This is Bugs93-I.gz in view mode; [Up]
Newsgroups: comp.sys.next.bugs From: Zacharias J. Beckman <zac@dolphin.com> Subject: HELP: Crash in 3.0 results in failure to boot Message-ID: <1993Jan7.165156.9501@dolphin.com> Sender: zac@dolphin.com Organization: Dolphin Software Distribution: usa Date: Thu, 7 Jan 1993 16:51:56 GMT Hello, A few days ago, our server had a horrible crash. Everything on the network locked up tight and needed to be restarted the hard way. When the server came back up, it refused to boot in multi-user mode. The boot parameter of "-nbu=128" was being ignored, and the default 16 buffers were used. When the system started to boot Mach, it declared that it couldn't write to mtab because the file system was read only!!! The only way around this at present is to boot in single-user mode, then continue the multi-user startup. Has anyone run into this sort of problem before, or have any idea where I an look to cure it? It may not be a 3.0 bug, but on the other hand, perhaps it is... Thanks in advance. -- Zacharias J. Beckman - Dolphin Software Inc. - zac@dolphin.com - use NeXTMAIL! To be "matter of fact" about the world is to blunder into fantasy.... and dull fantasy at that, as the real world is strange and wonderful. --- R. A. Heinlen Those opinions I express herein are my own, I'm fairly sure. --- Z. J. Beckman
Newsgroups: comp.sys.next.bugs From: Zacharias J. Beckman <zac@dolphin.com> Subject: HELP: Bug in FAX LATER option... Message-ID: <1993Jan7.165404.9562@dolphin.com> Sender: zac@dolphin.com Organization: Dolphin Software Distribution: usa Date: Thu, 7 Jan 1993 16:54:04 GMT Hello, Ever since installing 3.0 (final release as opposed to pre-release) the FAX LATER option of the fax panel has not worked. Trying to use it results in an immediate fax of whatever is going out. I've heard from a few other people that have this same problem, and have asked me to summarize if I find a solution. If you know how to fix this, I would appreciate a reply by email. Thanks in advance! -- Zacharias J. Beckman - Dolphin Software Inc. - zac@dolphin.com - use NeXTMAIL! To be "matter of fact" about the world is to blunder into fantasy.... and dull fantasy at that, as the real world is strange and wonderful. --- R. A. Heinlen Those opinions I express herein are my own, I'm fairly sure. --- Z. J. Beckman
Newsgroups: comp.sys.next.bugs From: grdetil@scrc.umanitoba.ca (Gilles Detillieux) Subject: Re: ownership of /bin/df and /etc/mtab Message-ID: <C0I2qE.Gn2@ccu.umanitoba.ca> Sender: news@ccu.umanitoba.ca Organization: University of Manitoba, Winnipeg, Canada References: <1993Jan6.064038.6744@proximus.north.de> Date: Thu, 7 Jan 1993 20:11:01 GMT In article <1993Jan6.064038.6744@proximus.north.de> gemoe@proximus.north.de (Gerhard Moeller) writes: > Well, maybe it's my fault, but I wonder why the permissions of mtab and df > are not right set, since 3.0: > > -rwxr-sr-x 1 root operator 6692 Jul 21 17:40 /bin/df* > -rw-r----- 1 root wheel 81 Jan 5 15:44 /etc/mtab > > With permissions like that only root and wheel are able to execute df. > (Very bad for news, as rnews needs df.) > > The question: Is is my fault and which group is wrong set? Well, without laying any blame, /etc/mtab usually is world-readable. (i.e. mode 664 or 644) The owner and group assignments are the same as on my NeXT. -- Gilles Detillieux <Gilles@scrc.UManitoba.CA> Spinal Cord Research Centre or <grdetil@gilles.scrc.UManitoba.CA> Dept. of Physiology, U. of Manitoba Phone: (204)788-6766 Winnipeg, MB R3E 0W3 (Canada) Fax: (204)786-0932
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin From: jchin@prodigy.bc.ca (Joseph Chin) Subject: sendmail problem Message-ID: <1993Jan7.222028.4493@prodigy.bc.ca> Sender: news@prodigy.bc.ca Organization: Prodigy Technologies Corp. Date: Thu, 7 Jan 1993 22:20:28 GMT I've just noticed that since I upgraded to 3.0, I can no longer send email to accounts within my domain (prodigy.bc.ca). Is there a bug in the 3.0 sendmail.cf file? The NeXT is a mail client on the network, and the mail host is a Sun Sparc server. Everything worked fine before the 3.0 upgrade. Any ideas? Thanks. Joe -- Joseph Chin Prodigy Technologies Corp. jchin@prodigy.bc.ca (NeXTmail welcome) Vancouver, British Columbia phone 16046874636 / fax 16046871671 Canada **** "Heaven won't take me and Hell's afraid I'll take over!" ****
Newsgroups: comp.sys.next.bugs From: rpruess@yoyo.weeg.uiowa.edu (Rex Pruess) Subject: Re: ownership of /bin/df and /etc/mtab In-Reply-To: gemoe@proximus.north.de's message of Wed, 6 Jan 1993 06:40:38 GMT Message-ID: <RPRUESS.93Jan7160658@yoyo.weeg.uiowa.edu> Sender: news@news.weeg.uiowa.edu (News) Organization: U of Iowa, Iowa City, IA References: <1993Jan6.064038.6744@proximus.north.de> Date: Thu, 7 Jan 1993 22:06:58 GMT gemoe@proximus.north.de (Gerhard Moeller) writes: > Well, maybe it's my fault, but I wonder why the permissions of mtab > and df are not right set, since 3.0: > > -rwxr-sr-x 1 root operator 6692 Jul 21 17:40 /bin/df* > -rw-r----- 1 root wheel 81 Jan 5 15:44 /etc/mtab Ejecting disks messes up the correct setting of "644" on /etc/mtab. One way to deal with this problem is to update the mode setting every 30 minutes or so by putting the following line in /etc/crontab.local: 1,31 * * * * root /bin/chmod 644 /etc/mtab -- ---------------------------------------------------------------------------- Rex Pruess, Weeg Computing Center, University of Iowa, Iowa City, IA NeXTmail: Rex-Pruess@uiowa.edu ......... Phone: (319) 335-5452
From: matthews@oberon.umd.edu (Mike Matthews) Newsgroups: comp.sys.next.bugs Subject: niload "features" Message-ID: <17841@umd5.umd.edu> Date: 8 Jan 93 01:05:41 GMT Sender: news@umd5.umd.edu Organization: University of Maryland, College Park, MD In NeXTstep 3.0, each user gets an 'info' NetInfo directory added to their usual /users/<name> directory. This means you can't just niutil -destroy <domain> /users/<name> to get rid of the user because NetInfo can't destroy directories with children. Well, niload -d passwd is supposed to delete any password entries that are not in the file you load up. But it can't because of the info directory. niload should know how to delete that info directory (what *IS* it used for, anyway?). This also adds a step to automatic user deletion (niutil -destroy <domain> /users/<user>/info, which may not be all that easy depending on how you do the batch deletes). Also, didn't niload hosts bind to the IP address before? Now it seems to bind to the hostname, so if you dump the hosts to a flat file, change their name, and reload it, you have two hosts with the same IP address (doesn't *hurt*, but it is annoying). At least niload -d now overwrites any existing data, and tries to delete information not in the new file as opposed to blowing away that directory and reloading it (as I think it did before, anyone care to verify or disprove that?). ------ Mike Matthews, matthews@oberon.umd.edu (NeXTmail accepted) ------ ...difference of opinion is advantageious in religion. The several sects perform the office of a common censor morum over each other. Is uniformity attainable? Millions of innocent men, women, and children, since the introduction of Christianity, have been burnt, tortured, fined, imprisoned; yet we have not advanced one inch towards uniformity. - Thomas Jefferson, "Notes on Virginia"
