installed the 3.3 patch on 4 m68k machines and an intel machine. then did my hppa/ns server and nameserving no longer functioned for me. could not mail out from this machine (no MX or A record), could not get out to the net... basically could only go to places in my hosttable w/o using the ip addr. what i finally did, was reinstall the whole 3.3 patch. now the machine is working fine again. i got no errors on the first install so it's a mystery of thought to think about... kelley wittmeyer dept of atmospheric science colorado state university
Newsgroups: comp.sys.next.bugs From: gerald@kurt.in-berlin.de (Gerald Erdmann) Subject: Re: Are there NeXTSTEP Viruses? Message-ID: <1995Nov29.133443.1832@kurt.in-berlin.de> Sender: news@kurt.in-berlin.de References: <48rsfc$crc@news.internetmci.com> Date: Wed, 29 Nov 1995 13:34:43 GMT american@internetMCI.COM (Charles C. Hocker) wrote: > > Hello, > I am not trying create a scare. I am just wondering if there are any known NS > viruses, trojan horses, or worms that one should watch-out for. And if there are, is > there anyway to protect against them? > Hi! As far as I know there is NO VIRUS, NO TROJAN HORSES, NO WORMS. Thanks god ... Gerald --------------------------------------------------------------------- | GERALD ERDMANN | email: gerald @ kurt.in-berlin.de (NeXTmail welcome) | voice: +49 30 372 43 10 (Germany - Berlin) | crypt: pgp2 public key available |
From: giddings@menominee.menominee.chem.wisc.edu (Michael Giddings) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: No Sendmail Queueing (Mail.app bug?) Date: 29 Nov 1995 19:22:50 GMT Organization: University of Wisconsin, Madison Distribution: world Message-ID: <49ibua$1jbi@news.doit.wisc.edu> I recently installed the 3.3 patch. Before the patch, when mail was sent while the computer (a portable) was disconnected from the network, the mail would get deferred to the queue with a message such as "Host name lookup failure". Now, after the patch, the mail is immediately returned with the error message "Host unknown". So, thinking I would outsmart the system, I went into the Sendmail.cf file (which is linked to Sendmail.mailhost.cf, also, the Netinfo locations points to this file), and I set the delivery mode to queuing using the line "Odq" in place of "Odbackground". It seems this should force all mail to be queued until the queue is run (which happens when the PPP link is brought up). It works fine if I use the command line mail (/usr/ucb/mail) - it queues up messages nicely. However, Mail.app seems to do something that completely ignores this queue setting. Instead, it still attempts immediate delivery (without queueing), and immediately bounces the message back. The only two fixes I have found (neither of which I like) are: - Downgrade to using the unix mail program (arg!) - Start lookupd in /etc/rc with the -R option, bringing back the old lookup behavior, which though somtimes painful, at least works when the computer is not connected to the network. Any other suggestions? Is this a bug in Mail.app? -- ---------------------------------------------------------------------------- - Michael Giddings \ Tcl definitions that apply: UW Madison Dept. of Chemistry \ set job "Consultant and Graduate Student" Madison, Wisconsin \ set specialty "Scientific Computation" (608) 692-2851 \ set InRealLife "Whitewater kayaker and\ giddings@whitewater.chem.wisc.edu or\ outdoor enthusiast" giddings@students.wisc.edu \ ---------------------------------------------------------------------------- --
From: gfin@psych.ualberta.ca (Gary Finley) Newsgroups: comp.sys.next.bugs Subject: NS 3.3 got me Date: 30 Nov 1995 01:40:30 GMT Organization: University of Alberta, Edmonton, Canada Message-ID: <49j22e$9te@pulp.ucs.ualberta.ca> I just ran the NS3.3 Upgrader.app on my P90 system (which was running 3.2), and got an un-bootable machine for my trouble. The error message says: vfs_mountroot: error=6 panic: (Cpu0) vfs_mountroot: cannot mount root It's clearly an Intel driver issue (what else ever goes wrong with Intel NeXTs?), but I don't know what this mount failure is telling me about my configuration. If anyone can give me some hints, I'd be very grateful for some help. -- ---------------------------------------------- Gary Finley, Psychology Dept. Univ. of Alberta Network manager, Web manager, and postmaster. gfin@psych.ualberta.ca (NeXTmail welcome) http://web.psych.ualberta.ca/staff_bios/gary.finley.htmld/
From: paul@plsys.co.uk (Paul Lynch) Newsgroups: comp.sys.next.bugs Subject: Re: NS 3.3 got me Date: 30 Nov 1995 10:26:18 GMT Organization: P & L Systems, Ltd. Message-ID: <49k0sa$hti@ironhorse.plsys.co.uk> References: <49j22e$9te@pulp.ucs.ualberta.ca> In article <49j22e$9te@pulp.ucs.ualberta.ca> gfin@psych.ualberta.ca (Gary Finley) writes: > > I just ran the NS3.3 Upgrader.app on my P90 system (which was > running 3.2), and got an un-bootable machine for my trouble. > The error message says: > > vfs_mountroot: error=6 > panic: (Cpu0) vfs_mountroot: cannot mount root > > It's clearly an Intel driver issue (what else ever goes wrong > with Intel NeXTs?), but I don't know what this mount failure is > telling me about my configuration. If anyone can give me some > hints, I'd be very grateful for some help. You don't give enough information to diagnose this. One problem that will cause the same result is a card/Config mismatch. For example, if you upgrade a Canon objectStation with BusLogic SCSI adaptor, NeXT for some reason changed the default port address from x330 to x334 (or vice versa :-). In which case, you are best advised to chnage the jumpers on the card. Paul -- Paul Lynch (NeXTmail) paul@plsys.co.uk Tel: (01494)432422 P & L Systems Fax: (01494)432478 http://www.plsys.co.uk/~paul
From: beerli@genetics.washington.edu (Peter Beerli) Newsgroups: comp.sys.next.bugs Subject: Re: Sendmail Queueing - NS3.3/patch Date: 30 Nov 1995 07:08:19 GMT Organization: University of Washington Message-ID: <49jl93$loc@nntp5.u.washington.edu> References: <48da0k$1ekc@news.doit.wisc.edu> <30ABF199.41C6@genetics.washington.edu> <48i4c6$unc@msunews.cl.msu.edu> <30ACE370.41C6@genetics.washington.edu> <490t1q$g75@nntp5.u.washington.edu> <49fq21$ba4@news.its.com> [I tried some days ago, but my mail was refused?] Thanks to Matthew Reichman and Bill Bereza the sendmail queueing problem is solved now. The problem was: After I installed the 3.3 System patch my mail stopped working. My machine is not permanently on the net and I use slip to do so. Also my machine is no mailhost or nameserver. The solution: In the pre-patched 3.3 System with an older sendmail version it was possible to send mail away without knowing which machine is the local mailhost, with the new sendmail 8.7.* this was obviously not possible anymore. The catch is to add the local mailhost to the netinfo database. Add the mailhost and its name, aliases and IP address as an entry in the machines directory of your computer and add them also into the /etc/hosts file [(I followed Matthew and Bill with this, but I am not sure if this is necessary), I used nidump hosts /] Peter beerli@genetics.washington.edu
From: Garance A Drosehn <gad@eclipse.its.rpi.edu> Newsgroups: comp.sys.next.bugs Subject: Re: My swapfile solution Date: 30 Nov 1995 03:19:40 GMT Organization: Rensselaer Polytechnic Institute, Troy NY, USA Message-ID: <49j7sc$t5g@usenet.rpi.edu> References: <DIn0q9.56u@news.hawaii.edu> altenber@uhunix.uhcc.Hawaii.Edu (Lee Altenberg) wrote: > It seems to me that the obvious answer to the swapfile problems > is: allocate a fixed amount of space on the > disk to the swapfile, using an entry in /etc/swaptab such as: > > /private/vm/swapfile lowat=134217728,hiwat=134217728 > > I chose such a high value (134 MB) because Mathematica is a VM hog. > But anyway, this allocates a fixed chunk of the disk to the swapfile. > When you reboot, the swapfile stays at 134 MB. Just one minor note. Lowat doesn't allocate anything, it indicates what the size will be trimmed *back* to. Ie, a lowat=25000 will cause a 35000 byte swapfile to be trimmed down to 25000 bytes at reboot, but it will not cause a 15000 byte swapfile to be increased to 25000 bytes. So, the only change I'd suggest is that you do a mkfile for the size you want (134217728 in this case) right after setting up your disk and it's swaptab (or /etc/rc.swap) values. Check the man page for mkfile. --- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
From: Garance A Drosehn <gad@eclipse.its.rpi.edu> Newsgroups: comp.sys.next.bugs Subject: Re: Swapfiles under NeXTStep3.3 White Date: 30 Nov 1995 03:26:57 GMT Organization: Rensselaer Polytechnic Institute, Troy NY, USA Message-ID: <49j8a1$t5g@usenet.rpi.edu> References: <48u5n5$2e9@mars.earthlink.net> <498r4t$alq@usenet.rpi.edu> <499i7l$87e@news.tuwien.ac.at> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit wilkie@cg.tuwien.ac.at (Alexander Wilkie) wrote: > Garance A Drosehn <gad@eclipse.its.rpi.edu> wrote: > > 3) If you're running NeXTSTEP 3.3, apply the just-released > > patch for 3.3 > Does this have any benefits with respect to vm/swapfiles apart > from allegedly fixing the memory leak in Preferences.app? Actually, I'm not sure that the patch does fix the memory leak in preferences.app. I haven't checked. However it does claim to fix: Libraries --------- Reference: 54226 Problem: _openDefaults can leak 1024 bytes per invocation. Fix: This has been fixed. Window Server ------------- Reference: 54132 Problem: The WindowServer should use vm_allocate instead of malloc for backing store˜currently, it almost never gives VM back to the operating system. Fix: This has been fixed. among a number of other useful things. Check NeXTanswers 2608 and 2608 for complete details. --- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
From: moellney@michi.bota.uni-bonn.de (Michael Moellney) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Q: Workspace won't detect CDROM on secondary IDE ...? Date: 30 Nov 1995 21:53:05 GMT Organization: University of Bonn, Germany Message-ID: <49l942$nur@news.rhrz.uni-bonn.de> References: <49f4pv$l5@leia.eric.on.ca> bmw@leia.eric.on.ca (Bruce Walker) wrote: > This is a variation on the "IDE-only Intel system and CD-ROMs" > theme. > > I'm using an HP Vectra (VL/3) with an HP/Sony CDU76E-S [1.0c], and > believe it or not, I was able to get NeXTSTEP installed just fine > from the CD-ROM. I connected it as the slave to the harddisk on > the primary controller and used the new and improved install disk > (NeXTAnswers #1921 & #1984) with the Drivers disk (NA #1921 & > #1923), although I modified the Drivers disk by adding the Beta > EIDE driver to it. > How many EIDE Driver did you install in Configure. It seems, you to have to install one driver for each Controller. So in most systems Primary EIde is on IRQ 14 and secondary is on IRQ 15. So Configure on EIDE on IQR14 and a second on IRQ15. (Kinda configuring the new SerialDriver, one for each port) This worked for me on an Intel Plato... > This worked just fine, and I didn't even have to install a SCSI > driver to fool the install process; it just worked. OK, so NeXTSTEP > comes up, I can login and other than an occaisional bogus message > about "SCSI disk is not initialized: format or eject?" when I login > with a CD in the drive, all's well. > > But the doc's suggest installing the CD-ROM as the master device > on the secondary IDE controller, so I did that. Now, when I login, > the CD-ROM is effectively ignored. That is, the Workspace never > "sees" a CD in the device, so no icon appears in the FileViewer. > The CD-ROM device is seen by the kernel as it boots, I get lines in > /usr/adm/messages to that effect. And I can manually mount the CD > with > > % mount /mnt /dev/sd0a > > and then it's browseable and all. Mmmmmmm? Funny... > > But automagic recognition by the Workspace is gone. Any ideas anyone? > > Thanks! > Hope it helps, Michael > -- > Bruce M. Walker | Visible Genetics Inc. | bmw@visgen.com
Newsgroups: comp.sys.next.software,comp.sys.next.bugs From: Mark_Tarbell@Radical.Com (Mark Tarbell) Subject: Re: Anyone running DoomII on a NeXTcube? Message-ID: <1995Nov30.182923.488@Radical.Com> Sender: news@Radical.Com Organization: Radical System Solutions, Inc. References: <1995Nov20.180218.10576@Radical.Com> <magnan-2211951939200001@10.0.2.15> Date: Thu, 30 Nov 1995 18:29:23 GMT In <magnan-2211951939200001@10.0.2.15> Francois Magnan wrote: > It s not worth it. Even if you make it work the game is not playable on a > NeXT Cube. Much too slow. Sorry about forgetting to followup, but I did get it to work! Thanks to Tim Wood at OmniGroup who pointed out that there are 2 versions of the doom1.wad file, and they have incompatible internal differences. This means that .wad files which work with DoomI won't work with DoomII, even though they're named identically, and have the same size; I inadvertently mixed the versions, thus the error. And yes, DoomII is agonizingly slower than DoomI on a cube! Maybe they're in cahoots with the Pyro board people? :) At any rate, my 6 year old son doesn't mind! (Honest - he's the one who plays it, not me. No, really.) Mark
From: berezaw@river.it.gvsu.edu (Bill Bereza) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: Re: No Sendmail Queueing (Mail.app bug?) Date: 1 Dec 1995 01:22:28 GMT Organization: My Bedroom, Ravenna, MI Message-ID: <49llck$8r8@news.it.gvsu.edu> References: <49ibua$1jbi@news.doit.wisc.edu> In article <49ibua$1jbi@news.doit.wisc.edu>, Michael Giddings <giddings@whitewater.chem.wisc.edu> wrote: > >Any other suggestions? Is this a bug in Mail.app? Put the name & IP of your mail-relay host in the /machine directory of NetInfo (or put it in /etc/hosts and do niload hosts . < /etc/hosts). That fixed the problem for me. -- Bill Bereza berezaw@river.it.gvsu.edu <NeXT/MIME> Beware of all enterprises that require new clothes.
From: Mike Duggan <usd62656@interramp.com> Newsgroups: comp.sys.next.hardware,comp.sys.next.bugs Subject: Re: BIOS error with Symbios (NCR) driver?! Date: Thu, 30 Nov 1995 23:50:14 -0500 Organization: AIS Message-ID: <30BE8986.7C98@interramp.com> References: <Pine.HPP.3.91.951124232053.4507H-100000@hphalle0.informatik.tu-muenchen.de> <49c81t$1pn@hagen.amg.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit CC: mduggan@ADVIS.COM Send Start Unit Command in NCR BIOS must be YES, or set jumper on SCSI drive to send start unit command.
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.hardware,comp.sys.next.bugs Subject: Re: BIOS error with Symbios (NCR) driver?! Date: Fri, 1 Dec 1995 11:32:59 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951201112459.6914B-100000@hphalle7a.informatik.tu-muenchen.de> References: <Pine.HPP.3.91.951124232053.4507H-100000@hphalle0.informatik.tu-muenchen.de> <49c81t$1pn@hagen.amg.de> <30BE8986.7C98@interramp.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <30BE8986.7C98@interramp.com> On Thu, 30 Nov 1995, Mike Duggan wrote: > Send Start Unit Command in NCR BIOS must be YES, or set jumper on > SCSI drive to send start unit command. > ??? Am I incorrect, if I think that 'send start unit' just makes the drive turning (start the spindle motor?) If so, this isn't the solution because the drive is accessible and the the DOS parititions are even writeable. If not, what is this command doing additionally? Best regards, Boerny.