Organization: Queen's University at Kingston Date: Fri, 8 Jan 1993 09:41:21 EST From: <HARRAPR@QUCDN.QueensU.CA> Message-ID: <93008.094121HARRAPR@QUCDN.QueensU.CA> Newsgroups: comp.sys.next.bugs Subject: PLI Floppy Drive under 3.0 - Help... I've been using 3.0 on a machine with a PLI floppy drive, the standard 2.8 meg one. It reads DOS disks just fine, but won't write to them.... When I try to write, a message comes up in the Process panel saying 'Invalid Argument' and the task terminates, regardless of whether I press Proceed or Stop. What gives? I'll summarize if there are answers. Rob Harrap Dept. Geological Sciences Queen's University Kingston, Ont. Canada Harrapr@qucdn.queensu.ca
Newsgroups: comp.sys.next.bugs From: eoliver@ralph.cs.haverford.edu (Erik Oliver) Subject: Re: niload "features" Message-ID: <TYKWBN1T@cc.swarthmore.edu> Sender: news@cc.swarthmore.edu (USENET News System) Organization: Haverford College Computer Science Department References: <17841@umd5.umd.edu> Date: Fri, 8 Jan 1993 15:41:11 GMT With regard to account deletion: USE nu -d, it works. Also, I have not encountered the problems you describe with the host.equiv files in netinfo. -Erik
From: madler@cco.caltech.edu (Mark Adler) Newsgroups: comp.sys.next.bugs Subject: Installer.app command-line launched misses floppy change Date: 8 Jan 1993 16:33:31 GMT Organization: California Institute of Technology, Pasadena Message-ID: <1ikacrINN3g2@gap.caltech.edu> On 3.0, if I launch Installer.app from a Terminal window su'ed to root, it does not think that I have changed the floppy when in fact I have (and put in the right one). If I log out and log in on the console as root, Installer.app works fine. I've tried this with systems that have both and internal and an external floppy. This is definitely a pain, since I like to be able to stay logged in to my normal account and still do the occasional system administration things, like installing new software. For every other sysadmin thing I've done, that is possible. Mark Adler madler@cco.caltech.edu
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin From: kinch@valve.heart.rri.uwo.ca (Dave Kinchlea) Subject: What does NeXTMail do during `undelete' command? Organization: Heart Valve Group, John P. Robarts Research Institiute, London, Ont. Canada Date: Fri, 8 Jan 1993 18:00:37 GMT Message-ID: <1993Jan8.180037.23392@julian.uwo.ca> Followup-To: comp.sys.next.sysadmin Sender: news@julian.uwo.ca (USENET News System) Howdy I am having a problem with NeXTMail. My users' directories are NFS mounted onto a Personal Iris but the /usr/spool/mail is local. When one of my users issues a `Undelete' command from inside of NeXTMail, my PI disk fills up AND the load on the PI machine continues to grow until there is nothing left to do but pull the plug. In trying to determine what the problem is it seems necessary to determine what NeXTMail is up to during this `Undelete' operation. Could anyone in the know please tell me. I assumed that it would simply make a copy of mbox sans any deleted messages but that simply does not seem to be the case. I can provide more info to any who want but understanding the problem isn't necessary in order to explain what NeXTMail is *supposed* to do with this command. cheerskinch -- Dave Kinchlea email: kinch@valve.heart.rri.uwo.ca Snail-mail: System Administrator/Research Programmer John P. Robarts Research Institute. P.O. Box 5015, 100 Perth Drive, London, Ont, Canada N6A 5K8 A Real Live Voice: (519)-663-5777-x4406 ".... 2 to the power 759 million to 1 and falling ..." HHGTG
Newsgroups: comp.sys.next.bugs From: David.Kelman@launchpad.unc.edu (David Kelman) Subject: DL and indexing third party manuals Message-ID: <1993Jan8.183532.9368@samba.oit.unc.edu> Sender: usenet@samba.oit.unc.edu Organization: University of North Carolina Extended Bulletin Board Service Date: Fri, 8 Jan 1993 18:35:32 GMT I encountered a problem with Digital Librarian indexing the reference manual for WordPerfect. With the manuals provided with 3.0 (man pages, etc.), you can search either the title or contents. DL indexed the WP manual, but would not search be contents; it could only find a citation if the word was in the title of the file. This is really annoying, as even searching for 'rtf' turns up nothing anywhere in the WP manual. Has anyone had a similar experience? David Kelman -- The opinions expressed are not necessarily those of the University of North Carolina at Chapel Hill, the Campus Office for Information Technology, or the Experimental Bulletin Board Service. internet: laUNChpad.unc.edu or 152.2.22.80
Newsgroups: comp.sys.next.bugs From: noyau@pelane.cubx.com (Eric Noyau) Subject: Re: ownership of /bin/df and /etc/mtab Message-ID: <1993Jan8.082445.8580@pelane.cubx.com> Sender: noyau@pelane.cubx.com Organization: Cub'X systemes References: <C0I2qE.Gn2@ccu.umanitoba.ca> Date: Fri, 8 Jan 1993 08:24:45 GMT In article <C0I2qE.Gn2@ccu.umanitoba.ca> grdetil@scrc.umanitoba.ca (Gilles Detillieux) writes: > In article <1993Jan6.064038.6744@proximus.north.de> gemoe@proximus.north.de > (Gerhard Moeller) writes: > > Well, maybe it's my fault, but I wonder why the permissions of mtab and df > > are not right set, since 3.0: > > > > -rwxr-sr-x 1 root operator 6692 Jul 21 17:40 /bin/df* > > -rw-r----- 1 root wheel 81 Jan 5 15:44 /etc/mtab > > > > With permissions like that only root and wheel are able to execute df. > > (Very bad for news, as rnews needs df.) > > > > The question: Is is my fault and which group is wrong set? > > Well, without laying any blame, /etc/mtab usually is world-readable. > (i.e. mode 664 or 644) > > The owner and group assignments are the same as on my NeXT. > Workspace do "chmod 640 /etc/mtab" sometimes...(when you insert/eject a disk?) fix : modify your crontab.local and add [...] 02,17,32,47 * * * * root /bin/chmod 644 /etc/mtab [...] It's ugly... Eric -- (Eric Noyau - e_noyau@cubx.com - (Small)NeXTMail Ok) /R{rand 2147483647 div mul}def 1 1 90{pop dup gsave/Helvetica findfont 40 R scalefont setfont 500 R 800 R moveto 360 R rotate 1 R setgray show grestore}for pop showpage % A bug can be changed to a feature by documenting it. Developpers know! %
Newsgroups: comp.sys.next.bugs From: tilley@ccu.umanitoba.ca (Richard Tilley) Subject: Re: HELP: Crash in 3.0 results in failure to boot Message-ID: <C0Kxow.98H@ccu.umanitoba.ca> Sender: news@ccu.umanitoba.ca Organization: University of Manitoba, Winnipeg, Canada References: <1993Jan7.165156.9501@dolphin.com> Date: Sat, 9 Jan 1993 09:14:56 GMT In <1993Jan7.165156.9501@dolphin.com> Zacharias J. Beckman <zac@dolphin.com> writes: >A few days ago, our server had a horrible crash. Everything on the network >locked up tight and needed to be restarted the hard way. When the server >came back up, it refused to boot in multi-user mode. The boot parameter of >"-nbu=128" was being ignored, and the default 16 buffers were used. Guess 1: You need a blank between the "-" and the "nbu=128". Guess 2: Use the "p" command in the ROM monitor to look at your settings. If they are all wrong, try using the "p" command to change one of them. Now use "p" to list them again. If the change was ignored, then cycle the power. Are the settings now all correct? -- This space is not blank!