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: NFS crashes on high load Date: 1 Dec 1995 14:37:32 GMT Organization: Technical University of Berlin, Germany Message-ID: <49n3vc$nuq@news.cs.tu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi all, After fighting with server crashes on a regular basis for the last week, I'm getting desperate. I think I narrowed it down to the following, which in an easy way to bring every Intel System (I tried it on) down in a second.... (NSI 3.3 with and without patch) export some directory via NFS mount this directory to the local machine via NFS cd to this directory execute 'iozone 10 8192' The machine will run for a while but lock up after short time, especially when doing a disk access from another application or terminal window. While I admit that it is not very sensible to access a filesystem which is available locally via NFS, I made this test the first time to see the maximum throughput of NFS if the network is not the limiting factor. As I stated above, I am under the impression that high NFS load is what crashed our server 8 times during the last 48 hours. Would anyone dare to comment? Anyone having the same experience? I'm also interested if this test works on HPPA or SPARC. Axel -- Axel Habermann kiwi@cs.tu-berlin.de \\|// Muellerstr. 145 kiwi@buran.fb10.tu-berlin.de )o o( D-13353 Berlin (Wedding) \ | / Fon: +49 30 45478986 (privat) 030 314 24 764 (uni) \~/
Newsgroups: comp.sys.next.bugs From: markus g <tm@burrow.muc.de> Subject: Apps ignore preferences Message-ID: <7xn39cnwr2.fsf@burrow.muc.de> Sender: tm@burrow.muc.de Organization: hardly any... Date: Sat, 2 Dec 1995 00:18:57 GMT I recently encounter the problem that Apps ignore set (or even default) preferences. For example Mail.app 3.3 doesn't wrap lines after 75 characters; I have to confirm this anew every time I start Mail. The same seems to be the case with other Apps. Any suggestions what might be wrong are highly appreciated. Tia, markus g a/k/a the mole -- / . . . . . . . . . . . . . . . . . . . . . . . . . . __ __/ __ __ . . . . . . . . . . . . . . . . . . . . . the mole / / / / . . . . . . . . . . . . . . . . . tm@burrow.muc.de __/ __/ __/ __/ . . . . . . . . . . . . http://www.muc.de/~mgloede/
From: Borek Lupomesky Newsgroups: comp.sys.next.bugs Subject: Problems with sending mail Date: 1 Dec 1995 10:09:53 GMT Organization: Czech Technical University Distribution: world Message-ID: <49mk9h$7h0@ns.felk.cvut.cz> I am experiencing problems with sending mail under NS 3.3/Intel. I use dial-up PPP conection to our university. Wherever I send my mail I get this response: -=-=-=- Date: Fri, 1 Dec 95 11:05:54 +0100 From: NeXT Mail Agent <Mailer-Agent> Subject: Returned mail: Host unknown To: root ----- Transcript of session follows ----- 550 mailhost (ether)... 550 Host unknown 554 lupomesk@ujep.cz... 550 Host unknown (Authoritative answer from name server) ----- Unsent message follows ----- Return-Path: <root> Received: by next.ujep.cz (NX5.67e/NX3.0S) id AA00384; Fri, 1 Dec 95 11:05:55 +0100 Message-Id: <9512011005.AA00384@next.ujep.cz> Content-Type: text/plain Mime-Version: 1.0 (NeXT Mail 3.3 v118.2) Received: by NeXT.Mailer (1.118.2) From: Borek Lupomesky <root> Date: Fri, 1 Dec 95 11:05:54 +0100 To: lupomesk@ujep.cz Subject: Test Reply-To: lupomesk@sun.ujep.cz Test mail -=-=-=- I really don't know what's wrong. I didn't mess with sendmail config - it's default set. Bye Borek -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- BOREK LUPOMESKY, student University of J.E.Purkyne, Usti nad Labem, Czech Republic lupomesk@sun.ujep.cz, http://www.ujep.cz/~lupomesk/ \/\/\ PGP'ed mail welcome, key available via finger /\/\/ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Newsgroups: comp.sys.next.bugs Subject: Re: My swapfile solution Message-ID: <1995Dec2.034340.20700@indyvax.iupui.edu> From: rsanchez@indyunix.iupui.edu (Richard Sanchez) Date: 2 Dec 95 03:43:39 -0500 References: <DIn0q9.56u@news.hawaii.edu> <49j7sc$t5g@usenet.rpi.edu> I have been following this thread regarding the swap file in Nextstep. I too have had a couple crashes after exhausting all of my memory and swap disk space. After reading several of the comments in this thread, I think the best solution would be to either add more memory or get another hard disk just for swapping. My question is does anyone know how to create a swap file on a second disk. At this point, I have added the following items in my swaptab file as indicated in Next documentation "optimizing virtual memory with swaptab". # # /etc/swaptab # /private/vm/swapfile lowat=16777216,hiwat=58720256,prefer /disk2/swapfile lowat=1024 The problem is that I do not know how to correctly create the swapfile and swapfile.front files. I am not sure if I should use the mkfile command. Can someone please help me. Thanks, -- ---------------------------------------------------------------------- Richard Sanchez email: rsanchez@indyunix.iupui.edu Indiana University School of Medicine Indianapolis, Indiana ----------------------------------------------------------------------
From: trey@hsv.tybrin.com (Trey McClendon) Newsgroups: comp.sys.next.bugs Subject: Re: Apps ignore preferences Date: 2 Dec 1995 07:31:55 -0600 Organization: TYBRIN Corporation Message-ID: <49pkgb$om@tybrin1.hsv.tybrin.com> References: <7xn39cnwr2.fsf@burrow.muc.de> markus g (tm@burrow.muc.de) wrote: : I recently encounter the problem that Apps ignore set (or even : default) preferences. For example Mail.app 3.3 doesn't wrap lines : after 75 characters; I have to confirm this anew every time I start : Mail. The same seems to be the case with other Apps. I noticed this behavior immediately after installed the new EnhanceMail bundle. I haven't removed it yet to verify, but I'm pretty sure it didn't happen before it was installed. If you cycle through the Mail types (NextMail to MIME and back to ASCII) it will properly wrap to 75 characters, or to whatever your settings dictate. Trey -- Trey McClendon TYBRIN Corporation trey@hsv.tybrin.com Huntsville, AL NeXT / MIME Mail Accepted Fax: 205-837-3472
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: Re: Problems with sending mail Date: 2 Dec 1995 15:33:13 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-h-64.usc.edu Message-ID: <49prjp$sc5@usc.edu> References: <49mk9h$7h0@ns.felk.cvut.cz> This has been gone over a lot around here (FAQ) Borek Lupomesky wrote: > > I am experiencing problems with sending mail under NS 3.3/Intel. > I use dial-up PPP conection to our university. Wherever I > send my mail I get this response: [SNIP] > ----- Transcript of session follows ----- > 550 mailhost (ether)... 550 Host unknown > 554 lupomesk@ujep.cz... 550 Host unknown (Authoritative answer from name > server) Here we go: You are getting a "Host unknown" because your mail is going out with the wrong domain - it's going out as "lupomesk@ujep.cz" it should be "lupomesk@sun.ujep.cz" (1) What version sendmail are you using? There's a sendmail.8.7.2.pkg at the ftp sites (probably ftp.cs.orst.edu) that you should upgrade to. (2) Use the "MASQUERADE" feature in sendmail.cf to give yourself the right domainname. (3) Make sure to Fully Qualify your domain names (the new sendmail works well with NetInfo and /etc/hostconfig), e.g. using NetInfo, along with your /etc/hostconfig file. You should enter (a) your mailserver (and its aliases) as a new machine with their ip addresses, (b) you should do this for your own machine, and (c) you should do this for your internet domain. Once NetInfo and /etc/hostconfig are updated, and I think you might want to reboot, you should have no more of those problems queuing messages. -- Be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV For PGP key, send email with subject "request_PGP"
From: stevem@RADium.ca (Stephen MacDougall) Newsgroups: comp.sys.next.bugs Subject: EOF Header bug Date: 2 Dec 1995 20:03:34 GMT Organization: ftn Internet Message-ID: <49qbem$52k@master.ftn.net> I just discovered a bug in the following header file in EOF: eoaccess/EOAdaptorChannel.h This file makes reference to NSCalendarDate, which is defined in NSDate.h. You must add the line: #import <foundation/NSDate.h> in EOAdaptorChannel.h. This will fix the problem. I guess someone forgot an import. I'm suprised that NeXT didn't catch this one. -- ==================================================== Stephen MacDougall RADium Technology Centre (Canada) stevem@RADium.ca NeXT and MIME mail gladly accepted ====================================================
From: felix.rauch@limmat.ch (Felix Rauch) Newsgroups: comp.sys.next.bugs Subject: Re: My swapfile solution Date: 28 Nov 1995 22:25:22 GMT Organization: Private NEXTSTEP-site Distribution: world Message-ID: <49g28i$89@harka.limmat.net.ch> References: <DIn0q9.56u@news.hawaii.edu> Lee Altenberg (altenber@uhunix.uhcc.Hawaii.Edu) wrote: > I would think this would prevent disk fragmentation caused by repeatedly > growing and shrinking the swapfile. If you just set the lowat to a high level, the swapfile will also slowly grow until it reaches the mark. I think it's better if you first create your swapfile with `mkfile' (You shouldn't do that while the system is using the swapfile! Rebooting in single-user mode and doing it from there would probably be safe). - Felix -- Felix Rauch, CS-Student @ ETH Zurich, Switzerland. E-mail: felix@nice.ch NeXT/MIME-mail welcome. Member of NiCE - NeXT User Group Homepage: http://nice.ethz.ch/~felix (includes pgp public key)
From: mow@navigator.de (Markus Wenzel) Newsgroups: comp.sys.next.bugs Subject: Re: My swapfile solution Date: 1 Dec 1995 22:24:16 +0100 Organization: Navigator Message-ID: <49nrq0$4cl@marsu.pilhuhn.de> References: <DIn0q9.56u@news.hawaii.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Lee Altenberg (altenber@uhunix.uhcc.Hawaii.Edu) wrote: > It seems to me that the obvious answer to the swapfile problems > is: allocate a fixed amount of space on the > disk to the swapfile, using an entry in /etc/swaptab such as: > > /private/vm/swapfile lowat=134217728,hiwat=134217728 > > I chose such a high value (134 MB) because Mathematica is a VM hog. > But anyway, this allocates a fixed chunk of the disk to the swapfile. > When you reboot, the swapfile stays at 134 MB. As I understand it, > the memory within this space just keeps getting automatically recycled. > I would think this would prevent disk fragmentation caused by repeatedly > growing and shrinking the swapfile. Also, you know that the swapfile > can't grow to fill up your disk, once it reaches hiwat. > Now, am I missing something about swapfile management that would > make this a bad idea? 1. 134 MB is rather small swapfile size for me. Try Tailor and TIFFany... 2. You're wasting disk space if your swap tasks need less than 134 MB. On the other hand, you are not tempted to fill up your harddisk totally :-) 3. My solution is 64 MB of RAM and a 1 GB swapdisk. Works great for me! ;_)) -- who? // Dipl.-Phys. Markus Wenzel work? // Navigator - IT Consulting & System administration mail? // mow@navigator.de more? // http://www.pilhuhn.de/~marsu/
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 4 Dec 1995 05:15:10 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <49u04u$i1l@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Help: Strange PPP behaviour: ping not possible, telnet ok?! Date: Mon, 4 Dec 1995 17:49:14 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951204174123.3625A-100000@hphalle7a.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Without a noticable change to the system, I encountered a serious problem. I'm using ppp2.2_0.4.x and everything works fine: the deamons exchange their IPs and my pppd states: ipcp up. But then.... Then I try a ping to the remote IP which I just received during the connection phase: result is just a hanging pinng (NO network unreachable, but no reply). Tracerouts to the same IP result in stars (*). Telnet to the remote IP works ! But the connection is going to drop in about 1 minute or so. Isn't this strange? I did a complete new installation of NS3.3 on a spare drive and installed only ppp with the same cause. I have to state that everything went fine for weeks with the same configuration so: Might there be hardware problem with my serial port? I already tried different baud rates with no luck. Any tips would be very much appreciated, Boerny (going to change the ppp provider). -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/ scholz@ve1.rm.op.dlr.de http://www.leo.org/~scholz/
From: info@object-tech.com Newsgroups: comp.sys.next.bugs Subject: RE: TEST ONLY Date: 4 Dec 1995 16:55:27 GMT Organization: SuperNet Inc. (303)-296-8202 Denver Colorado Message-ID: <49v95v$b6u@news-2.csn.net> TEST POST
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: Creating a swapfile on 2nd drive (was Re: My swapfile solution) Date: Mon, 4 Dec 1995 11:55:24 -0500 Organization: Princeton University Message-ID: <Pine.NXT.3.91.951204114326.1161A-100000@capitalist.princeton.edu> References: <DIn0q9.56u@news.hawaii.edu> <49j7sc$t5g@usenet.rpi.edu> <1995Dec2.034340.20700@indyvax.iupui.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <1995Dec2.034340.20700@indyvax.iupui.edu> On 2 Dec 1995, Richard Sanchez wrote: > I have been following this thread regarding the swap file in Nextstep. I too > have had a couple crashes after exhausting all of my memory and swap disk > space. After reading several of the comments in this thread, I think the > best solution would be to either add more memory or get another hard disk > just for swapping. My question is does anyone know how to create a swap file > on a second disk. Of course, if you can do both (swapdisk and more RAM) that's better. For your question, I need more info. Are you talking about dedicating a second drive to swapping? If so, then you do 1 of 2 things. 1) change the disk label to swapdisk (using either the disk -L or disk -l command, see man page for 'disk' I can't remember which) or 2) setup a fstab entry for the disk to use it for swapping. Since I don't really know how to do #2, I can only really suggest #1, as long as you realize that you will have to do more than just set the disk label. Also realize that the /tmp directory will be loaded onto the swapdisk rather than onto your primary drive. Whether or not this is is correct is left up to you. If you don't like it, change the references to /private/swapdisk/tmp to /private/tmp (well, I did that and it works OK for me, use at your own risk). You will have to go into rc.swap (I believe that is the right file) and edit the lowat and hiwat for the swapdisk to suit your disk size. Remember that the HIWAT is based on the swapfile.front size, which is not the same as your swapfile size (more on this below). This is probably considered a bug, because it really should be based on the size of the actual swapfile. > > At this point, I have added the following items in my swaptab file as > indicated in Next documentation "optimizing virtual memory with swaptab". > > # > # /etc/swaptab > # > /private/vm/swapfile lowat=16777216,hiwat=58720256,prefer > /disk2/swapfile lowat=1024 > > The problem is that I do not know how to correctly create the swapfile and > swapfile.front files. I am not sure if I should use the mkfile command. Can > someone please help me. DO NOT create the swapfile.front file with mkfile!!!!!!!!!!!! If you use compression of the swapfile, then swapfile.front will be created automatically on its own so you don't have to. The best way to make the swapfile (at the LOWAT mark) is to use the mkfile command from single user mode. > > Thanks, > Richard Sanchez > email: rsanchez@indyunix.iupui.edu Someone with a big heart really should put together a FAQ on this subject... it could be done almost as simply as going through posts from the last few months. -- Timothy J. Luoma luomat@capitalist.princeton.edu (MIME fine! NeXTMail if necessary) mail sent to: 476tjl@ptsmail.ptsem.edu is forwarded to above address. "If I understood the man page, I wouldn't have asked the question."