From: dimitri@dutiws.twi.tudelft.nl (Dimitri Tischenko) Newsgroups: comp.sys.next.bugs Subject: mach IO error -- solved Keywords: summary Message-ID: <C0LA1u.BJL@dutiws.twi.tudelft.nl> Date: 9 Jan 93 13:41:53 GMT References: <C0FELB.KA3@dutiws.twi.tudelft.nl> Organization: Delft University of Technology The reason I got this message: Jan 4 14:14:11 dutiea mach: IO error on pagein: error = 22. was the fact that there was no room for the swapfile to grow. Thanks to Scott Hess, Rob Kedoin and Mike Hovan. Dimitri +----------------------------------------------------------------------------+ | Dimitri Tischenko | D.B.Tischenko@IS.TWI.TUDelft.NL | NeXTmail preferred! | +---------------------------------------+------------------------------------+ | Delft University of Technology | Technische Universiteit Delft | | Fac Applied Math & Computer Science | Fac. Techn. Wiskunde & Informatica | | The Netherlands | Nederland | +---------------------------------------+------------------------------------+
From: glenn@rightbrain.com (Glenn Reid) Newsgroups: comp.sys.next.bugs Subject: Re: My old Font-Problem with NS3.0 Message-ID: <1058@rtbrain.rightbrain.com> Date: 10 Jan 93 01:11:23 GMT References: <981@rtbrain.rightbrain.com> Sender: glenn@rightbrain.com Borris Balzer writes > > Hi there, > > posting to other groups didn't bring any hints to solve my font-problem in an > acteptable way (BTW: Thanks to all who answered). > > The problem I've got: > > NS2.1 installed, I used without problems fonts with the following names: > - Garamond (the ITC Garamond) > - AdobeGaramond > - StempelGaramond > - Garamond Three > After installing NS3.0 the AdobeGaramond and StempelGaramond disappeared from > the font panel. I posted a response to this. Maybe it didn't make it out. There is a known bug in NS 3.0 related to the Garamonds, and you have hit it squarely on the head. I have been in correspondence with NeXT about it, and a fix should make it into the next release, whenever that may be. In the interim, your only choice is to edit the AFM files for fonts that are disappearing to change the FamilyName property to add some extra characters, so it won't sort into the same pile as the other Garamonds. This won't affect anything except the way the fonts are displayed in the font panel. -- Glenn Reid NeXTmail: glenn@rightbrain.com RightBrain Software 415-326-2974 (NeXTfax 326-2977) Palo Alto, California Electronic Frontier Foundation, member #054
Newsgroups: comp.sys.next.bugs From: CCGREG@mizzou1.missouri.edu (Greg Johnson) Subject: Re: HELP: Crash in 3.0 results in failure to boot Message-ID: <16B52EE0A.CCGREG@mizzou1.missouri.edu> Sender: news@mont.cs.missouri.edu Organization: University of Missouri References: <1993Jan7.165156.9501@dolphin.com> Date: Sun, 10 Jan 93 16:55:19 CST In article <1993Jan7.165156.9501@dolphin.com> Zacharias J. Beckman <zac@dolphin.com> writes: >When the system started to boot Mach, it declared that it couldn't write to >mtab because the file system was read only!!! I found a colleague with a read-only root not long after doing a 3.0 install. We finally discovered that he copied from the CD-ROM /NeXTCD or some such file where the extended version packages are. If a file with that name is on /, the new /etc/rc mounts the file-system read-only!
Newsgroups: comp.sys.next.bugs From: CCGREG@mizzou1.missouri.edu (Greg Johnson) Subject: Re: niload "features" Message-ID: <16B52F542.CCGREG@mizzou1.missouri.edu> Sender: news@mont.cs.missouri.edu Organization: University of Missouri References: <17841@umd5.umd.edu> Date: Sun, 10 Jan 93 17:26:22 CST In article <17841@umd5.umd.edu> matthews@oberon.umd.edu (Mike Matthews) writes: >Well, niload -d passwd is supposed to delete any password entries that are >not in the file you load up. But it can't because of the info directory. >niload should know how to delete that info directory (what *IS* it used for, >anyway?). Right. I can appreciate the safety feature, but for the next increment of niload, NeXT, would you add a "-f" option to force deletion even if there are kids? Until then, I'm inclined to delete all the "info" subdirectories with niutil. It'll make the database smaller. They'd be easy to recreate. Does anyone know if "info" is actually used by anything? The 3.0 Netinfo Release Notes describes it as a miscellaneous repository; the _password entry preloaded there when an ID is created just allows the user to update this portion of their NetInfo data. It might be used for .plan info, or maybe you could put your resume in it.
Newsgroups: comp.sys.next.bugs From: CCGREG@mizzou1.missouri.edu (Greg Johnson) Subject: niload exports bombs if there are # comments in the exports Message-ID: <16B52F640.CCGREG@mizzou1.missouri.edu> Sender: news@mont.cs.missouri.edu Organization: University of Missouri Date: Sun, 10 Jan 93 17:30:40 CST Nathan Janette mentioned this problem to me. In 3.0 a bug causes niload to bomb if the exports file has any comment hashs in it. One gets a bus error, and the exports may or may not be loaded.
Newsgroups: comp.sys.next.bugs From: death@netcom.com (David Burrowes) Subject: NeXT's dealing with rtf mac charset Message-ID: <1993Jan11.055225.28258@netcom.com> Organization: Netcom - Online Communication Services (408 241-9760 guest) Distribution: usa Date: Mon, 11 Jan 1993 05:52:25 GMT Does anyone have details or explanations about this problem? I found that NS 3.0 understands the \mac charset directive in rtf documents now, and does a reasonable job of converting the characters. However, I also found that it strangely mapped the Mac's open and close single printer's quotes, not to proper curley quotes on the NeXT, but instead to the accent and neuter quote. The end result is that if you used normal single quotes on the Mac, you get curley ones on the NeXT, and if you use curley ones on the Mac, you get the strange mac-normal ones on the NeXT. What am I misunderstanding? \david john burrowes death@kira.net.netcom.com
Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs From: ben@fizz.fdn.org (Benoit Grange) Subject: Strange things in make depend Message-ID: <1993Jan10.144622.1447@fizz.fdn.org> Sender: ben@fizz.fdn.org (Benoit Grange) Date: Sun, 10 Jan 1993 14:46:22 GMT If you issue a make depend command in a project, the file Makefile.dependencies contains references of *ALL* the appkit headers for each object file. This produces HUGES dependencies list (more than 100Kb), slowing down the make process. To build the dependencies make depends issues 'cc -MM <all sources> | awk...' and the cc manual says -MM Like -M but the output mentions only the user-header files included with #include "file". System header files included with #include <file> are omitted. So why are all imported <appkit/xxx.h> listed ? Is that because of #import instead of #include ? I don't care having dependencies with them. I currently use a sed script to remove all this stuff. My makes are now going somewhat faster. The first command issued appears after one second instead of something like 5 or more seconds. -- ---- Benoit Grange - PARIS - FRANCE Mail to : ben@fizz.fdn.org (SMALL NeXTMails accepted, < 10kb)
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: CCC_HART@rzmain.rz.uni-ulm.de (Hardy) Subject: NxBrowser (Bug ?) Message-ID: <1993Jan11.132437.16200@wega.rz.uni-ulm.de> Sender: news@wega.rz.uni-ulm.de (News Net) Organization: University of Ulm, Germany Date: Mon, 11 Jan 1993 13:24:37 GMT Hello world, i have a big problem with the NeXT Interface-Builder (and there with NXBrowser): When i generate a window in IB and place a NXBrowser in that window the method _setPath:(const char *)aPath_ didn't work. It returns only self and nothing happened. All other methods works fine (ex. _loadColumnZero_, after loading the browser manually (using _loadColumnZero_) even _getPath_). So what do i wrong ? Is this a (known) bug of NeXTStep 3.0 ? Any hints welcome. Please answer by e-mail (ccc_hart@rzmain.rz.uni-ulm.de) since i didn't read this group frequently. Awaiting your replies and thousand thanx (sorry for mistakes, but english isn't my native language) so long... hardy ######################################################################## ## Wiedergabe, auch ## INTERNET: ccc_hart@rzmain.rz.uni-ulm.de ## ## auszugsweise, nur ## BITNET: ccc_hart@dulruu51.bitnet ## ## mit schriftlicher ## DECnet: 50186::ccc_hart (BelWue) ## ## Genehmigung des ## PSImail: 0262450502601::rzmain::ccc_hart ## ## Authors !!! ## x.400: s=ccc_hart;ou=rz;p=uni-ulm;a=dbp;c=de ## ######################################################################## Real programmers don't comment their code. It was hard to write, it should be hard to understand. - a real programmer
Newsgroups: comp.sys.next.bugs From: jeffs@next-demo.acs.ohio-state.edu (Jeff Schluep) Subject: Re: sendmail problem Message-ID: <1993Jan11.154634.29195@magnus.acs.ohio-state.edu> Sender: news@magnus.acs.ohio-state.edu Organization: The Ohio State University References: <1993Jan7.222028.4493@prodigy.bc.ca> Date: Mon, 11 Jan 1993 15:46:34 GMT In article <1993Jan7.222028.4493@prodigy.bc.ca> jchin@prodigy.bc.ca (Joseph Chin) writes: > I've just noticed that since I upgraded to 3.0, I can no longer send email > to accounts within my domain (prodigy.bc.ca). Is there a bug in the 3.0 > sendmail.cf file? The NeXT is a mail client on the network, and the mail > host is a Sun Sparc server. Everything worked fine before the 3.0 upgrade. > > Any ideas? Thanks. > > Joe > I had a similar problem. Mail to host faxgate.acs.ohio-state.edu would not work correctly for me (my domain is acs.ohio-state.edu). When I investigated the problem I discovered that all mail sent to a host inside my domain was considered "local". What this means is that next's sendmail is configured to send local mail without the domain. Thus mail to "joe@faxgate.acs.ohio-state.edu" would be sent as "joe@faxgate". The host that handles mail for faxgate wanted to see the whole domain and so it bounced the message. I reconfigured the sendmail.mailhost.cf (this was the sendmail config that I was using) to send mail to the next's in my netinfo config as "local" and everything else as "non-local". This solved my problem. I'm not sure if my problem was the same as yours but it sounds like it. --- Jeff Schluep Email: JEFF+@osu.edu The Ohio State University 1971 Neil Avenue Phone: (614) 292-4843 Columbus, Ohio 43210 Fax: (614) 292-1268
From: baer@psssun (Brian Baer) Newsgroups: comp.sys.next.bugs Subject: HELP: Problems printing from WordPerfect under 3.0 Date: 11 Jan 1993 15:11:08 GMT Organization: Michigan State University Message-ID: <1is2mcINNchj@msuinfo.cl.msu.edu> Summary: Installed version 3.0 of NeXTstep and now WordPerfect prints black page Keywords: help wordperfect printing nextstep3.0 I just installed version 3.0 of NeXTstep and my version of WordPerfect (dated 16 Sep 1991) now prints a black page instead of my document. I can send the file to Preview and it prints fine from there. Any idea of how to fix this? Do I need to get a new version form WordPerfect? Thanks. Brian Baer baer@psssun.pss.msu.edu
From: wade@hobbes.ucsd.edu (Wade Blomgren) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: csh redirection fails when run under cron Message-ID: <1isocbINNas7@network.ucsd.edu> Date: 11 Jan 93 21:21:15 GMT Followup-To: poster Organization: University of California, San Diego After upgrading to 3.0 I am having a problem with csh scripts run from cron with user ids other than root. The scripts are run, but any redirection of output from a command WITHIN the script goes to the standard output of the script, not to the target of the redirection. For example, the following script (saved in, say, /tmp/script, mode 755): #! /bin/csh echo foo > /tmp/bar run by the following crontab entry: * * * * * me /tmp/script > /tmp/script.out 2>&1 results in the word "foo" appearing in /tmp/script.out, NOT /tmp/bar. The same is true for piped output: cat /etc/hosts | grep localhost > /tmp/bar will send the entire hosts file to the output of the script, and nothing goes through the grep or into /tmp/bar. NOTE: the output files _are created_. That is to say, in the first simple example, the file /tmp/bar will exist, but it will be empty. This behavior does not seem to happen when the cron command is run as root. When the script is run from the command line by the user account it works fine. Adding additional levels of indirection (script calling a script) seems to result in all output being lost. The problem does not occur on a self contained machine in my office but it does occur on our other machines. The only significant difference between my machine and the failing machines is the failing machines are running NIS. Can anyone else running non-root csh scripts with cron who is using NIS (or not) under 3.0 comment? Am I hallucinating? Thanks for any input - I will summarize responses. Wade Blomgren wade@hobbes.ucsd.edu