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: Re: Creating a swapfile on 2nd drive (was Re: My swapfile solution) Date: Mon, 4 Dec 1995 11:59:53 -0500 Organization: Princeton University Message-ID: <Pine.NXT.3.91.951204115802.1328A-100000@capitalist.princeton.edu> References: <DIn0q9.56u@news.hawaii.edu> <49j7sc$t5g@usenet.rpi.edu> <1995Dec2.034340.20700@indyvax.iupui.edu> <Pine.NXT.3.91.951204114326.1161A-100000@capitalist.princeton.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <Pine.NXT.3.91.951204114326.1161A-100000@capitalist.princeton.edu> One other very important thing I forgot to mention is that on my NeXT, which uses the swapdisk method, reaches the HIWAT, the swapping goes back to /private/vm/swapfile. This might also be considered a "bug" but I like it because I have written a small script that tells me when the 2nd swapfile has begun to grow, and I use that opportunity to reboot. -- Timothy J. Luoma luomat@capitalist.princeton.edu (MIME fine! NeXTMail if necessary) mail sent to: 476tjl@ptsmail.ptsem.edu is forwarded to above address. "If I understood the man page, I wouldn't have asked the question."
From: robert@steffi.accessone.com (Robert Nicholson) Newsgroups: comp.sys.next.bugs Subject: Re: EOF Header bug Date: 04 Dec 1995 14:04:37 -0800 Organization: x Sender: robert@steffi.accessone.com Message-ID: <e91kszdsa.fsf@steffi.accessone.com> References: <49qbem$52k@master.ftn.net> To: stevem@RADium.ca (Stephen MacDougall) CC: eof@omnigroup.com Q: Why does it have all those imports in the header file? It's uses @class sparingly. Doesn't this add the foundation headers in the precompiled header file? Which eoaccess.h seems to do already deliberately although it misses NSCalendarDate Anybody know the format of a precompiled header file? -- "Mary ate a little lamb and punk rock isn't dead" (PGP key: send email with Subject: request pgp key)
From: reichman@scf.usc.edu Newsgroups: comp.sys.next.bugs Subject: Re: Creating a swapfile on 2nd drive (was Re: My swapfile solution) Date: 4 Dec 1995 22:09:43 GMT Organization: University of Southern California, Los Angeles, CA Sender: reichman@comserv-h-59.usc.edu Message-ID: <49vrj7$5o@usc.edu> References: <DIn0q9.56u@news.hawaii.edu> <49j7sc$t5g@usenet.rpi.edu> <1995Dec2.034340.20700@indyvax.iupui.edu> <Pine.NXT.3.91.951204114326.1161A-100000@capitalist.princeton.edu> "Timothy J. Luoma" <luomat@capitalist.princeton.edu> wrote: > You will have to go into rc.swap (I believe that is the right file) and > edit the lowat and hiwat for the swapdisk to suit your disk size. Actually, I spoke with a NeXT support person and they advised me to keep the LOWAT definition but take out the HIWAT completely as the system will use whatever is available on your swapdisk anyway. By the way, thanks tons TJL for the instructions. They worked quite well, though I was able to make other mistakes that made me sweat a bit. oh well. But now everything is fine. -- Be well, Matthew ================================================================== Matthew Reichman | NeXTStep v.3.3 m68k reichman@scf.usc.edu | NeXTMAIL & MIME welcome USC-CNTV For PGP key, send email with subject "request_PGP"
From: perkins@cps.msu.edu (Stephen J. Perkins) Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Re: Help: Strange PPP behaviour: ping not possible, telnet ok?! Date: 5 Dec 1995 00:10:12 GMT Organization: Michigan State University Distribution: world Message-ID: <4a02l4$lg9@msunews.cl.msu.edu> References: <Pine.HPP.3.91.951204174123.3625A-100000@hphalle7a.informatik.tu-muenchen.de> In <Pine.HPP.3.91.951204174123.3625A-100000@hphalle7a.informatik.tu-muenchen.de > Bernhard Scholz wrote: > > Without a noticable change to the system, I encountered a serious > problem. I'm using ppp2.2_0.4.x and everything works fine: the deamons > exchange their IPs and my pppd states: ipcp up. But then.... > > Then I try a ping to the remote IP which I just received during the > connection phase: result is just a hanging pinng (NO network unreachable, > but no reply). Tracerouts to the same IP result in stars (*). Telnet to > the remote IP works ! But the connection is going to drop in about 1 > minute or so. This is not necessarily a problem. Can you ping some other host (other than the PPP peer) by IP address? There are some PPP servers that are not built on top of a full TCP/IP stack. These servers wouldn't understand or reply to ICMP messages directed at them. They can however, correctly forward IP packets to a LAN. As to the telnet, they could have a proxy... I don't know. Do other hosts work OK? - Steve -- ============================================================== Stephen J. Perkins | mailto:perkins@cps.msu.edu Dept. of Comp. Science | NeXT, MIME, finger for PGP Michigan State University | NeXT OS 3.3 using PPP-2.3 NeXT PPP-2.2 info at http://www.thoughtport.com:8080/PPP/
From: chuck@its.com (Chuck Swiger) Newsgroups: comp.sys.next.bugs Subject: Re: EOF Header bug Date: 5 Dec 1995 04:51:55 GMT Organization: Information Technology Solutions, Inc. Message-ID: <4a0j5b$n5h@news.its.com> References: <49qbem$52k@master.ftn.net> <e91kszdsa.fsf@steffi.accessone.com> robert@steffi.accessone.com (Robert Nicholson) wrote: > Q: Why does it have all those imports in the header file? > > It's uses @class sparingly. Either someone was lazy setting up the headers, or they needed to. :-) > Doesn't this add the foundation headers in the precompiled header > file? Which eoaccess.h seems to do already deliberately although it > misses NSCalendarDate Yes. It's almost certainly cheaper to read the precompiled headers and intergrate them than it would be to actually wander through the filesystem reading in all of the .h files anyway. > Anybody know the format of a precompiled header file? Can't you determine that by examining the source NeXT provides for the -precomp option to the compiler....? The precompiled format is likely to be fairly close to the internal representation that the compiler would generate itself by reading in all of the .h files. -Chuck -- Charles Swiger -- chuck@its.com | Information Technology Solutions, Inc. --------------------------------+--------------------------------------- CrashCatcher Development, Systems and Networking Administrator
From: kay@kauri.vuw.ac.nz (Cameron Kay) Newsgroups: comp.sys.next.bugs Subject: Trouble booting off a floppy or CD-ROM Date: 05 Dec 1995 09:16:54 GMT Organization: Victoria Uni. of Wellington, NZ. Distribution: world Message-ID: <KAY.95Dec5221654@kauri.vuw.ac.nz> I've got a NeXTstation Colour Turbo that I wanted to install a new system on from scratch. Looking at the Installation Manual it say to hold down the Command and ~ keys while the system does its self test and then to type bfd at the NeXT> prompt after inserting the boot floppy and NEXTSTEP 3.3 CD-ROM. Unfortunately when I try this it always comes up the an Exception #3 error and fails to boot. Anyone know how to fix this? I'm also tried to boot of the CD-ROM directly using the bsd(n,0,0)sdmach rootdev=sdna when n is the SCSI device number of the CD-ROM. This also doesn't work and I get another Exception #3. Lastly I tried booting off the CD-ROM by setting the CD-ROM drive to SCSI ID 0. This attempts to boot the CD-ROM as a normal drive and fails part-way through the boot process. Can anyone help me out? - Cameron -- Email Cameron.Kay@kauri.vuw.ac.nz Post Computer Science Department Phone + 64 4 472 1000 x7032 (Work) Victoria University + 64 4 237 5895 (Home) P.O.Box 600 Fax + 64 4 495 5232 Wellington, New Zealand
From: flight@mathi.uni-heidelberg.de (Gregor Hoffleit) Newsgroups: comp.sys.next.bugs Subject: Floppy error msg only to /usr/adm/messages Date: 5 Dec 1995 11:23:29 GMT Organization: University of Heidelberg, Germany Message-ID: <4a1a3h$pd6@sun0.urz.uni-heidelberg.de> Error messages regarding floppy disk drives (only tested with DOS fs) always go to /usr/adm/messages, and don't use syslog. This is annoying since there's no notification about these errors in the UI, and - due to the lack of syslog support - it's hard to catch these messages with a custom tool and notify the user. This is how these errors look like. In fact, these files were corrupted during write, but due to the lack of notification, I didn't find out until reading back: Excerpt from /usr/adm/messages: Dec 4 22:54:41 thefly mach: fd0: Sector 72 cmd = Read; Missing Address Mark: RETRYING Dec 4 22:54:42 thefly mach: fd0: Sector 72 cmd = Read; Missing Address Mark: RETRYING Dec 4 22:54:42 thefly mach: fd0: Sector 72 cmd = Read; Missing Address Mark: RECALIBRATING Dec 4 22:54:42 thefly mach: fd0: Sector 72 cmd = Read; Missing Address Mark: RETRYING Dec 4 22:54:42 thefly mach: fd0: Sector 72 cmd = Read; Missing Address Mark: ... Dec 4 22:59:50 thefly mach: fd0: Sector 2801 cmd = Read; Media error - data CRC: RETRYING Dec 4 22:59:50 thefly mach: fd0: Sector 2801 cmd = Read; Media error - data CRC: RETRYING Dec 4 22:59:50 thefly mach: fd0: Sector 2801 cmd = Read; Media error - data CRC: RECALIBRATING Dec 4 22:59:51 thefly mach: fd0: Sector 2801 cmd = Read; Media error - data CRC: RETRYING -- | Gregor Hoffleit admin MATHInet / contact HeidelNeXT | | MAIL: Mathematisches Institut PHONE: (49)6221 56-5771 | | INF 288, 69120 Heidelberg / Germany FAX: 56-3812 | | EMAIL: flight@mathi.uni-heidelberg.de (NeXTmail) |
From: mike@hobbs.chem.usu.edu (Michael Emmel) Newsgroups: comp.sys.next.bugs Subject: NS3.3 sleeping Date: 5 Dec 1995 15:35:02 GMT Organization: SouthWind Internet Access, Inc. Distribution: world Message-ID: <4a1or6$i57@opal.southwind.net> Hello I have a non networked home intel machine and I tend to stay logged on and turn the screen on sometimes it won't come back from down mode were the display is dimmned. No amount of playing with the keyboard or mouse will reactivate the window server ? Then I must reboot. I suspect that rlogins would probably work but I have know way to test it. It seems to happen when the machines sets for several days without me logging out. Anybody else seen this ? mike@hobbs.chem.usu.edu Mike
Newsgroups: comp.sys.next.bugs From: bobcook@slac.stanford.edu (Bob Cook) Subject: Re: NS3.3 sleeping Message-ID: <DJ574G.qz@unixhub.SLAC.Stanford.EDU> Sender: news@unixhub.SLAC.Stanford.EDU Organization: Stanford Linear Accelerator Center References: <4a1or6$i57@opal.southwind.net> Date: Wed, 6 Dec 1995 02:05:03 GMT Michael Emmel writes > >Hello I have a non networked home intel machine and I tend to stay logged >on and turn the screen on sometimes it won't come back from down mode were >the display is dimmned. No amount of playing with the keyboard or mouse >will reactivate the window server ? Then I must reboot. I have seen behavior like this (on a black NeXT, not Intel) when the machine has in fact panic'd and crashed. In my case the screen was dimmed a great deal but not black, so I could dimly see the panic window. -- Bob Cook (415) 926-2769 bobcook@slac.stanford.edu Stanford Linear Accelerator Center NeXT mail okay PGP public key: ftp://ftp.slac.stanford.edu/users/bobcook/pgp.publickey
From: majordomo@singnet.com.sg Newsgroups: comp.sys.next.bugs Subject: * Free Newsletter on Passion & Romance * Date: 5 Dec 1995 22:31:49 GMT Organization: PassionFRUIT Distribution: world Message-ID: <4a2h8l$f34@lantana.singnet.com.sg> =================================================================== Introducing * PassionFRUIT * ... ... =================================================================== First Online Newsletter Dedicated to Passion & Romance ... ... Distributed Fortnightly FREE includes:- - how to meet your dream partner - how to get that all important first date - What to do on dates -- how to make your dates sizzle ... - how to keep your dates coming - how to build a deeper & more intimate relationship - how to make YOU an Attractive person & lots of other love tips 1 SIZZLING love tip per fortnightly issue - delivered to Your Mailbox FREE on 1st & 15th monthly! To subscribe, simply e-mail to majordomo@singnet.com.sg -- In the body/content, type SUBSCRIBE PASSIONFRUIT. It's as simple as that! * Serving the Network Community Since 1980 * -------------------------------------------------------------------
From: majordomo@singnet.com.sg Newsgroups: comp.sys.next.bugs Subject: cancel: * Free Newsletter on Passion & Romance * Date: 5 Dec 1995 22:31:49 GMT Organization: PassionFRUIT Distribution: world Message-ID: <cancel.4a2h8l$f34@lantana.singnet.com.sg> Control: cancel <4a2h8l$f34@lantana.singnet.com.sg> Spam killing
From: Bernhard Scholz <scholz@informatik.tu-muenchen.de> Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Re: Help: Strange PPP behaviour: ping not possible, telnet ok?! Date: Wed, 6 Dec 1995 13:38:03 +0100 Organization: Technische Universitaet Muenchen, Germany Distribution: world Message-ID: <Pine.HPP.3.91.951206133054.16447B-100000@hphalle4h.informatik.tu-muenchen.de> References: <Pine.HPP.3.91.951204174123.3625A-100000@hphalle7a.informatik.tu-muenchen.de> <4a02l4$lg9@msunews.cl.msu.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <4a02l4$lg9@msunews.cl.msu.edu> On 5 Dec 1995, Stephen J. Perkins wrote: > > Then I try a ping to the remote IP which I just received during the > > connection phase: result is just a hanging pinng (NO network unreachable, > > but no reply). Tracerouts to the same IP result in stars (*). Telnet to > > the remote IP works ! But the connection is going to drop in about 1 > > minute or so. > > > This is not necessarily a problem. Can you ping some other host (other than > the PPP peer) by IP address? There are some PPP servers that are not built > on top of a full TCP/IP stack. These servers wouldn't understand or reply > to ICMP messages directed at them. They can however, correctly forward IP > packets to a LAN. As to the telnet, they could have a proxy... I don't > know. Do other hosts work OK? > Thanks Steve for commenting on this. I tried extra hard to figure out some additional information: Our PPP provider changed the OS the PPP software is running on, since than I don'te get any valid ping (although I can telnet as I stated). There are >100 other people on e.g. Linux boxes who don't have any problems (and I believe they use the same package, don't they?). I didn't try your lates release x.6, yet, but I tried a different provider and everything is ok with him, so I'm really wondering what's going on here! While I switched pppd to send alive messages I encountered a loss of rcv messages and then pppd brings down the connection, so I'm pretty sure the remote side is doing something wrong, but I don't know what or why, and if so, why there are no other problems with the others connected to the site. I already got a new account from the same provider but there was no change in the bahaviour. Many greetings, and a happy Santa Clause day, Boerny. -- Bernhard Scholz (IRC: Boerny) scholz@informatik.tu-muenchen.de http://peanuts.leo.org/ scholz@ve1.rm.op.dlr.de http://www.leo.org/~scholz/
From: Borek Lupomesky Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Re: Help: Strange PPP behaviour: ping not possible, telnet ok?! Date: 6 Dec 1995 08:38:10 GMT Organization: Czech Technical University Distribution: world Message-ID: <4a3kpi$k9h@ns.felk.cvut.cz> References: <Pine.HPP.3.91.951204174123.3625A-100000@hphalle7a.informatik.tu-muenchen.de> <4a02l4$lg9@msunews.cl.msu.edu> In <4a02l4$lg9@msunews.cl.msu.edu> Stephen J. Perkins wrote: > > Then I try a ping to the remote IP which I just received during the > > connection phase: result is just a hanging pinng (NO network unreachable, > > but no reply). Tracerouts to the same IP result in stars (*). Telnet to > > the remote IP works ! But the connection is going to drop in about 1 > > minute or so. > > This is not necessarily a problem. Can you ping some other host (other than > the PPP peer) by IP address? There are some PPP servers that are not built > on top of a full TCP/IP stack. These servers wouldn't understand or reply > to ICMP messages directed at them. They can however, correctly forward IP > packets to a LAN. As to the telnet, they could have a proxy... I don't > know. Do other hosts work OK? I have same problem. My peer is Cisco CS508 terminal server, so that it certainly should reply ICMP echos. I think there is something wrong with routing that PPP automatically sets. But, I didn't explore this too much, because everything else works fine. Bye Borek -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- BOREK LUPOMESKY, student University of J.E.Purkyne, Usti nad Labem, Czech Republic lupomesk@sun.ujep.cz, http://www.ujep.cz/~lupomesk/ PGP and NeXT mail welcome, key available via finger =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.misc From: Fabien_Roy@free.fdn.org Subject: EOF: multiple declarations for method `endEditing' Message-ID: <1995Dec7.005259.3295@free.fdn.org> Sender: news@free.fdn.org Organization: Fabien Roy Consultant, Paris, France Date: Thu, 7 Dec 1995 00:52:59 GMT How to get rid of this warning? This message arise when I started to use the method -endEditing to force an update on an NXTableview. BTW the program seems to work. Prog.m:203: warning: multiple declarations for method `endEditing' /NextDeveloper/Headers/eointerface/EOController.h: In function `endEditing': /NextDeveloper/Headers/eointerface/EOController.h:197: warning: using `-(void)endEditing' /NextDeveloper/Headers/eointerface/NXEditableFormatter.h: In function `endEditing': /NextDeveloper/Headers/eointerface/NXEditableFormatter.h:69: warning: also found `-(BOOL)endEditing' /NextDeveloper/Headers/eointerface/NXTableView.h: In function `endEditing': /NextDeveloper/Headers/eointerface/NXTableView.h:166: warning: also found `-endEditing' Fabien -- --------------------------------------------------------------------- Fabien_Roy@free.fdn.org (NextMail/MIME accepted) Fabien Roy Consultant NEXTSTEP/OPENSTEP/EOF Consultant, SYBASE DBA 10 rue de la DEFENSE 93100 MONTREUIL, France Tel: 33 1 45 28 32 23 Fax: 33 1 48 55 09 90
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: Re: Floppy error msg only to /usr/adm/messages Date: Wed, 6 Dec 1995 16:44:16 -0500 Organization: Princeton University Message-ID: <Pine.NXT.3.91.951206164043.2127C-100000@capitalist.princeton.edu> References: <4a1a3h$pd6@sun0.urz.uni-heidelberg.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <4a1a3h$pd6@sun0.urz.uni-heidelberg.de> This isn't really a "solution" but you might want to grab "Spy.app" from ftp://peanuts.leo.org/pub/comp/platforms/next/Tools/workspace/Spy.1.3.NIHS.b.tar.gz It is basically a console watcher, but there is a dwrite (dwrite Spy File /path/to/file) which lets you keep an eye on any file, not just console.log. I use it to watch my .procmail.log file. A real solution would be to find out where it is setup and change it, but I don't know where that is. If you find out, please let us know. TjL -- Timothy J. Luoma luomat@capitalist.princeton.edu (MIME fine! NeXTMail if necessary) mail sent to: 476tjl@ptsmail.ptsem.edu is forwarded to above address. "If I understood the man page, I wouldn't have asked the question."