From: d1h1883@sc.tamu.edu (Dave Hess) Newsgroups: comp.sys.next.bugs Subject: gcc preprocessor and -traditional problem Date: 12 Jan 1993 00:20:13 GMT Organization: Texas A&M University, College Station, TX Distribution: world Message-ID: <1it2rtINNk7i@tamsun.tamu.edu> I was trying to put together a package (detex 2.4) that needed to be compiled with -traditional and gcc puked on some NeXT include files. gcc apparently won't interpret the funky macros that NeXT uses in some of their #include statements (for the architecture independent stuff) if the -traditional switch is on. This seems like a bug to me. Has anybody else hit this or have a workaround? Are there #pragmas that will do the same thing for me? The program dies when compiled with normal gcc. Thanks. Dave -- David K. Hess Graduate Assistant David-Hess@tamu.edu Supercomputer Center Texas A&M University
Newsgroups: comp.sys.next.bugs From: waldvoge@nessie.cs.id.ethz.ch (Marcel Waldvogel) Subject: /bin/sh $# bug with nested functions Message-ID: <1993Jan12.160259.7994@bernina.ethz.ch> Sender: news@bernina.ethz.ch (USENET News System) Organization: Swiss Federal Institute of Technology (ETH), Zurich, CH Date: Tue, 12 Jan 1993 16:02:59 GMT Another nasty bug (which prevents me using 'shql' for now): #!/bin/sh nested_func(){ echo "inside: $#" } func(){ echo "before: $#" nested_func 6 7 8 echo "after : $#" } func 1 2 3 4 5 gives the following output (on 2.1 and on 3.0): before: 5 inside: 3 after : 3 Every other system I've tried returns before: 5 inside: 3 after : 5 This is in accordance to what I expect from the behaviour of parameter passing. -Marcel
From: d1h1883@sc.tamu.edu (Dave Hess) Newsgroups: comp.sys.next.bugs Subject: Re: gcc preprocessor and -traditional problem Date: 12 Jan 1993 17:46:51 GMT Organization: Texas A&M University, College Station, TX Distribution: world Message-ID: <1iv06bINNqft@tamsun.tamu.edu> References: <1it2rtINNk7i@tamsun.tamu.edu> In article <1it2rtINNk7i@tamsun.tamu.edu> d1h1883@sc.tamu.edu (Dave Hess) writes: > > I was trying to put together a package (detex 2.4) that needed to be compiled > with -traditional and gcc puked on some NeXT include files. gcc apparently > won't interpret the funky macros that NeXT uses in some of their #include > statements (for the architecture independent stuff) if the -traditional > switch is on. > Apparently this is a bug and is known. I was forwarded the following. > From: Kate Smith <Kate_Smith@NeXT.COM> > Date: Wed, 30 Dec 92 14:43:55 -0800 > To: <serge@Princeton.EDU> > Subject: [37670] @!compiler problems - something about the comments in math.h > > Serge, > We have a bug with our machine dependant include files when using the -traditional flag. We have an include file that is used to build the name of the machine dependant include file that really needs to be included. This file uses an ansi directive that is invalid in traditional mode. > > So in order for you to get the math.h file included, you need to specify the fullpath in your program, or modify math.h to include the proper file. > > So your program would look like this: > #include </NextDeveloper/Headers/ansi/m68k/math.h> > > main() > { > double x; > x=cos(1.2); > printf("%f\n",x); > } > > Or you could change math.h to include /NextDeveloper/Headers/ansi/m68k/math.h instead of doing the machine dependant stuff. > > Sorry for the inconvenience. > > Kate It looks like the workaround is to fully qualify pathnames. The problem is that this won't work on include files that include others. i.e. <sys/param.h>. I would have to expand by hand all of macros. Ugh. Thanks to Serge J. Goldstein for the info. Dave -- David K. Hess Graduate Assistant David-Hess@tamu.edu Supercomputer Center Texas A&M University
From: chet@odin.ins.cwru.edu (Chet Ramey) Newsgroups: comp.sys.next.bugs Subject: Re: /bin/sh $# bug with nested functions Date: 12 Jan 1993 18:35:11 GMT Organization: Case Western Reserve University, Cleveland OH (USA) Message-ID: <1iv30vINNe8q@usenet.INS.CWRU.Edu> References: <1993Jan12.160259.7994@bernina.ethz.ch> In article <1993Jan12.160259.7994@bernina.ethz.ch> waldvoge@nessie.cs.id.ethz.ch (Marcel Waldvogel) writes: $ Another nasty bug (which prevents me using 'shql' for now): $ $ #!/bin/sh $ nested_func(){ $ echo "inside: $#" $ } $ func(){ $ echo "before: $#" $ nested_func 6 7 8 $ echo "after : $#" $ } $ func 1 2 3 4 5 $ $ gives the following output (on 2.1 and on 3.0): $ $ before: 5 $ inside: 3 $ after : 3 $ $ Every other system I've tried returns $ $ before: 5 $ inside: 3 $ after : 5 $ $ This is in accordance to what I expect from the behaviour of parameter passing. The NeXT sh is based on the System V.2 sh (probably filtered through BRL, though I don't know for sure). The V.2 sh had the behavior you've observed regarding $# and functions -- there was only a single array of arguments, and it was overwritten each time a function was called. The V.3 sh changed that so that $# is saved and restored around each function call. Posix.2 has standardized the V.3 (and ksh, and bash, and the new BSD sh, and ...) behavior, but NeXT has not upgraded its version of sh. Chet -- ``The use of history as therapy means the corruption of history as history.'' -- Arthur Schlesinger Chet Ramey, Case Western Reserve University Internet: chet@po.CWRU.Edu
Newsgroups: comp.sys.next.bugs From: gtoews@engr.UVic.CA (Greg Toews) Subject: What is NeXTs policy on software upgrades Message-ID: <1993Jan12.210939.3468@sol.UVic.CA> Sender: news@sol.UVic.CA Organization: University of Victoria, Victoria, BC, Canada Date: Tue, 12 Jan 93 21:09:39 GMT I'm trying to decide whether or not to go to NS 3.0. Does anyone know whether an improved version (3.1?) would again cost me? What was the story from going 2.0 -> 2.1 -> 2.2? Thanks Greg
Newsgroups: comp.sys.next.bugs From: gemoe@proximus.north.de (Gerhard Moeller) Subject: NeXTmail and fonts in it... Message-ID: <1993Jan12.151436.520@proximus.north.de> Sender: gemoe@proximus.north.de Organization: Gerhard Moeller, German NeXT User Group, Oldenburg. Date: Tue, 12 Jan 1993 15:14:36 GMT Hi, Another funny thing: (I can't explain it, but it seems to be a bug.) I use MailHelper to include my .signatures to my mailings with NeXTmail. (Please add that functionality to the next edition of NeXTmail, NeXT!) Now I have a .signature.rtf in Courier Medium 10.0 Pt. as I always did. If I include this rtf into a NeXTmail sendwindow, the space between the lines suddenly is 12pt, though the font itself is still 10pt. It seems to be that 12pt is the minimum line space I can use. Funny, because: (a) this was not like this in 2.1 (b) this is not true if I use NeXTmail as root!!!! (c) this is not true for Edit. ad (b) The only difference between gemoe and root is that gemoe uses German and root English in the prefs. Is that the reason? Puzzled, Gerhard -- +---------------------------< principiis obsta! >---------------------------+ N Gerhard Moeller, Teichstrasse 12, 2900 Oldenburg (FRG) [*: 02/21/1968] N e Private: gemoe@proximus.north.de Phone (voice): +49-441-75520 e X Uni: Gerhard.Moeller@arbi.Informatik.Uni-Oldenburg.DE NeXTmail X T Z-Net: Gerhard.Moeller@uniol.zer encouraged! T +-> NoGeNUG - Northern German NeXT User Group: NoGeNUG@proximus.north.DE <-+