From: Ron Wood <ron> Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: Re: Help: Strange PPP behaviour: ping not possible, telnet ok?! Date: 6 Dec 1995 18:15:32 GMT Organization: CE-CERT Message-ID: <4a4mk4$qdk@galaxy.ucr.edu> References: <Pine.HPP.3.91.951204174123.3625A-100000@hphalle7a.informatik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Check in NetInfo protocols to see if UDP packets are enabled. -- Ron Wood, Systems Administrator CE-CERT, UCR
From: johnz@shellx.best.com (John Zollinger) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.misc Subject: Re: EOF: multiple declarations for method `endEditing' Followup-To: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.misc Date: 6 Dec 1995 19:15:02 -0800 Organization: Best Internet Communications Message-ID: <4a5m7m$7jn@shellx.best.com> References: <1995Dec7.005259.3295@free.fdn.org> Fabien_Roy@free.fdn.org wrote: : How to get rid of this warning? : This message arise when I started to use the method -endEditing to force an : update on an NXTableview. : BTW the program seems to work. : Prog.m:203: warning: multiple declarations for method `endEditing' : /NextDeveloper/Headers/eointerface/EOController.h: In function `endEditing': : /NextDeveloper/Headers/eointerface/EOController.h:197: warning: using : `-(void)endEditing' : /NextDeveloper/Headers/eointerface/NXEditableFormatter.h: In function : `endEditing': : /NextDeveloper/Headers/eointerface/NXEditableFormatter.h:69: warning: also : found `-(BOOL)endEditing' : /NextDeveloper/Headers/eointerface/NXTableView.h: In function `endEditing': : /NextDeveloper/Headers/eointerface/NXTableView.h:166: warning: also found : `-endEditing' Because the object you are sending the 'endEditing' method to is of type 'id' (so you can hook it up in IB), and the compiler finds multiple objects that have that method, it just picks one that it thinks you mean. It is easy to fix. Just type cast the object when you use it. Like so: [(NXTableView *)myTableView endEditing]; Then it will know what object you mean, and the warning will go away. This is a good way to do things anyway, so that the compiler can tell you if you screw something up. But, then I'm the type that makes sure my compiles have no warnings so that when one shows up, it catches my attention. Good luck, John Zollinger ProActive Consulting, Inc. proactiv@sunrem.com
From: katzlbt@vuse.vanderbilt.edu (Thomas Katzlberger) Newsgroups: comp.sys.next.bugs Subject: NeXT->DOS filesystem copy bug ? Date: 7 Dec 1995 05:50:44 GMT Organization: a black NeXT Message-ID: <4a5vbk$gjo@news.vanderbilt.edu> I was copying file from my NeXT to my DOS partition as I saw that files with size of power of two end up with 0 bytes on the DOS partition. They show up right in the WS browser and are also readable. Once I log out and start DOS I get - makeing a dir - filesize 0 of those files. Logging in again in NS I get also size 0, deleting them will free the original size of the file. Adding a single byte to those files makes the copy work again. -rw-r--r-- 1 cat 1048576 Oct 20 01:00 disk01.hm -rw-r--r-- 1 cat 2138112 Aug 14 16:08 sounds Those are the files producing this behavior reliably. Has anyone else seen that or can verify this bug ? Cat. -- _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ _/_/ _/_/ _/_/ Thomas Katzlberger _/_/ _/_/ katzlbt@vuse.vanderbilt.edu _/_/ _/_/ @aBlackNeXT.called.garfield _/_/ _/_/ http://www.vuse.vanderbilt.edu/~katzlbt/ _/_/ _/_/ _/_/ _/_/ "You can tune a file system, but you can't tune a fish." _/_/ _/_/ UNIX man page for tunefs. _/_/ _/_/ _/_/ _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
Newsgroups: comp.sys.next.bugs Subject: Re: My swapfile solution Message-ID: <1995Dec7.001528.20916@indyvax.iupui.edu> From: rsanchez@indyunix.iupui.edu (Richard Sanchez) Date: 7 Dec 95 00:15:27 -0500 References: <DIn0q9.56u@news.hawaii.edu> <49nrq0$4cl@marsu.pilhuhn.de> I have been able to create a separate swap disk by renaming the label of the drive to swapdisk which is then setup up with the rc.swap script on bootup. The problem is that the system will only swap about 130MB before the system hangs, mouse can still be moved, but I am unable to shutdown with the mouse or the keyboard sequence <alt> <alt> <num lock>. The disk is a Seagate 1GB scsi drive so it should be able to swap more than 130MB of space. I tried removing the hiwat argument in the swaptab file and in the rc.swap files. At this point, I am stuck. Any help will be greatly appreciated. -- ----------------------------------------------------------------- Richard Sanchez email: rsanchez@indyunix.iupui.edu Indiana University School of Medicine Indianapolis, Indiana -----------------------------------------------------------------
From: lehmann@iaeb.kloeckner-hb.de (Dirk Lehmann) Newsgroups: comp.sys.next.bugs,comp.sys.next.misc Subject: NSFiP: Sometimes no Sound ! Date: 7 Dec 1995 12:42:49 GMT Organization: Stahlwerke Bremen Sender: lehmann@iaeb (Dirk Lehmann) Distribution: world Message-ID: <4a6ng9$4r1@opuskl.kloeckner-hb.de> Hi, I have a strange problem with the sound output of my SoundBlaster16 VE. The first time this problem occured while playing sound with sndplay from a shell. The sound came out randomly. The same happened when I'm holding the backspace key in a workspace window (this causes NXBeep to play a sound). Sometimes the sound is playing after seconds, sometimes after a minute. The same thing happens while playing a modfile. This problem is also reported by a friend with the same hardware configuration. Configure.app reports _no_ collision with other devices. The soundcard is working fine under '95. My hardware configuration is : Intel Endeavor PCI/Triton Board, NS3.3 with patch SoundBlaster16 VE S3 Displayadapter SMC ELite Ultra 16 2940 SCSI Controller Thanks in advance Dirk
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <flexus!rfschtkt@maze.ruca.ua.ac.be> Date: Thu, 7 Dec 95 21:14:06 +0100 From: Raf Schietekat <flexus!rfschtkt@maze.ruca.ua.ac.be> Message-ID: <9512072014.AA02618@flexus> Subject: Re: Creating a swapfile on 2nd drive (was Re: My swapfile solution) Timothy J. Luoma <luomat@capitalist.princeton.edu> wrote: > One other very important thing I forgot to mention is > that on my NeXT, which uses the swapdisk method, reaches > the HIWAT, the swapping goes back to /private/vm/swapfile. > This might also be considered a "bug" but I like it > because I have written a small script that tells me when > the 2nd swapfile has begun to grow, and I use that > opportunity to reboot. That's a feature, not a bug (IMHO). If you don't want this to happen, you could specify a HIWAT for the /private/vm/swapfile too. Raf Schietekat, RfSchtkt@maze.ruca.ua.ac.be (NeXTmail), Flanders, Belgium If I don't answer: my mail relay can't handle !, % or .uucp, I think *** The year 2000 will be the last year of the 20th century. ***
From: sven.hennig@wiesbaden.netsurf.de (Sven Hennig) Newsgroups: comp.sys.next.bugs Subject: ---Please Read This--- Date: Fri, 08 Dec 1995 12:34:03 GMT Organization: Vistec GmbH Message-ID: <4a9bl2$i6m@luna.vistec.com> Hi, I just wanted to give you the chance to earn some money in a very easy way.A friend of mine turned me on a company called Intercall Marketing, which offers a nice, uncomplicated way to get some more cash, just by callingthem and advertising a bit. The number of the company is 011-24-831-831 (if you're calling from the USA, in other countries it's (i think) ++248-313131). The call has a fee of $1.50, so I think (hope) this isn't to expensive for the possibility to get some more money. Now I'll try to explain how it works: After calling the number a recording gives you deatiled informationon what to to next. After a brief description you will be asked to enter a six digit account number of the person that gave you this number, and this should be my account number. It is -3 8 0 4 9 9- (this one is mine) and then you can start to earn big money for nearly doing nothing. To explain a little how it works, when you type in my account number, the computer will give you you're own account number and a personal PIN Number for your use only. Then it will record name and address for future accounting needs, i.e. payments. Here is where the money comes from: For every call made to this number, intercall receives $1.50. 50 Cents go to the account number you first typed in. But of course, now as you've got you're own account, you'll want people to call and type in your account number right away ! And I think, with the help of all the networking systems around, it won't be a problem to reach many people. Up to now its nothing special, BUT IT GETS EVEN BETTER. You not only make 50 cents from the first people who call using your number but also when they get people to call using their number you get another 25 Cents from them, plus another 25 cents from the people they get to call. So that means you get paid 3 Levels deep, which could amount to some serious $$$. An example: If you get 100 people to call (by advertising in the net, or in the local news papers), then you get 50$. If these people, let's say get another 50 people to call, it's already $1,250. But the third Level is the most important one(so you should perhaps hurry a bit), because if these people get 25 people each, that means that you'll get exactly $31,250 (just imagine, if every Level would get 100 people, you would get $ 250.000). Some good cash for nearly doing nothing, isn't it ? If you're wondering why Intercall does this, realize that they get 50 Cents for every call. 1st Level=$0.50, 2nd Level=$0.25, 3rd Level=$0.25 => $1.00. Remember that they'll get $1.50 for each call, so that leaves 50 cents per call. I normally dont try this things out. But it's (IMHO) not possible to loose much money, and just paying $1.50 for calling, and then perhaps getting much more cash, it's not too bad ? But now I stop trying to convince you. You have to decide if this possibility of earning cash is worth $1.50. Thank you very much for reading, i hope it didn'T bother you. C ya Sven H. ----------------Sven Hennig---------------- ------------Zum Kohlwaldfeld 2a------------ -------------D-65817 Eppstein------------- ------------Tel. ++49-6198-9218------------ ------------Data ++49-6198-9208------------ -E-Mail: sven.hennig@wiesbaden.netsurf.de-
From: mike@hobbs.chem.usu.edu (Michael Emmel) Newsgroups: comp.sys.next.bugs Subject: Re: NS3.3 sleeping Date: 8 Dec 1995 16:16:50 GMT Organization: SouthWind Internet Access, Inc. Distribution: world Message-ID: <4a9odi$5rc@opal.southwind.net> References: <DJ574G.qz@unixhub.SLAC.Stanford.EDU> In article <DJ574G.qz@unixhub.SLAC.Stanford.EDU> bobcook@slac.stanford.edu (Bob Cook) writes: > Michael Emmel writes > > > >Hello I have a non networked home intel machine and I tend to stay logged > >on and turn the screen on sometimes it won't come back from down mode were > >the display is dimmned. No amount of playing with the keyboard or mouse > >will reactivate the window server ? Then I must reboot. > > I have seen behavior like this (on a black NeXT, not Intel) when the machine > has in fact panic'd and crashed. In my case the screen was dimmed a great deal > but not black, so I could dimly see the panic window. > I can see my screen no panic window I know all about that. PPP takes my machine down all the time. I'm not up on my PPP conection. I wish I could telnet in and see : ( I really think the window server goes to sleep and for some reason not checking the event cue. Mabye it has something to do with mach's multitasking ? As far as i know the rom monitor jumps in front of every window ? My monitor is a ViewSonic 21PS and is pretty bright even dimmed so I can still see all the windows. I checked and my display brightness is almost maxed out. The only way to repeat it is to leave the machine for several days while logged in. Common on a home/personal system and I suspect off the internet. Mike