From: glenn@rightbrain.com (Glenn Reid) Newsgroups: comp.sys.next.bugs Subject: Re: My old Font-Problem with NS3.0 Message-ID: <1070@rtbrain.rightbrain.com> Date: 14 Jan 93 01:54:41 GMT References: <981@rtbrain.rightbrain.com> Sender: glenn@rightbrain.com [this didn't make it out before; reposted] Borris Balzer writes > > Hi there, > > posting to other groups didn't bring any hints to solve my font-problem in an > acteptable way (BTW: Thanks to all who answered). > > The problem I've got: > > NS2.1 installed, I used without problems fonts with the following names: > - Garamond (the ITC Garamond) > - AdobeGaramond > - StempelGaramond > - Garamond Three > After installing NS3.0 the AdobeGaramond and StempelGaramond disappeared from > the font panel. I posted a response to this. Maybe it didn't make it out. There is a known bug in NS 3.0 related to the Garamonds, and you have hit it squarely on the head. I have been in correspondence with NeXT about it, and a fix should make it into the next release, whenever that may be. In the interim, your only choice is to edit the AFM files for fonts that are disappearing to change the FamilyName property to add some extra characters, so it won't sort into the same pile as the other Garamonds. This won't affect anything except the way the fonts are displayed in the font panel. -- Glenn Reid NeXTmail: glenn@rightbrain.com RightBrain Software 415-326-2974 (NeXTfax 326-2977) Palo Alto, California Electronic Frontier Foundation, member #054
Newsgroups: comp.sys.next.bugs From: hinkle@sundance.cso.uiuc.edu (Steven Hinkle) Subject: Where is putenv or setenv??? Message-ID: <C0up7z.DI@news.cso.uiuc.edu> Originator: hinkle@sundance.cso.uiuc.edu Sender: usenet@news.cso.uiuc.edu (Net Noise owner) Followup-To: comp.sys.next.bugs Organization: University of Illinois at Urbana/Champaing - CCSO Date: Thu, 14 Jan 1993 15:47:57 GMT Anybody know where NeXT put the routines "putenv" or "setenv"? I can use "getenv" but what good does it do me, if I can't change the environment. Anyone have a work around for this? Thanks, -- Steven Hinkle Computing and Communication Services Office s-hinkle@uiuc.edu University of Illinois at Urbana/Champaign
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: Known Bugs in NeXTStep 3.0 Message-ID: <9301141953.AA00817@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Thu, 14 Jan 1993 19:55:32 GMT rfschtkt%banruc60.BITNET@pucc.Princeton.EDU is maintaining a list of Known Bugs in NeXTStep 3.0 (along with his suggestions for improvements). Version 3 is available from the Purdue archives, under /lore (I believe). You can contact him for further information after obtaining the archives. He cannot read this group, however; so postings here won't be noticed by him. --- Subrata Sircar|ssircar@canon.com (NextMail ok)|Prophet & SPAMIT Charter Member Canon Information Systems and I do not share the same views on everything. "I'm just mad that I missed the sexual revolution." - me "Yes, but you dress much better as a result." - Mike
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: Typed streams library error: file inconsistency Message-ID: <9301150042.AA01308@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Fri, 15 Jan 1993 00:44:18 GMT Begin forwarded message: From: Ali Ozer <Ali_Ozer@next.com> Date: Thu, 14 Jan 93 14:21:01 -0800 To: lam@tds.com Subject: Re: Typed streams library error: file inconsistency Cc: next-prog@cpac.washington.edu This is due to a bug in NXSizeBitmap(); it will give bad results when reading bitmaps from 12-bit windows (such as those on the NSC) and the width is not an even number. As a result, the wrong number of bytes is written out when a cached image is archived. However, upon unarchiving, the correct number of bytes is requested, so the unarchive fails. In your example, for a 611 x 520 image, the correct number of bytes would be 476840, which is what is expected. However 477360 bytes were written out, because that's what NXSizeBitmap() reports... One workaround is to always write out an even number of pixels when you have 12-bit images. Another is to not archive the actual image (which is inefficient anyway) but somehow get the image stored in an image file (.tiff, or .eps, if appropriate), separate from the typedstream. In your case the NXImage you archive has its backing store in a NXCachedImageRep, and the actual bug is in NXCachedImageRep's write: method, which uses NXSizeBitmap(). This is fixed in 3.1. Ali Begin forwarded message: Date: Wed, 13 Jan 93 19:37:08 PST From: TDS.COM!lam@tds.com To: next-prog@cpac.washington.edu Subject: Typed streams library error: file inconsistency Hello all, My application runs on NeXTstep 2.2 and dealt with archiving/unarchiving an object of type NXImage (using NXWriteRootObjectToBuffer()/NXReadObjectFromBuffer()). Occasionally, the application encounters the following error during the unarchiving (using NXReadObjectFromBuffer()): Typed streams library error: file inconsistency: read '[477360c]', expecting '[476840c]' (The values in side the square brackets are not always 477360c and 476840c). partial code: ... tempPic = NXReadObjectFromBuffer(data, bytes); ... The curious thing is that this error only happens occassionally and expecially when I try to unarchive an NXImage of Digital Librarian of 611 (width) by 520 (height). Any suggestion as how I might go about fixing this error is greatly appreciated. Sincerely, --Dennis Please reply to lam@tds.com Dennis Lam Trident Data Systems 5933 W. Century Blvd., Ste. 700 Los Angeles, CA 90045 (310) 645-6483
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin From: jdyer.wbst129@xerox.com (John D. Dyer) Subject: Trouble upgrading to 3.0 Message-ID: <1993Jan14.174441.13524@spectrum.xerox.com> Keywords: 3.0,upgrade,help Sender: news@spectrum.xerox.com Organization: Xerox Corporation, Webster NY Date: Thu, 14 Jan 1993 17:44:41 GMT Well, I'm probably one of the last NeXT users to 'try' to upgrade, and I'm having much difficulty. Here's my current configuration: NeXTStation Mono 040 400 Meg Hard Drive, 130 Megs free Release 2.0 (current) NeXT CD-ROM Drive/3.0 bundle Installing the software from the Tools floppy is no problem. After the CD-ROM shows up in the Workspace Manager, I access it. I then double click on the 3.0upgrade.app (sp?). I choose Update, and then Edit Plan since I want to keep some things the way they were (my uucp config files for example). While updating it says "Writing the control file to disk". The gauge starts moving to the right. After a while one of two things happens: the upgrade.app terminates in the middle of writing the control file, or the entire machine locks up, and I have to do a reboot from the monitor. Has anyone seen this happen before. My apologies if this is in a faq somewhere, but I haven't been able to read the news lately. If anyone can give me any information, I will be forever indebted. Please email at jdyer.wbst129@xerox.com Thanks in advance. John Dyer
From: hacker@access.digex.com (Dark Hacker) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: Bug with nested macros? Date: 15 Jan 1993 15:23:41 GMT Organization: Express Access Online Communications, Greenbelt, MD USA Distribution: world Message-ID: <1j6kttINNa7c@mirror.digex.com> Is there some kind of problem with the GCC delivered with NS 3.0 and macros? I have a program that uses nexted macros and nested include files extensively ... all of this in ANSI C mind you... no Objcetive C or anything having to do with appkit. Now when I compile this program I get errors that look something like: blah_dee_func 620: Syntax error following ')' 642: Syntax error following ')' 649: Syntax error following ')' 680: Syntax error following ')' This is as close as I can remember it. Now the errors all occur on macro invocations. All of this worked under NS 2.1 too but uner NS 3.0, well.... Anybody know what's going on here? Should I back out to an old GCC or non-NeXT GCC? Is there a way to kludge or fix the existing system of even my code? - Hacker -- Dark Hacker @ Black Silicon, Fortress Of Computation hacker@black-silicon.mclean.va.us "Life itself is... COMPUTATION!"