Date: 8 Dec 1995 20:41:02 GMT From: sven.hennig@wiesbaden.netsurf.de (Sven Hennig) Message-ID: <cancel.4a9bl2$i6m@luna.vistec.com> Newsgroups: comp.sys.next.bugs Subject: cmsg cancel <4a9bl2$i6m@luna.vistec.com> Control: cancel <4a9bl2$i6m@luna.vistec.com> Spam cancelled by clewis@ferret.ocunix.on.ca
From: samurai@random.cs.mcgill.ca (Darcy Brockbank) Newsgroups: comp.sys.next.bugs Subject: Re: NeXT->DOS filesystem copy bug ? Date: 08 Dec 1995 18:30:32 GMT Organization: disorganized Message-ID: <SAMURAI.95Dec8133032@random.cs.mcgill.ca> References: <4a5vbk$gjo@news.vanderbilt.edu> In-reply-to: katzlbt@vuse.vanderbilt.edu's message of 7 Dec 1995 05:50:44 GMT <katzlbt@vuse.vanderbilt.edu> writes: >I was copying file from my NeXT to my DOS partition as I >saw that files with size of power of two end up with 0 bytes >on the DOS partition. They show up right in the WS browser >and are also readable. Once I log out and start DOS >I get - makeing a dir - filesize 0 of those files. >Logging in again in NS I get also size 0, deleting them will >free the original size of the file. >Adding a single byte to those files makes the copy work again. >-rw-r--r-- 1 cat 1048576 Oct 20 01:00 disk01.hm >-rw-r--r-- 1 cat 2138112 Aug 14 16:08 sounds >Those are the files producing this behavior reliably. >Has anyone else seen that or can verify this bug ? Yes, it happens to me all the time. I sent in a bug report a long time ago to NeXT but never heard from them. It's a terrible bug. One tends to count on the filesystem to work. - db -- I have a heart condition... if you hit me, it's murder. -- Withnail -- We are millionares! -- and I -- To you, we're not deep -- Housemartins -- A bit small for a share certificate -- Cleese -- Smoke me a kipper, skipper, I'll be back for breakfast. -- Ace Rimmer -- He's the tallest man by a dam site. -- Crow --
From: dave@turbocat.snafu.de (David Wetzel) Newsgroups: comp.sys.next.bugs Subject: S3 Generic Display Driver and screen dimming? Date: 6 Dec 1995 10:54:29 GMT Organization: Turbocat's Development, Germany Message-ID: <4a3sp5$75@turbocat.snafu.de> Hi! Why does the S3 Generic Display Driver (Driver Version 3.32, Prerelease) still not support screen dimming? This cannot be a big thing to code -- since it works fine on HPs (and NeXT has both sources). _ _ _(_)(_)_ David Wetzel, Turbocat's Development, (_) __ (_) Buchhorster Strasse, D-16567 Muehlenbeck/Berlin, FRG, _/ \_ Phone +49 33056 82151, Fax +49 33056 82152 (______) dave@turbocat.snafu.de (NeXTMail)
From: karl@rivendell.osu.edu (Karl N. Matthias) Newsgroups: comp.sys.next.bugs Subject: NeXT Serial Driver Problem? Date: 10 Dec 1995 20:51:36 GMT Organization: InfiNet Message-ID: <4afh8o$27f@horus.infinet.com> I have read several things saying that there is some problem with the NeXT serial drivers and that a new version is available. Can someone tell me what the error is? What is the Mux driver? Thanks Karl
From: kay@kauri.vuw.ac.nz (Cameron Kay) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Trouble with the Harvest cache Date: 11 Dec 1995 03:19:43 GMT Organization: Victoria Uni. of Wellington, NZ. Distribution: world Message-ID: <KAY.95Dec11161943@kauri.vuw.ac.nz> I seem to be having trouble getting the harvest web cache to work under NEXTSTEP 3.3 (Patch 1 installed) for Intel. The cache refuses to create its swap file when I run RunCache and keeps restarting its self claiming there is an error with the DNS lookup. There is the contents of /tmp/cache/cache.log [Monday, 11-Dec-95 02:46:53 GMT] main.c:362: Starting Harvest Cache (version 1.3 pl 2)... [Monday, 11-Dec-95 02:46:53 GMT] ipcache.c:1191: ipcache_init: DNS name lookup appears to be broken on this machine. And heres the contents of /tmp/cache/cached.out cached: WARNING: Cannot write to /tmp/cache/00 for storage swap area. Forcing a *full restart* (e.g., cached -z)... WARNING: No log file specified? messages will be sent to 'stderr'. ipcache_init: DNS name lookup appears to be broken on this machine. cached: WARNING: Cannot write to /tmp/cache/00 for storage swap area. Forcing a *full restart* (e.g., cached -z)... WARNING: No log file specified? messages will be sent to 'stderr'. ipcache_init: DNS name lookup appears to be broken on this machine. Does anyone have harvest running successfully and know how to fix my problem? Thanks. - Cameron -- Email Cameron.Kay@kauri.vuw.ac.nz Post Computer Science Department Phone + 64 4 472 1000 x7032 (Work) Victoria University + 64 4 237 5895 (Home) P.O.Box 600 Fax + 64 4 495 5232 Wellington, New Zealand
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 11 Dec 1995 05:15:09 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <4ageot$9sd@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 200+ ISV company pages - 400+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://nova.cc.purdue.edu: Lots of older stuff ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: steve@tweedledee.ucsb.edu (Steve Trainoff) Newsgroups: comp.sys.next.bugs Subject: Indexing WriteNow files Date: 12 Dec 1995 04:34:43 GMT Organization: University of California, Santa Barbara Distribution: world Message-ID: <4aj0p3$ktb@yuggoth.ucsb.edu> Hi all, I have a lot of WriteNow files which I was attempting to index. I had a lot of trouble which I eventually tracked down to a bug in WriteNow.app/wn-rtf. It didn't properly deal with the case when the WN file is an app wrapper. In that case the actual text is in the file foo.wn/WNDocument.wn. I hacked this simple fix. To use it compile with: cc -O -o fixed-wn-rtf fixed-wn-rtf.c and put it in /LocalApps/WriteNow.app (If you have WN installed elsewhere you will have to change a #define at the top). To get Librarian.app to use the fixed version put the following lines into /LocalLibrary/ThirdPartyFilters.service Filter: Port: NXUNIXSTDIO Send Type: NXTypedFilenamePboardType:wn Return Type: NXRTFPboardType Executable: /LocalApps/WriteNow.app/fixed-wn-rtf ---------------- fixed-wn-rtf.c -------------- /* Steven Trainoff (C)opyright 1995 */ #include <stdlib.h> #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <sys/param.h> #define WN_RTF "/LocalApps/WriteNow.app/wn-rtf" int main(int argc, char **argv) { char actualFile[MAXPATHLEN]; struct stat buf; if (argc != 2 || stat(argv[1], &buf)) { fprintf(stderr, "Usage: fixed-wn-rtf filename\n"); exit(-1); } /* Check if we were handed a directory */ strcpy(actualFile, argv[1]); if (buf.st_mode & S_IFDIR) strcat(actualFile, "/WNDocument.wn"); execl(WN_RTF, "wn-rtf", actualFile, NULL); exit(0); /* Not reached */ } --------------- END --------------------- -- ..STeve ------------------------------------ Insert pithy maxim here... steve@tweedledee.ucsb.edu (NeXT mail)
From: toddw@radium.ca (Todd White) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: Re: Trouble with the Harvest cache Date: 14 Dec 1995 07:37:12 GMT Organization: ftn Internet Distribution: world Message-ID: <4aok78$otl@master.ftn.net> References: <KAY.95Dec11161943@kauri.vuw.ac.nz> In <KAY.95Dec11161943@kauri.vuw.ac.nz> Cameron Kay wrote: > I seem to be having trouble getting the harvest web cache to work > under NEXTSTEP 3.3 (Patch 1 installed) for Intel. The cache refuses to > create its swap file when I run RunCache and keeps restarting its self > claiming there is an error with the DNS lookup. > > > Does anyone have harvest running successfully and know how to fix my > problem? > > Thanks. > > - Cameron > > Is this the RADium Technology Centre (Canada) port of Harvest 1.3? If so, just send E-mail to support@radium.ca and someone will help you. Cheers, -- ===================================== Todd White RADium Technology Centre (Canada) toddw@RADium.ca NeXT and MIME mail gladly accepted =====================================
From: "Biblioteca F. Trisi Lugo (RA) ITALY" <bibtrisilug@ra.nettuno.it> Newsgroups: comp.sys.next.bugs Subject: ATI MACH 32 PCI flicker Date: 14 Dec 1995 16:02:14 GMT Organization: Cineca Message-ID: <4aphq6$mrd@sirio.cineca.it> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit My computer configuration is: -motherboard: Genoa PCI Turboexpress G version; chipset SATURN-II 06/23/94 number 2A498F21-00 -bios: of motherboard AWARD V4.50G; of NCR53810 version 3.04 -hard disk: FUJITSU M2684S-512 rev 2026 -cdrom: NEC model 55 bios 1.0 revision 01 at ID 02 -graphic card: ATI mach32 ULTRA PRO 2mb VRAM BIOS rev 1.4 (1993/12/6) I have installed NextStep 3.3 and I view some pixel flickering, expecially in 256 gray mode at the login prompt. My hardware works well under dos linux os2 so it is not faulty, perhaps a bios or pci or bug problem ??? Please Help me if You can! Please mail me to giammarc@cs.unibo.it thank You in advance!
From: alii@eden.rutgers.edu (Syed Ali) Newsgroups: comp.sys.next.hardware,comp.sys.next.bugs,comp.sys.next.advocacy,comp.sys.next.misc Subject: Adaptec SCSI 1522 SCSI host adapterm NEC SCSI external 6x CD_ROM Date: 16 Dec 1995 00:50:53 -0500 Organization: Rutgers University Message-ID: <4atmnt$535@er5.rutgers.edu> Summary: Adaptec SCSI 1522 SCSI host adapterm NEC SCSI external 6x CD_ROM Keywords: Adaptec SCSI 1522 SCSI host adapterm NEC SCSI external 6x CD_ROM Hi. I am trying to install NEXTSTEP 3.2 on a gw2000 pentium 60 with 40MB of RAM, EIDE HD, Adaptec SCSI host adapter 1522, NEC SCSI external 6x cd-rom, but am having problems making the boot disk load the SCSI adapter driver. I bought the cd-rom ($600) and adapter ($200) just so that I can install NEXSTEP because my EIDE Mitsumi 4x is not supported. http://www.next.com shows that NEXTSTEP supports Adaptec host adapter series 154xx so I figured 1522 would also be supported. I tried to set my NEC cd-rom with ID 0 and also 6 ( I got this info from the home page of NEXTSTEP) w/o any luck. I downloaded the additional drivers for 3.2 compressed file but I cannot uncompress on my PC running NT since uncompress is a UNIX utitlity. Besides I do not know if the 3.2 additional drivers disk will support the SCSI host adapter I have. ( Adaptec 1522 ) I saw NEXTSTEP running on a SPARC and liked it a lot and would love to be able to install it. Anyone has any clue on how I can install it w/o having to return my NEC and Adaptec SCSI adapter and buying ones that are explicitly listed as supported on the NEXT home page? Thank you for any help! Regards, Syed Ali. ---------------------------------------------------------------- Systems Operator, LCSR-CF, Rutgers University, New Brunswick, NJ.