Newsgroups: comp.sys.next.bugs From: apf@io.fct.unl.pt (Antonio Pascoa) Subject: HZ Message-ID: <1993Jan15.174931.27713@fct.unl.pt> Sender: news@fct.unl.pt (USENET News System) Organization: Universidade Nova de Lisboa, PORTUGAL Date: Fri, 15 Jan 1993 17:49:31 GMT The header <bsd/sys/param.h> fails to include <bsd/m68k/param.h> where the definition for HZ is. BTW the manual says 60 when in fact it's 64. Is there any official way to report bugs? Boas, Pascoa PS: The GUI seems to be losing events more often. Menus keep forgetting to un-highlight the MenuCell.
From: louie@sayshell.umd.edu (Louis A. Mamakos) Newsgroups: comp.sys.next.bugs Subject: Re: HZ Date: 17 Jan 1993 16:04:07 GMT Organization: The University of Maryland, College Park Message-ID: <1jc01nINN7q2@ni.umd.edu> References: <1993Jan15.174931.27713@fct.unl.pt> In article <1993Jan15.174931.27713@fct.unl.pt> apf@io.fct.unl.pt (Antonio Pascoa) writes: >The header <bsd/sys/param.h> fails to include <bsd/m68k/param.h> where the >definition for HZ is. BTW the manual says 60 when in fact it's 64. The value of HZ is likely just fiction. It's likely there just for compatibility reasons so that the times(3) library routine "works:" TIMES(3C) UNIX Programmer's Manual TIMES(3C) NAME times - get process times SYNOPSIS #include <sys/types.h> #include <sys/times.h> times(buffer) struct tms *buffer; DESCRIPTION This interface is obsoleted by getrusage(2). Times returns time-accounting information for the current process and for the terminated child processes of the current process. All times are in 1/HZ seconds, where HZ is 60. >Is there any official way to report bugs? You can send bug reports over the event horizon, into the black hole to this email address: bug_next@NeXT.COM. I get automated advisories back with a bug report number. You will likely never hear anything else back about the status of your bug report after that. It either gets fixed, or you get to report it again in the next release. Hope you don't encounter a "Mission Critical" bug in NeXT software, where you need to know if or when it will be fixed. Of course, if you bought NeXT support, you could pay big bucks for the opportunity to tell NeXT about their software problems. In this case, you might get a different (any?) response. Who knows? Louis Mamakos
Newsgroups: comp.sys.next.bugs From: davisre@sage.cc.purdue.edu (Robert Davis) Subject: 3.0 Workspace's Attributes Inspector Message-ID: <C19tn1.KGt@mentor.cc.purdue.edu> Sender: news@mentor.cc.purdue.edu (USENET News) Organization: Purdue University Computing Center Date: Fri, 22 Jan 1993 19:47:25 GMT When I used the Attributes Inspector to take a look at the size of /private/vm/swapfile.front, I was surprised to find that it reported 328 MB. I have a 400/20 mono NeXTStation with the entire Extended release installed (documentation, TeX, etc.). Hmm. ls shows it to be about 15 MB. Rob -- | Robert Davis davis@sonata.cc.purdue.edu | "Look up, Hannah. Look up." NeXT Mail accepted --
Control: cancel <C19tn1.KGt@mentor.cc.purdue.edu> Newsgroups: comp.sys.next.bugs From: davisre@sage.cc.purdue.edu (Robert Davis) Subject: cmsg cancel <C19tn1.KGt@mentor.cc.purdue.edu> Message-ID: <C1A578.5J6@mentor.cc.purdue.edu> Originator: davisre@sage.cc.purdue.edu Sender: news@mentor.cc.purdue.edu (USENET News) Organization: Purdue University Computing Center References: <C19tn1.KGt@mentor.cc.purdue.edu> Date: Fri, 22 Jan 1993 23:57:07 GMT <C19tn1.KGt@mentor.cc.purdue.edu> was cancelled from within rn. -- | Robert Davis davis@sonata.cc.purdue.edu | "Look up, Hannah. Look up." NeXT Mail accepted --
Newsgroups: comp.sys.next.bugs From: davisre@sage.cc.purdue.edu (Robert Davis) Subject: /etc/mtab permissions Message-ID: <C1A5Ep.6nI@mentor.cc.purdue.edu> Sender: news@mentor.cc.purdue.edu (USENET News) Organization: Purdue University Computing Center Date: Sat, 23 Jan 1993 00:01:36 GMT This may have been mentioned before (I recently upgraded to 3.0), but is there a workaround or fix for this: ejecting a mounted floppy changes the permissions of /etc/mtab so that you can't do a df. % df Filesystem kbytes used avail capacity Mounted on df: /etc/mtab: Permission denied Rob -- | Robert Davis davis@sonata.cc.purdue.edu | "Look up, Hannah. Look up." NeXT Mail accepted --
Newsgroups: comp.sys.next.bugs From: Gregory W. Gee <gee@gaul.csd.uwo.ca> Subject: font problem again. Organization: Relayed-by-Sendmail Date: Sat, 23 Jan 1993 01:18:14 GMT Message-ID: <9301230118.AA21235@gaul.csd.uwo.ca> To: comp.sys.next.bugs@newshost.uwo.ca Sender: daemon@julian.uwo.ca (Julian System Daemon Account) I sorry if I missed a fix for this problem but I'm not sureif it's the same one. I am having problems with my word perfect. I can't set things to bold, it just puts a 1 on the document. When I try to start an underline, it won't shut off. I can do things fine from the font panel but not from the app menu. Soory if this is a repeat problem but I might have missed the first one. Thanks. **************************************************************************** Greg Gee gee@gaul.csd.uwo.ca NeXT Mail accepted Honours Computer Science, University of Western Ontario >>> Developing in the NeXT generation <<< ****************************************************************************
Control: cancel <9301230118.AA21235@gaul.csd.uwo.ca> Newsgroups: comp.sys.next.bugs From: gee@gaul.csd.uwo.ca (Gregory W. Gee) Subject: cmsg cancel <9301230118.AA21235@gaul.csd.uwo.ca> Organization: Computer Science Dept., Univ. of Western Ontario, London, Canada Date: Sun, 24 Jan 1993 17:16:44 GMT Message-ID: <1993Jan24.171644.20385@julian.uwo.ca> Originator: gee@gaul.csd.uwo.ca References: <9301230118.AA21235@gaul.csd.uwo.ca> Sender: news@julian.uwo.ca (USENET News System) <9301230118.AA21235@gaul.csd.uwo.ca> was cancelled from within trn. -- **************************************************************************** Greg Gee gee@gaul.csd.uwo.ca NeXT Mail accepted Honours Computer Science, University of Western Ontario