From: katzlbt@vuse.vanderbilt.edu (Thomas Katzlberger) Newsgroups: comp.sys.next.bugs Subject: How to activate DPMS ? Date: 16 Dec 1995 22:17:24 GMT Organization: a black NeXT Message-ID: <4avghk$nt3@news.vanderbilt.edu> How can I activate Display power management in NS ? The CPU slows down, ok but I can't get the display to power down while the system is unused. It works in DOS, although it is strange that the system when tranfering to Suspend mode speeds up to doze speed instead of slowing down or staying at standby speed. hmmm, and I can't get any PM event to match the BUS mouse, if I activate IRQ4 (mouse) I disable the power management, only disabeling all IRQ's helps. Another strange thing sometimes when I login it takes 5 minutes to startup loginwindow.app with 3 errormessages that it can't find the window server. however after the 5 min's everyting works fine. Has anyone seen compareable things ? Cat. -- _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ _/_/ _/_/ _/_/ Thomas Katzlberger _/_/ _/_/ katzlbt@vuse.vanderbilt.edu _/_/ _/_/ @aBlackNeXT.called.garfield _/_/ _/_/ http://www.vuse.vanderbilt.edu/~katzlbt/ _/_/ _/_/ _/_/ _/_/ "You can tune a file system, but you can't tune a fish." _/_/ _/_/ UNIX man page for tunefs. _/_/ _/_/ _/_/ _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.programmer,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.soft-sys.nextstep Subject: NEXTSTEP Resources on the Internet Date: 18 Dec 1995 05:15:11 GMT Organization: Digital Fix Development Distribution: inet Message-ID: <4b2tcv$1i9@digifix.digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - 188+ ISV company pages - 433+ ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - NEXTSTEP User Group Directory - comp.sys.next archives - User Group information - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Additionally there is a Mail Server available. You can get information on using the mail server at ns-products@stepwise.com Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news:comp.sys.next.advocacy This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. news:comp.sys.next.announce Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Archives are available by ftp at ftp://ftp.stepwise.com/pub/Next_Announce_Archives Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. news:comp.sys.next.bugs A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT- specific groups as well. news:comp.sys.next.hardware Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. news:comp.sys.next.marketplace NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. news:comp.sys.next.misc For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! news:comp.sys.next.programmer Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news:comp.soft-sys.nextstep Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news:comp.object Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep-d next-advocacy-d next-announce-d next-bugs-d next-hardware-d next-marketplace-d next-misc-d next-programmer-d next-software-d next-sysadmin-d (For a full description, send mail saying LISTS to <digestif@antigone.com>). The subscription syntax is essentially the same as LISTSERV's. To subscribe, send a message to <digestif@antigone.com> saying: SUB Listname YourName Example: SUB next-hardware-d John Doe The ftp sites ============= ftp://ftp.cs.orst.edu: The main site for North American submissions ftp://ftp.informatik.uni-muenchen.de: (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next Peanuts mirror in the US ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) and ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next/ eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO <your-address> If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! Written by: Eric P. Scott (mailto:eps@toaster.SFSU.EDU) and Scott Anguish (mailto:sanguish@digifix.com) Additions from: Greg Anderson (mailto:Greg_Anderson@afs.com) Michael Pizolato (mailto:alf@epix.net) Dan Grillo (mailto:dan_grillo@next.com)
From: altenber@acpub.duke.edu (Lee Altenberg) Newsgroups: comp.sys.next.bugs Subject: Socket() problems in mixed network Date: 18 Dec 1995 22:47:44 -0500 Organization: Duke University, Durham, NC, USA Message-ID: <4b5cl0$1vi@news.duke.edu> I am having problems sending data over a socket between a NS/Intel 3.3 machine and other Unix boxes (SGI Onyx and IBM AIX RS/6000). I am running a simple test program that is run on both machines connecting by the socket, which then pass blocks of data back and forth between them. When this program is run between two NS/Intel 3.3 machines, it works fine. When it runs on a NS/Intel 3.3 machine and either an AIX RS/6000 or an SGI Onyx, the data don't get sent (or received) properly. Furthermore, blocking doesn't seem to be working, as one machine will race ahead of the NS/Intel 3.3 machine. The program works fine when running between combinations of SGI and RS/6000 machines. (It's embarrassing trying to be a NS advocate to the systems people here when this kind of stuff happens). Has anyone heard of socket problems with NS/Intel 3.3? Some protocol flags needed perhaps? NeXTAnswers has nothing, nor does the "Known Bugs in NS" FAQ. Any clues would greatly be appreciated. Lee Altenberg altenber@mhpcc.edu altenber@acpub.duke.edu
From: rdieter@mathlab44.unl.edu (Rex Dieter) Newsgroups: comp.sys.next.bugs Subject: Re: Adaptec SCSI 1522 SCSI host adapterm NEC SCSI external 6x CD_ROM Date: 19 Dec 1995 15:00:04 GMT Organization: University of Nebraska--Lincoln Message-ID: <4b6k1l$f5p@crcnis3.unl.edu> References: <4atmnt$535@er5.rutgers.edu> In article <4atmnt$535@er5.rutgers.edu> alii@eden.rutgers.edu (Syed Ali) writes: > I bought the cd-rom ($600) and adapter ($200) just so that I can > install NEXSTEP because my EIDE Mitsumi 4x is not supported. FYI, the EIDE Mitsumi 4x IS supported by the new EIDE/ATAPI driver. I've used it to install NEXTSTEP-3.3. > I tried to set my NEC cd-rom with ID 0 and also 6 ( I got this info > from the home page of NEXTSTEP) w/o any luck. > > I downloaded the additional drivers for 3.2 compressed file but I cannot > uncompress on my PC running NT since uncompress is a UNIX > utitlity. Besides I do not know if the 3.2 additional drivers disk > will support the SCSI host adapter I have. ( Adaptec 1522 ) I would recommend that you use NEXTSTEP-3.3, which has many more, robust device drivers available. -- Rex A. Dieter rdieter@math.unl.edu (NeXT/MIME) Computer System Manager Voice: (402)472-9747 Department of Mathematics and Statistics FAX: (402)472-8466 University of Nebraska - Lincoln http://www.math.unl.edu/~rdieter/
From: dmi@demandmedia.com Newsgroups: comp.sys.next.bugs Subject: Workspace dies when accessing NFS mounted filesystems... Date: 19 Dec 1995 21:32:31 GMT Organization: Pencom Software Message-ID: <4b7b1f$i1g@digdug.pencom.com> I have a _very_ repeatable bug. If I log in as 'root', and mount an NFS filesystem from a server -- that has not exported the filesystem with "root" access -- when I change the Fileviewer into that directory, Workspace dies. Has anyone else seen this behavior? I'm on NS 3.3 Intel, with the user and developer patches. -- ----------------------------------------------------------------------------- *** These are my opinions... Mine! All Mine! Minemineminemineminemine! *** ----------------------------------------------------------------------------- Robin D. Wilson robin@pencom.com Pencom Software 701 Canyon Bend Dr. 9050 Capital of Texas Hwy Pflugerville, TX 78660 Austin, TX 78759 (512) 251-1737 (512) 343-6666
From: doyle@zeke.lanl.gov Newsgroups: comp.sys.next.bugs Subject: Re: Workspace dies when accessing NFS mounted filesystems... Date: 19 Dec 1995 23:20:44 GMT Organization: Los Alamos National Laboratory Message-ID: <4b7hcc$7rl@newshost.lanl.gov> References: <4b7b1f$i1g@digdug.pencom.com> dmi@demandmedia.com wrote: > I have a _very_ repeatable bug. If I log in as 'root', and mount an NFS > filesystem from a server -- that has not exported the filesystem with "root" > access -- when I change the Fileviewer into that directory, Workspace dies. > > Has anyone else seen this behavior? > > I'm on NS 3.3 Intel, with the user and developer patches. Yes, I have seen it in NS 3.0 (or maybe even 2.1) on black. If I remember correctly, the problem goes away if you make the directory world-executable. I believe I reported it to NeXT at the time (about 2 years ago). It does seem like the proper behavior not to be able to read the filesystem, but to crash Workspace is a terrible bug in my opinion. Just pop up a panel with an error message. Cheers, Mark
From: Paul Lynch <Paul_Lynch@plsys.com> Newsgroups: comp.sys.next.bugs Subject: Re: Workspace dies when accessing NFS mounted filesystems... Date: Tue, 19 Dec 1995 23:36:12 GMT Organization: P & L Systems Sender: news@seer.demon.co.uk Message-ID: <1995Dec19.233612.2354@seer.demon.co.uk> References: <4b7b1f$i1g@digdug.pencom.com> In article <4b7b1f$i1g@digdug.pencom.com> dmi@demandmedia.com writes: > I have a _very_ repeatable bug. If I log in as 'root', and mount an NFS > filesystem from a server -- that has not exported the filesystem with "root" > access -- when I change the Fileviewer into that directory, Workspace dies. > > Has anyone else seen this behavior? > > I'm on NS 3.3 Intel, with the user and developer patches. It's a feature; I've seen this on all versions of NeXTSTEP that I use. Once qualification: this should only happen if world doesn't have read permission on that directory, and so root (aka nobody) can't read it. Paul -- Paul Lynch (NeXTmail) http://www.plsys.co.uk/~paul (under construction)
From: mike@starburst.cbl.cees.edu (Michael F. Santangelo) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs,comp.sys.next.hardware Subject: /bin/login -and- new ISASerialPorts/PortServer Date: 20 Dec 1995 20:09:07 GMT Organization: University of Maryland, Chesapeake Biological Lab. Message-ID: <4b9qh3$af6@gamera.cbl.cees.edu> Summary: /bin/login via getty not passing term characteristics properly? Keywords: /bin/login,ISASerialPort,PortDriver I had 2 internal MODEMs (28.8 GATEWAY TELEPATH II's, simulated 16550 UARTs internal to each) on COM2 and COM3 on my P5-100 Intel TRITON motherboard running with the old NeXTstep 3.3 serial driver running getty's at 19200. I had it working okay despite some quirkyness here and there and except for occasional overrun messages in /usr/adm/messages. So I went poking around www.next.com and found the new ISASerialPort and PortServer drivers and installed them. The new drivers themselves seem to work great, no overruns, kermit can talk to the ports without a hitch now (had some problems before), etc. HOWEVER: getty'ing the ports for dialin has problems. I have "p8" set on the default option, and that prints the initial banner and login message okay, accepts the username, and then when that forks /bin/login everything breaks down, I get jibberish and it won't read anything. I am of course coming in at 8 bits no parity 1 stop (and need to run the port at 8 bit clean all the way through). I have tried all manner of gettytab parity settings and only "zp:ap" seems to at least accept a password from the keyboard despite the string "Password" and the /etc/motd file being echo'd in jibberish until it gets to the user's .login where "stty pass8" is executed then all is well again. This however is HIGHLY annoying and didn't screw up like this with the old serial port driver (despite its other problems). It appears as if /bin/login can't set its modes correctly to the new driver when forked by getty. Can f0, f1, or f2 in gettytab help? I have tried them somewhat with no success (no effect). ANY ideas appreciated here but I would rather not go back to the old (broken) serial port driver. Thanks! -Mike -- -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- Mike F. Santangelo, Dept. Head-Computer & Network Systems, UMCEES/CBL Solomons
From: jq@papoose.quick.com (James E. Quick) Newsgroups: comp.sys.next.bugs Subject: Re: Socket() problems in mixed network Date: 20 Dec 1995 15:14:23 -0500 Organization: PHCS Message-ID: <4b9qqv$4rg@papoose.quick.com> References: <4b5cl0$1vi@news.duke.edu> In article <4b5cl0$1vi@news.duke.edu>, Lee Altenberg <altenber@acpub.duke.edu> wrote: >I am having problems sending data over a socket between a NS/Intel 3.3 >machine and other Unix boxes (SGI Onyx and IBM AIX RS/6000). I am >running a simple test program that is run on both machines connecting >by the socket, which then pass blocks of data back and forth between them. > >When this program is run between two NS/Intel 3.3 machines, it works >fine. When it runs on a NS/Intel 3.3 machine and either an AIX RS/6000 >or an SGI Onyx, the data don't get sent (or received) properly. (munch) Your problem is that the SGI and RS6000 boxes are both big-endian architectures while Intel chips are little-endian. This means that all values whose size is larger than a byte have different representations. Big endian short 1-2 Little endian short 2-1 Big endian int 1-2-3-4 Little endian int 4-3-2-1 In order to be portable, any multi-byte values must be converted to a common byte order (usually big endian) before being sent. The values are then converted to the native host order upon receipt. The following documentation should help you: First read Concepts/PortabilityGuide.rtf (in the online developer docs). also available online are the man pages for: byteorder(3) defines the ntohs(), ntohl() etc. functions. xdr(3) describes the implementation of the sun xdr library. You migh also want to consult the bible: W. Richards Stevens network programming book. >Furthermore, blocking doesn't seem to be working, as one machine will >race ahead of the NS/Intel 3.3 machine. I do not know what you mean by blocking here. TCP/IP is a stream oriented protocol, not block oriented. You do need to be sure that when you post a read for a fixed block of date that, you followup with subsequent reads when you got less then you wanted. -- ___ ___ | James E. Quick jq@quick.com / / / | Private HealthCare Systems NeXTMail O.K. \_/ (_\/ | Systems Integration Group (617) 895-3343 ) | - My other car has a mouse.
From: jmbettems@ping.ch (Jean-Marie Bettems) Newsgroups: comp.sys.next.bugs Subject: Weird bug Date: 21 Dec 1995 20:40:28 GMT Organization: CyberLink, Switzerland Message-ID: <4bcgns$dta@www.cyberlink.ch> Hello! Since a couple of weeks I experience some weird bugs. I do not know if it is a software problem or a hardware problem (memory). Any help is welcomed. My configuration: - NeXTstation color with 32MB RAM and a Pyro accelerator - NEXTSTEP 3.3 user + developer, patch 1 Bugs description: any of the following randomly occurs - DPS errors, e.g.: Fiend[198]: DPS client library error: PostScript program error, DPSContext b2b1c Fiend[198]: %%[ Error: undefined; OffendingCommand: get ]%% - Workspace errors, e.g.: Workspace, internal error 1000 or Workspace[483]: *** Unreadable data for apps in search path: /LocalApps Remarks: once such an error has occurred, new errors will follow as long as I haven't made a power_down/power_on cycle. In particular rebooting does not help! --- _____________________________________________________________________ Jean-Marie Bettems Phone (prof.): +41 1 256 93 03 Obstgartenstrasse 7 Phone (priv.): +41 1 803 07 30 CH-8302 Kloten E-Mail jmbettems@ping.ch Switzerland _____________________________________________________________________
Newsgroups: comp.sys.next.bugs From: csaldanh@mae.carleton.ca (Chris Saldanha) Subject: Re: NeXT->DOS filesystem copy bug ? Message-ID: <DJyH65.CwC@cunews.carleton.ca> Sender: news@cunews.carleton.ca (News Administrator) Organization: Carleton University References: <4a5vbk$gjo@news.vanderbilt.edu> <SAMURAI.95Dec8133032@random.cs.mcgill.ca> Date: Thu, 21 Dec 1995 21:32:29 GMT Darcy Brockbank (samurai@random.cs.mcgill.ca) wrote: : >and are also readable. Once I log out and start DOS : >I get - makeing a dir - filesize 0 of those files. : >Logging in again in NS I get also size 0, deleting them will : >free the original size of the file. : >Adding a single byte to those files makes the copy work again. : Yes, it happens to me all the time. I sent in a bug report a long time : ago to NeXT but never heard from them. It's a terrible bug. One tends to : count on the filesystem to work. I find it only happens when I write to a DOS partition that's getting very full (ie < 5MB from filling). This behaviour is much worse since I installed Win95 over DOS, but I'm not sure if it's a coincidence. DOS ScanDisk is capable of restoring the FAT entries for the file properly, without having to append bytes to them or anything. --Chris Chris Saldanha, Software Analyst -------------------------------------- computerActive, Inc |"The telephone was not invented by | chris@computerActive.on.ca (NeXTMail) | Alexander Graham Unitel" -Bell Ad | http://www.mae.carleton.ca/~csaldanh --------------------------------------
From: Mark Crispin <mrc@CAC.Washington.EDU> Newsgroups: comp.sys.next.advocacy,comp.sys.next.bugs Subject: NEXTSTEP sd driver doesn't allow more than 7 spindles! Date: Thu, 21 Dec 1995 20:17:47 -0800 Organization: Networks & Distributed Computing Message-ID: <Pine.NXT.3.92.951221195955.1077C-100000@Tomobiki-Cho.CAC.Washington.EDU> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII NNTP-Posting-User: guest So, I go and buy a CD-ROM changer which can hold 7 CDs. I hook it up to one of the NeXTs and try to use it. OK, so it thrashes a bit doing all those changes from CD to CD, but it settles down and seems to work. Uh-oh. I'm only seeing 6 CDs. What's wrong? I look at the system startup log. Sure enough, it's only recognizing lun 0 through lun 5 on target 4, so the last assigned sd is sd6. Damn, is something wrong with the changer??? Take a deep breath. This is NEXTSTEP, after all. It's probably the software, not the hardware. I hook up one of the old black CD-ROM drives at SCSI target 5 and reboot. Hey, that's wierd, it doesn't see that drive at all. Is the CD-ROM changer whomping the bus somehow? Hmm, I wonder what happens if I see the black CD-ROM drive at target 3, a lower number than the CD-ROM changer (target 4). Hmm, it sees the black drive just fine. Hmm, now it only recognizes lun 0 through lun 4 on the CD-ROM changer, so the last assigned sd is sd6. Ah hah! Let's try putting in another black CD-ROM drive. First at target 5. Nope, it doesn't see it. Now at target 2. There it is, and now the changer stops at lun 3. But, but, the sd device goes up to sd15. Let's try to access /dev/sd7a. "Please mount sd7a in drive 1". What the hell? Read the man page. device_name as sd[0-6] at sc0 target [0-6] lun [0-7] But, that's only 7 sd's for 56 possible spindles. What are the other sd's for? Removable volumes!! Gack, they couldn't have given these a different device name. So, if you were thinking about using a CD-ROM changer with NEXTSTEP, remember that you're limited to 7 spindles including your internal hard disk. You'll have to use a real operating system from a real vendor to be able to make full use of the changer. -- Mark -- DoD #0105, R90/6 pilot, FAX: (206) 685-4045 ICBM: N 47 39'35" W 122 18'39" Science does not emerge from voting, party politics, or public debate.
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: G.C.Th.Wierda@AWT.NL (Drs. G. C. Th. Wierda) Subject: How a cabling problem hung my machine Message-ID: <DJzCs3.9v@AWT.NL> Sender: gerben@AWT.NL (Drs G. C. Th. Wierda) Organization: Adviesraad voor het Wetenschaps- en Technologiebeleid Date: Fri, 22 Dec 1995 08:55:15 GMT Well, sort of. I have configured NetWare to be used. At boot time, NetWare is started if some Netinfo property is set. But if the net is bad you have to Ctrl-C the NetWare startup. The result is that the loginwindow des not appear. Reason: npsd is not running or not working properly. I think it is a bug if NEXTSTEP doesn't start up properly if there is such a NetWare problem. Solution: I created a rc.noNetWare which contains the NetWare stuff commented out. That way, I can in a case like this boot single-user, replace rc with rc.noNetWare and boot normally. Yours, -- Gerben Wierda, Stafmedewerker Adviesraad voor het Wetenschaps- en Technologiebeleid. Javastraat 42, 's-Gravenhage, The Netherlands Tel (+31) 70 3639922 Fax (+31) 70 3608992
Newsgroups: comp.sys.next.bugs Organization: Antigone Press gateway, San Francisco Return-Path: <flexus!rfschtkt@maze.ruca.ua.ac.be> Date: Thu, 21 Dec 95 13:56:00 +0100 From: Raf Schietekat <flexus!rfschtkt@maze.ruca.ua.ac.be> Message-ID: <9512211256.AA27598@flexus> Subject: Re: Workspace dies when accessing NFS mounted filesystems... Paul Lynch <Paul_Lynch@plsys.com> writes: > It's a feature; I've seen this on all versions of NeXTSTEP > that I use. You *must* be kidding! This is about KBNS.33.1.010, which was hitherto reported for 3.2 only. ftp://ftp.cs.orst.edu/software/NeXT/documents/KBNS.33.1.README.rtf.gz ftp://ftp.cs.orst.edu/software/NeXT/documents/KBNS.33.1.rtf.gz Raf Schietekat, RfSchtkt@maze.ruca.ua.ac.be (NeXTmail), Flanders, Belgium If I don't answer: my mail relay can't handle !, % or .uucp, I think *** Santa Claus is a fraud. Fortunately for him, Sinterklaas (or Saint Nicholas for the English spoken) is a *real* saint, and he's not considering litigation. Sinterklaas brings presents on the 6th of December each year. If you think I'm biased because I live in Sint-Niklaas (sic), you don't have to take my word for it: look it up in Webster's! ***
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.advocacy,comp.sys.next.bugs Subject: Re: NEXTSTEP sd driver doesn't allow more than 7 spindles! Date: 22 Dec 1995 18:30:26 GMT Organization: monoChrome, Inc. Message-ID: <4betg2$4e4@news4.digex.net> References: <Pine.NXT.3.92.951221195955.1077C-100000@Tomobiki-Cho.CAC.Washington.EDU> Mark Crispin <mrc@CAC.Washington.EDU> wrote: > So, if you were thinking about using a CD-ROM changer with > NEXTSTEP, remember that you're limited to 7 spindles including > your internal hard disk. You'll have to use a real operating > system from a real vendor to be able to make full use of the > changer. Yikes, that's stinky... Definatly worse than the 2gig partition limit. I hope NeXT fixes this for 4.0 as well. Thanks for finding that Mark, I was thinking of getting one of thsoe 7 CD things as well (only 350). -- Thanks, take care, later, John Kheit )^> %^) =^) monoChrome, Inc. | New York Law School NEXTSTEP Developer | Opinions expressed represent me only... Telepathy, It's coming | MIME & NeXTmail OK--PPP (3P) jkheit@cnj.digex.net | http://cnj.digex.net/~jkheit
From: rragner@panther.vm.iastate.edu (Rod Ragner) Newsgroups: comp.sys.next.bugs Subject: Re: NEXTSTEP sd driver doesn't allow more than 7 spindles! Date: 22 Dec 1995 18:59:20 GMT Organization: Iowa State University, Ames, Iowa Message-ID: <4bev68$mu8@news.iastate.edu> References: <4betg2$4e4@news4.digex.net> In article <4betg2$4e4@news4.digex.net> John Kheit <jkheit@cnj.digex.net> writes: > Mark Crispin <mrc@CAC.Washington.EDU> wrote: > > So, if you were thinking about using a CD-ROM changer with > > NEXTSTEP, remember that you're limited to 7 spindles including > > your internal hard disk. You'll have to use a real operating > > system from a real vendor to be able to make full use of the > > changer. > > Yikes, that's stinky... Definatly worse than the 2gig partition > limit. I hope NeXT fixes this for 4.0 as well. > > Thanks for finding that Mark, I was thinking of getting one of > thsoe 7 CD things as well (only 350). > > -- > Thanks, take care, later, John Kheit )^> %^) =^) > > monoChrome, Inc. | New York Law School > NEXTSTEP Developer | Opinions expressed represent me only... > Telepathy, It's coming | MIME & NeXTmail OK--PPP (3P) > jkheit@cnj.digex.net | http://cnj.digex.net/~jkheit Will not this be true of all operating systems, because the SCSI adapter only supports 7 devices? NEXTSTEP does support multiple SCSI controllers, so you could buy one controller and a 7 CD changer and another controller for your hard disk drives. Actually, many people suggest multiple controllers to improve system performance when you have more than 4 SCSI devices. -- Rod Ragner, UNIX Systems Administrator/NEXTSTEP Application Developer Veterinary Diagnostic Laboratory, College of Veterinary Medicine, Iowa State University, 2630 Vet. Med. Bldg., Ames, Iowa 50011 Voice: (515) 294-4751, FAX: (515) 294-6961 or 3564, (NeXT Mail accepted) Email: rragner@stallion.vm.iastate.edu or stryder@iastate.edu
From: Dominik Westner <dominik@gowest.ppp.informatik.uni-muenchen.de> Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs Subject: Categories not recognized by InterfaceBuilder Date: 22 Dec 1995 19:37:51 GMT Organization: TeDoc, Munich, Germany Message-ID: <4bf1ef$263@cube.ppp.informatik.uni-muenchen.de> Hi folks, I encountered the following problem concerning the use of categories within InterfaceBuilder: There are several cases to consider: 1) Category for a custom class myclass inheriting from Object: This works fine. I can drag any header file which contains a category interface and the method will show up for myclass. 2) Category for a NeXT class inheriting from Object: This does not work properly. The method does NOT show up in the classes view for the class. WORKAROUND: Before dragging the category header file into InterfaceBuilder, define it as follows: @interface Application:Responder (Test) - test: sender; @end then remove ":Responder" (This is not valid ObjectiveC). NOTE: Sometimes the other action methods defined in the class disapear. 3) Category for a custom class myclass inheriting from NSObject: This is the same as 2). 4) Category for a NeXT class inheriting from NSObject: This is the same as 2) but the NOTE seems to always hold in this case. Valid for Release 3.3 (v460) Is there any proper workaround for the cases 2) - 4)??? Is this a known problem? Am I making something wrong? Thanks and a Happy New Year. -- Dominik --- Dominik Westner "Everything is possible in an infinite universe" _____________________________________Willi, the painter Munich, Germany (NOT :-)
From: chris@miles.opensource.com (Chris Miner) Newsgroups: comp.sys.next.bugs Subject: Re: NEXTSTEP sd driver doesn't allow more than 7 spindles! Date: 22 Dec 1995 20:34:25 GMT Organization: OpenSource Inc. Distribution: world Message-ID: <4bf4oh$lb@trane.opensource.com> References: <4bev68$mu8@news.iastate.edu> In article <4bev68$mu8@news.iastate.edu> rragner@panther.vm.iastate.edu (Rod Ragner) writes: > > Will not this be true of all operating systems, because the SCSI adapter > only supports 7 devices? > The CD-Changer is one device (one id, one device), with several logical units. So a person might reasonably expect that all the disks would be accessible. Right? Chris
From: "Jonathan W. Hendry" <steeldrv@ix.netcom.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.bugs Subject: Re: NEXTSTEP sd driver doesn't allow more than 7 spindles! Date: Fri, 22 Dec 1995 21:13:18 -0500 Organization: Steel Driving Software, Inc. Message-ID: <30DB65BE.7D0@ix.netcom.com> References: <Pine.NXT.3.92.951221195955.1077C-100000@Tomobiki-Cho.CAC.Washington.EDU> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Mark Crispin wrote: >So, if you were thinking about using a CD-ROM changer with NEXTSTEP, >remember that you're limited to 7 spindles including your internal hard >disk.You'll have to use a real operating system from a real vendor to > be able to make full use of the changer. Hmm. SCSI is (or was) limited to 7 devices on a chain. Perhaps the individual CD's are being treated as separate SCSI devices, instead of the changer being treated as one device on the chain? I'm surprised the CD's aren't treated as removable. When you start up with the changer empty, how many CD's can you insert? Perhaps there's a difference in how they are mounted which depends on whether it's at bootup or later. - Jon -- Jonathan W. Hendry Steel Driving Software, Inc. Delphi and NeXTSTEP consulting and software development. Cincinnati, Ohio
From: Paul_Lynch@plsys.com (Paul Lynch) Newsgroups: comp.sys.next.bugs Subject: Re: Workspace dies when accessing NFS mounted filesystems... Date: Sat, 23 Dec 1995 08:10:17 GMT Organization: P & L Systems Sender: news@seer.demon.co.uk Message-ID: <1995Dec23.081017.11711@seer.demon.co.uk> References: <9512211256.AA27598@flexus> In article <9512211256.AA27598@flexus> Raf Schietekat <flexus!rfschtkt@maze.ruca.ua.ac.be> writes: > Paul Lynch <Paul_Lynch@plsys.com> writes: > > It's a feature; I've seen this on all versions of NeXTSTEP > > that I use. > You *must* be kidding! My apologies for any confusion. I was using "feature" in the relatively common sense of a bug that, for whatever reason, has become so well entrenched that we expect it to be around forever (unlike the product that contains it). Paul -- Paul Lynch (NeXTmail) http://www.plsys.co.uk/~paul (under construction)
From: dave@turbocat.snafu.de (David Wetzel) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs,comp.sys.next.hardware Subject: Re: /bin/login -and- new ISASerialPorts/PortServer Date: 22 Dec 1995 11:44:12 GMT Organization: Turbocat's Development, Germany Message-ID: <4be5mc$d8@turbocat.snafu.de> References: <4b9qh3$af6@gamera.cbl.cees.edu> mike@starburst.cbl.cees.edu (Michael F. Santangelo) wrote: (...) > HOWEVER: getty'ing the ports for dialin has problems. I have "p8" set > on the default option, and that prints the initial banner and login > message okay, accepts the username, and then when that forks > /bin/login everything breaks down, I get jibberish and it won't read (...) Use mgetty from ftp.leo.org. /etc/ttys: cufb "/usr/local/sbin/mgetty" unknown on Works better than the original one. _ _ _(_)(_)_ David Wetzel, Turbocat's Development, (_) __ (_) Buchhorster Strasse, D-16567 Muehlenbeck/Berlin, FRG, _/ \_ Phone +49 33056 82151, Fax +49 33056 82152 (______) dave@turbocat.snafu.de (NeXTMail)
From: caradoc@neta.com (Santa ) Newsgroups: comp.sys.next.bugs Subject: Santa's Xmas Gift to you all Date: Sun, 24 Dec 1995 23:44:01 GMT Organization: North Pole - Spam Central Message-ID: <4bkklb$a0u@news1.goodnet.com> Rudolf, the red-nosed raindeer. Had a very shiny nose - and if you ever saw it you would even say it glows. ................ Then one foggy Christmas night, Santa came and said, Rudolf with your nose so bright, won't you SPAM the world tonight.............................................. YES I WILL SANTA SPAM SPAM SPAM SPAM SPAM
From: caradoc@neta.com (Santa ) Message-ID: <cancel.4bkklb$a0u@news1.goodnet.com> Control: cancel <4bkklb$a0u@news1.goodnet.com> Newsgroups: comp.sys.next.bugs Subject: cmsg cancel <4bkklb$a0u@news1.goodnet.com> Date: Mon, 25 Dec 1995 08:16:49 KST UDP SPAM cancelled by jem@xpat.postech.ac.kr.
Newsgroups: comp.sys.next.advocacy,comp.sys.next.bugs From: rdl@world.std.com (Robert La Ferla) Subject: Re: NEXTSTEP sd driver doesn't allow more than 7 spindles! In-Reply-To: "Jonathan W. Hendry"'s message of Fri, 22 Dec 1995 21:13:18 -0500 Message-ID: <RDL.95Dec26012132@world.std.com> Sender: rdl@world.std.com (Robert La Ferla) Organization: The World Public Access UNIX, Brookline, MA References: <Pine.NXT.3.92.951221195955.1077C-100000@Tomobiki-Cho.CAC.Washington.EDU> <30DB65BE.7D0@ix.netcom.com> Date: Tue, 26 Dec 1995 06:21:32 GMT Does anyone else think that NeXT should comment on this? Robert La Ferla Registered OPENSTEP / NEXTSTEP Consultant Boston, MA Tel: + 1 (617) 252-0088 Fax: + 1 (617) 252-0004 E-mail: Robert_La_Ferla@hot.com In article <30DB65BE.7D0@ix.netcom.com> "Jonathan W. Hendry" <steeldrv@ix.netcom.com> writes: Xref: world comp.sys.next.advocacy:37862 comp.sys.next.bugs:3615 Path: world!news.kei.com!newsfeed.internetmci.com!swrinde!howland.reston.ans.net!ix.netcom.com!netnews From: "Jonathan W. Hendry" <steeldrv@ix.netcom.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.bugs Date: Fri, 22 Dec 1995 21:13:18 -0500 Organization: Steel Driving Software, Inc. Lines: 24 References: <Pine.NXT.3.92.951221195955.1077C-100000@Tomobiki-Cho.CAC.Washington.EDU> NNTP-Posting-Host: ix-cin1-21.ix.netcom.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-NETCOM-Date: Fri Dec 22 6:15:27 PM PST 1995 X-Mailer: Mozilla 2.0b3 (Win95; I) Mark Crispin wrote: >So, if you were thinking about using a CD-ROM changer with NEXTSTEP, >remember that you're limited to 7 spindles including your internal hard >disk.You'll have to use a real operating system from a real vendor to > be able to make full use of the changer. Hmm. SCSI is (or was) limited to 7 devices on a chain. Perhaps the individual CD's are being treated as separate SCSI devices, instead of the changer being treated as one device on the chain? I'm surprised the CD's aren't treated as removable. When you start up with the changer empty, how many CD's can you insert? Perhaps there's a difference in how they are mounted which depends on whether it's at bootup or later. - Jon -- Jonathan W. Hendry Steel Driving Software, Inc. Delphi and NeXTSTEP consulting and software development. Cincinnati, Ohio
From: "Timothy J. Luoma" <luomat@capitalist.princeton.edu> Newsgroups: comp.sys.next.bugs Subject: GNU uudecode and NeXTMail Date: Tue, 26 Dec 1995 11:37:19 -0500 Organization: Princeton University Message-ID: <Pine.NXT.3.91.951226112859.11078A-100000@capitalist.princeton.edu> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I dunno if this is really where to post this or what, but anyway: running the GNU version of 'uudecode' at the commandline to decode an ASCII message which was uuencode by NeXT's Mail.app usually (in my experience) produces an error "No end line" even when an 'end' line is demonstably present at the end of the message. HOWEVER, it appears that GNU uudecode _will_ properly decode the message anyway, but the error message might cause some to wonder. To avoid the error message, use '/usr/bin/uudecode' rather than the GNU version of uudecode. I'd just like to reiterate my personal belief that we still need a commandline version of Mail.app will can read/decode NeXTMail and MIME mail, use read-receipts, save to NeXTMailboxes (even nested mailboxes, as possible with Mail.app 3.3). I'd be more than willing to beta-test the hell out of a VT-100 binary which would do so... heck I'd write it if I could program, but I can't, so I just pass it on to others in the hope that someone has some Christmas cheer and Christmas vacation left which they might put to use on this project... ok, I'll stop before my rambling gets bigger than my actual post... TjL -- Timothy J. Luoma <luomat@capitalist.princeton.edu> "I'm deperate, I've got a glass and I can't find a coaster... Then a thought comes to mind... I search frantically through the mail and there it is... another @$%#&! AOL disk.... Finally a place to put down my drink."
From: ray@mayo.edu (Ray Ghanbari) Newsgroups: comp.sys.next.bugs Subject: Re: Workspace dies when accessing NFS mounted filesystems... Date: 26 Dec 1995 22:27:39 GMT Organization: Mayo Foundation Distribution: world Message-ID: <4bpssr$t8f@tribune.mayo.edu> References: <9512211256.AA27598@flexus> In article <9512211256.AA27598@flexus> Raf Schietekat <flexus!rfschtkt@maze.ruca.ua.ac.be> writes: > Paul Lynch <Paul_Lynch@plsys.com> writes: > > It's a feature; I've seen this on all versions of NeXTSTEP > > that I use. > You *must* be kidding! > > This is about KBNS.33.1.010, which was hitherto reported for 3.2 only. First time I saw it was the 3.0 beta (yes we reported it) You see this one mentioned in a lot of release notes (I think the 3.3 release notes lists it, but I would have to check) Anyway. must be some really funky stuff happening behind the scenes to keep a crasher like this one around. -- Ray Ghanbari Mayo Foundation ray@mayo.edu
From: Alvin Austin <alvin@cs.usask.ca> Newsgroups: comp.sys.next.bugs Subject: Using Samba smbd on NS3.2 White/Black causes fsck on reboot Date: 27 Dec 1995 00:49:48 GMT Organization: University of Saskatchewan, Saskatoon, CANADA Sender: alvin@cs.usask.ca (Alvin Austin) Distribution: world Message-ID: <4bq57c$kbd@tribune.usask.ca> Keywords: samba fsck Originator: alvin@after I have observed that the Samba SMB software (which compiles and runs great on NeXT systems) seems to have a little glitch somewhere... If you link to a shared directory or printer on the NeXT from an SMB client (ie: Windows for Workgroups), it will work fine, but some resource on the NeXT does not seem to be released when the share is deleted from the client. The instance of the 'smbd' daemon quits (according to 'ps'), but the NeXT will force an fsck of the root disk on the next reboot. I don't see this behavior on other unix platforms running Samba. Does anyone have any ideas on what might be causing this? I've observed this happening with many different revs of Samba, including the current samba-1.9.15p4. It's not such an issue for NeXT machines that are on continuously, but it is a pain on my home cube which gets turned off. I have received confirmation from a few other NeXT users that this happens with both black/white NS 3.2 systems. Any suggestions for solving this would be greatly appreciated. Thanks for any info... Alvin --- Alvin Austin (alvin@cs.usask.ca), Department of Computer Science University of Saskatchewan, Saskatoon, CANADA (NeXTmail OK!)
From: ccwf@acquine.klab.caltech.edu (Charles Fu) Newsgroups: comp.sys.next.bugs Subject: Re: GNU uudecode and NeXTMail Date: 28 Dec 1995 15:25:36 GMT Organization: California Institute of Technology, Pasadena Message-ID: <4buctg$r3o@gap.cco.caltech.edu> References: <Pine.NXT.3.91.951226112859.11078A-100000@capitalist.princeton.edu> In article <Pine.NXT.3.91.951226112859.11078A-100000@capitalist.princeton.edu>, Timothy J. Luoma <luomat@capitalist.princeton.edu> wrote: > running the GNU version of 'uudecode' at the commandline to decode an > ASCII message which was uuencode by NeXT's Mail.app usually (in my > experience) produces an error "No end line" even when an 'end' line is > demonstably present at the end of the message. I looked into this and a related problem at one point. The fundamental problem is that NeXT's uuencode encodes slightly improperly--it uses a "`" whereever a space " " is supposed to go in the encoded file. (The format is described in uuencode(5)). In particular, the line before the end line is supposed to consist of a single space, but NeXT uuencode produces a single "`" instead. The first character of each line indicates how many significant characters are encoded in that line, so the incorrect character causes GNU uuencode to think the "end" on the next line is actually part of the previous line's encoded text. NeXT uudecode and some other uudecode's look for and work around this bug. Of course, if NeXT uuencode did follow the standard, then it would likely run into Mail.app's problem of munging spaces at the end of lines (which is why uuencoded or btoa'ed files sent to you from non-NeXT machines are sometimes garbled if you read them with Mail.app). Perhaps that's why they have not fixed the bug. -ccwf
From: doyle@zeke.lanl.gov Newsgroups: comp.sys.next.bugs Subject: Re: GNU uudecode and NeXTMail Date: 28 Dec 1995 18:10:50 GMT Organization: Los Alamos National Laboratory Message-ID: <4bumja$e7n@newshost.lanl.gov> References: <Pine.NXT.3.91.951226112859.11078A-100000@capitalist.princeton.edu> <4buctg$r3o@gap.cco.caltech.edu> ccwf@acquine.klab.caltech.edu (Charles Fu) wrote: > I looked into this and a related problem at one point. The fundamental > problem is that NeXT's uuencode encodes slightly improperly--it uses a > "`" whereever a space " " is supposed to go in the encoded file. (The > format is described in uuencode(5)). In particular, the line before the > end line is supposed to consist of a single space, but NeXT uuencode > produces a single "`" instead. The first character of each line > indicates how many significant characters are encoded in that line, so > the incorrect character causes GNU uuencode to think the "end" on the > next line is actually part of the previous line's encoded text. NeXT > uudecode and some other uudecode's look for and work around this bug. Except that all modern uuencodes do this precisely because there are lots of mailers that strip trailing spaces corrupting uuencoded files. The backquote and the space are supposed to be rigorously interchangeable. It is not a bug to use backquotes instead of spaces. It is a bug not to recoginze the backquotes as the same as spaces. (I am just speaking from experience here; we receive lots of uuencoded files that have been corrupted by many mailers that strip trailing spaces. I don't know if there is some RFC around detailing all of this). > Of course, if NeXT uuencode did follow the standard, then it would > likely run into Mail.app's problem of munging spaces at the end of > lines (which is why uuencoded or btoa'ed files sent to you from > non-NeXT machines are sometimes garbled if you read them with > Mail.app). Perhaps that's why they have not fixed the bug. Sounds more like a bug for GNU to fix in their uudecode. (NeXT's munging of spaces in Mail.app is another issue --- seems better in 3.3 but still not quite right). Cheers, Mark
From: crath@bnr.ca (Christopher Rath) Newsgroups: comp.sys.next.bugs Subject: Feature request for GNU cpp Date: 28 Dec 1995 22:21:22 GMT Organization: Bell Northern Research, Ottawa, Canada Message-ID: <CRATH.95Dec28172122@bmerhe83.bnr.ca> /***** ***** ** ** This code demonstrates cpp behaviour that I would like to see ** cpp emit a warning message for. When a #define over-rides an ** enum, I would like cpp to warn me about what is taking place. ** This should only be enabled by specific user request, or by ** ``gcc -Wall''. ** ** This code appears to compile alright, but it will never ** execute as I had intended (because the enum's value is 0 and ** the #define's value is 1). Unfortunately, this problem is ** hard to find when it is one of the system header files that is ** over-riding an enum in my own code. ** ***** *****/ typedef enum { OK, NOTOK } okType; #define OK (1) okType main() { okType imok = OK; /* lots of code that may conditionally set imok to NOTOK */ if (imok == NOTOK) { printf("Premature return\n."); return imok; } else { printf("Returned properly\n."); } return imok; } /***** Here's what ``gcc -E'' emits for this code (with empty lines removed): ***** #1 "hassle.c" typedef enum { OK , NOTOK } okType ; okType main ( ) { okType imok = ( 1 ) ; if ( imok == NOTOK ) { printf ( "Premature return\n." ) ; return imok ; } else { printf ( "Returned properly\n." ) ; } return imok ; } ***** end of ``gcc -E'' output *****/ -- === Christopher Rath ===== crath@bnr.ca ===== (613) 765-3141 === Bell-Northern Research | Box 3511, Station `C' | ``Hydrogen is a colourless, odourless Ottawa, ON K1Y 4H7 | gas which, given enough time, turns FAX: (613) 763-4101 | into people.'' --- Henry Hiebert
Newsgroups: comp.sys.next.bugs From: wave@media.mit.edu (Michael B. Johnson) Subject: NEXTSTEP's Mac HFS filesystem suport is ridiculous... Message-ID: <1995Dec29.051859.10666@media.mit.edu> Sender: news@media.mit.edu (USENET News System) Organization: Pixar Animation Studios (East Coast Office) Date: Fri, 29 Dec 1995 05:18:59 GMT Having just trashed a Mac gig drive for the nth time under NEXTSTEP by doing nothing more innocuous than dragging a couple of folders on to it, I'm really sick of this... I've avoided writing up a bug report on this because it always happens when I'm frantic, and I don't have the time to deal with it properly, and then when I'm done, I'm too annoyed to go back to it. But since we just passed 100M BO and Steve gave us the week off, I've got some time on my hands... :-) For years I've been happily used NEXTSTEP's support for Mac's file system. Usually I only copied the odd file to a Mac floppy (usually a TIFF file or EPS or FreeHand/Virtuoso file), and while every once in awhile there was a glitch, it usually was no biggie. But lately, I've hooked up bigger drives (200MB, 400MB, 1GB) that have been formatted on a Mac and tried to copy large amounts of files from NEXTSTEP to a Mac formatted disk; I get an error ("invalid argument"). From that moment on, while I can see the disk under NEXTSTEP, a Mac won't touch it. I don't have a lot of Mac disk repair programs, but the few that I have can't repair it, and I'm forced to reformat the disk on the Mac side, losing all the data on the disk. The reason I need to do this is (among other things) I have a Mac-based CD-ROM burner, and I really want to burn CDs of things I'm doing under NEXTSTEP, but I can't get the damn files over there. The real bug is that most of my filesystems are very deep; too deep to make a single huge tar file under NEXTSTEP and copy it to the Mac and then untar it on the Mac side. I have to use gnutar for the directory depth, and while there are plenty of "tar"s on the Mac that can understand UNIX tar format, I have yet to find a gnutar. Please note: this happens on my newly patched 3.3 Intel machine; it certainly happened on my pre-patch 3.3 machine, and I'm reasonably certain this happens on my black hardware. I don't remember if it happened in 3.2. As you can imagine, this a rather annoying bug to reproduce... What I'm looking for is other Mac filesystem war stories, so I know how to formulate a coherent bug report for NeXT. I'm not sure if it's file size, file name length, directory depth, thefact that I'm the one logged in :-) or what. Any data appreciated; I'll try to formulate it into a coherent bug report. Better yet would be a work around, but I'm not optimistic. -- --> Michael B. Johnson, SMVS, Ph.D. -- wave@media.mit.edu|wave@pixar.com --> http://wave.www.media.mit.edu/people/wave/ --> alumnus, MIT Media Lab, Computer Graphics & Animation Group --> Media Arts Technologist, Pixar Animation Studios (East Coast Office)
From: tikin@hasc.ca (Martin Lavertu) Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs Subject: Distributed Object, NS 3.2 & 3.3 Date: 28 Dec 1995 21:19:36 GMT Organization: School of Computer Science, McGill U, Montreal Message-ID: <4bv1l8$f21@lisa.cs.mcgill.ca> Hi, Have you experienced problems using distributed objects under 3.2 and/or 3.3 or porting your applications from 3.2 to 3.3 that makes use of DOs? If so, please let me know. I have experienced a few problems with them lately and I'd like to know if I'm the only having problems. They are essentially related to the new feature in NS 3.3 where connections are closed if they are idle for more than 2.5 minutes. I have also some workarounds. Thanks, -- Martin Lavertu martin@hasc.ca
Newsgroups: comp.sys.next.bugs From: belka@il.us.swissbank.com (Andy Belk) Subject: Re: NEXTSTEP's Mac HFS filesystem suport is ridiculous... Message-ID: <1995Dec29.182850.3814@il.us.swissbank.com> Sender: root@il.us.swissbank.com (Operator) Organization: Swiss Bank Corporation CM&T Division References: <1995Dec29.051859.10666@media.mit.edu> Date: Fri, 29 Dec 1995 18:28:50 GMT In article <1995Dec29.051859.10666@media.mit.edu> wave@media.mit.edu (Michael B. Johnson) writes: < Problems with a Mac Hard disk deleted > I recently plugged my Mac Zip drive into a NeXT and happily copied over 80 MB of directories (untarred). The Zip disk appeared on my shelf as soon as I stuck it into the drive. Considering the Zip drive wasn't invented when the NeXT was built, I was quite impressed ... It did have problems copying _very_ long filenames (the Mac can't handle them). Sorry, this isn't very helpful ... but just to assure you it can be done with reasonable amounts of data. Andy
Newsgroups: comp.sys.next.bugs From: fozztexx@nvc.cc.ca.us (Chris Osborn) Subject: Re: Using Samba smbd on NS3.2 White/Black causes fsck on reboot Message-ID: <DKD5I6.1Go@nvc.cc.ca.us> Keywords: samba fsck Sender: news@nvc.cc.ca.us Organization: Napa Valley College References: <4bq57c$kbd@tribune.usask.ca> Date: Fri, 29 Dec 1995 19:44:30 GMT In article <4bq57c$kbd@tribune.usask.ca>, Alvin Austin <alvin@cs.usask.ca> wrote: >If you link to a shared directory or printer on the NeXT >from an SMB client (ie: Windows for Workgroups), it will work >fine, but some resource on the NeXT does not seem to be >released when the share is deleted from the client. The >instance of the 'smbd' daemon quits (according to 'ps'), >but the NeXT will force an fsck of the root disk on the >next reboot. I don't see this behavior on other unix >platforms running Samba. > >I have received confirmation from a few other NeXT users >that this happens with both black/white NS 3.2 systems. I'm also experiencing the same problems under 3.3 black. It's become quite a nuisance, so for the moment I've disabled Samba. -- Chris Osborn, Network Administrator Voice: 707 253 3130 Napa Valley College Fax: 707 253 3063 2277 Napa-Vallejo Hwy., Napa, CA, 94558 <fozztexx@nvc.cc.ca.us> MIME ok, NeXTMail tolerated
From: dave@turbocat.snafu.de (David Wetzel) Newsgroups: comp.sys.next.bugs Subject: No manual entry for machine Date: 29 Dec 1995 18:03:54 GMT Organization: Turbocat's Development, Germany Message-ID: <4c1aia$1kc@turbocat.snafu.de> dave@turbocat> man machine No manual entry for machine. The NEXTSTEP has no man page for machine. Here is the NetBSD page: dave@amiga> man machine MACHINE(1) NetBSD Reference Manual MACHINE(1) NAME machine - print machine type SYNOPSIS machine DESCRIPTION The machine command displays the machine type. SEE ALSO make(1) HISTORY The machine command is currently under development. BSD Experimental November 24, 1995 1 _ _ _(_)(_)_ David Wetzel, Turbocat's Development, (_) __ (_) Buchhorster Strasse, D-16567 Muehlenbeck/Berlin, FRG, _/ \_ Phone +49 33056 82151, Fax +49 33056 82152 (______) dave@turbocat.snafu.de (NeXTMail)
These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Marcel Waldvogel and Netfuture.ch.