From: rhess@consilium.COM (Richard L. Hess) Newsgroups: comp.sys.next.bugs Subject: Re: /etc/mtab permissions Message-ID: <RHESS.93Jan25093544@consilium.COM> Date: 25 Jan 93 17:35:44 GMT References: <C1A5Ep.6nI@mentor.cc.purdue.edu> Sender: root@cimshop.UUCP Organization: Consilium Inc., Mountain View, California. In-reply-to: davisre@sage.cc.purdue.edu's message of 23 Jan 93 00:01:36 GMT >>>>> "Robert" == Robert Davis <davisre@sage.cc.purdue.edu> writes: Robert> This may have been mentioned before (I recently upgraded to Robert> 3.0), but is there a workaround or fix for this: ejecting a mounted Robert> floppy changes the permissions of /etc/mtab so that you can't do a df. I had the same problem. It took me a while to figure it out, but I'd bet that you have very tight permissions set on your "File Creation Mask". If you just use "Expert Preferences" (Unix) in the Preferences tool to open up your permissions to allow "Group" & "Others" Read access, then you won't have any problems with the permissions changing on /etc/mtab. Evidently when the eject script/program executes it runs as root and uses your default permissions and so if your permissions are too "tight" you can't read /etc/mtab after the floppy get's ejected. Hope this helps, Dick... ps. I don't remember whether it was ejecting Macintosh or DOS format floppies that caused this problem, but one worked and the other didn't and I never had problems with the NeXT (Unix) format floppies... +---------------------------------------------------------- WHO: Richard Hess [ Staff Engineer ] CORP: Consilium Inc, 640 Clyde Court, Mtn View, CA 94043 VOICE: [415] 691-6342 EMAIL: rhess@consilium.COM <-----------------[ NeXTmail ] +
Newsgroups: comp.sys.next.bugs From: davisre@sage.cc.purdue.edu (Robert Davis) Subject: Re: /etc/mtab permissions Message-ID: <C1FBwD.5Br@mentor.cc.purdue.edu> Sender: news@mentor.cc.purdue.edu (USENET News) Organization: Purdue University Computing Center References: <C1A5Ep.6nI@mentor.cc.purdue.edu> <RHESS.93Jan25093544@consilium.COM> Date: Mon, 25 Jan 1993 19:09:45 GMT In article <RHESS.93Jan25093544@consilium.COM> rhess@consilium.COM (Richard L. Hess) writes: > > I had the same problem. It took me a while to figure it out, but I'd bet >that you have very tight permissions set on your "File Creation Mask". If >you just use "Expert Preferences" (Unix) in the Preferences tool to open up >your permissions to allow "Group" & "Others" Read access, then you won't have >any problems with the permissions changing on /etc/mtab. Evidently when the >eject script/program executes it runs as root and uses your default >permissions and so if your permissions are too "tight" you can't read >/etc/mtab after the floppy get's ejected. Hope this helps, > Thanks, that seems to be a good workaround. I had trouble with NeXT disks as well as DOS and Mac disks. Rob -- | Robert Davis davis@sonata.cc.purdue.edu | "Look up, Hannah. Look up." NeXT Mail accepted --
Newsgroups: comp.sys.next.bugs From: ravn@imada.ou.dk (Thorbjoern Ravn Andersen) Subject: NFS mounts -- time stamp problems? Message-ID: <1993Jan25.191725.23058@imada.ou.dk> Sender: news@imada.ou.dk (USENET News System) Organization: Dept. of Math. & Computer Science, Odense University, Denmark Distribution: comp Date: Mon, 25 Jan 1993 19:17:25 GMT We have a Unix network with Sun's and NeXT machines mixed together, with home directories on the NeXT hard disks. Apparently, date stamping works incorrectly when setting stamps explicitly, instead of inheriting the dates from the file system. Touching a file yields a date of January 1 1970, which is somewhat incorrect. I *think* this is date Zero, but why should that be? Anybody got any suggestions for circumventing this problem? -- Thorbj{\o}rn Andersen ravn@imada.ou.dk
Newsgroups: comp.sys.next.bugs From: gemoe@proximus.north.de (Gerhard Moeller) Subject: Re: /etc/mtab permissions Message-ID: <1993Jan25.062236.6355@proximus.north.de> Sender: gemoe@proximus.north.de Organization: Gerhard Moeller, German NeXT User Group, Oldenburg. References: <C1A5Ep.6nI@mentor.cc.purdue.edu> Date: Mon, 25 Jan 1993 06:22:36 GMT In article <C1A5Ep.6nI@mentor.cc.purdue.edu> davisre@sage.cc.purdue.edu (Robert Davis) writes: > > This may have been mentioned before (I recently upgraded to > 3.0), but is there a workaround or fix for this: ejecting a mounted > floppy changes the permissions of /etc/mtab so that you can't do a df. > > % df > Filesystem kbytes used avail capacity Mounted on > df: /etc/mtab: Permission denied Yes, I had the same problem. And it doesn't help to change the permissions again. (Except you do it with a cron job. But this is ugly!) Especially bad is the problem if you use cnews... What I did: I simply set the suid bit of df. But setting a suid always worries me... Any better solutions out there? Also strange: Since I installed the MOTD.app multi-volume disk copys don't work anymore!! Neither in a net nor at a single user station. When will 3.1 be out? Gerhard. -- +---------------------------< principiis obsta! >---------------------------+ N Gerhard Moeller, Teichstrasse 12, 2900 Oldenburg (FRG) [*: 02/21/1968] N e Private: gemoe@proximus.north.de Phone (voice): +49-441-75520 e X Uni: Gerhard.Moeller@arbi.Informatik.Uni-Oldenburg.DE NeXTmail X T Z-Net: Gerhard.Moeller@uniol.zer encouraged! T +-> NoGeNUG - Northern German NeXT User Group: NoGeNUG@proximus.north.DE <-+
Newsgroups: comp.sys.next.bugs From: gemoe@proximus.north.de (Gerhard Moeller) Subject: FontPanel select... Message-ID: <1993Jan26.204432.986@proximus.north.de> Sender: gemoe@proximus.north.de Organization: Gerhard Moeller, German NeXT User Group, Oldenburg. Date: Tue, 26 Jan 1993 20:44:32 GMT Hi, either I'm stupid or... It says in the 3.0 release notes: o The FontPanel, when key, will select in the Family scrolling list when you type the first few characters of a family name. But when I type letters they appear in the size field. No way to get the damn thing working. Any hints? Gerhard. -- +---------------------------< principiis obsta! >---------------------------+ N Gerhard Moeller, Teichstrasse 12, 2900 Oldenburg (FRG) [*: 02/21/1968] N e Private: gemoe@proximus.north.de Phone (voice): +49-441-75520 e X Uni: Gerhard.Moeller@arbi.Informatik.Uni-Oldenburg.DE NeXTmail X T Z-Net: Gerhard.Moeller@uniol.zer encouraged! T +-> NoGeNUG - Northern German NeXT User Group: NoGeNUG@proximus.north.DE <-+
Organization: Senior, Chemistry, Carnegie Mellon, Pittsburgh, PA Newsgroups: comp.sys.next.bugs Message-ID: <QfNRoLC00YUnJYCTBk@andrew.cmu.edu> Date: Tue, 26 Jan 1993 19:53:43 -0500 From: Charles William Swiger <infidel+@CMU.EDU> Subject: Re: NFS mounts -- time stamp problems? In-Reply-To: <1993Jan25.191725.23058@imada.ou.dk> Distribution: comp Excerpts from netnews.comp.sys.next.bugs: 25-Jan-93 NFS mounts -- time stamp pr.. by Thorbjoern Andersen@imad > Apparently, date stamping works incorrectly when setting stamps > explicitly, instead of inheriting the dates from the file system. > > Touching a file yields a date of January 1 1970, which is somewhat > incorrect. I *think* this is date Zero, but why should that be? 1/1/70 was chosen to be the "epoch" for Unix filesystems. All of the timestamps used count in seconds from epoch to the current time. (Some trivia: the values are stored in 4-byte longs, meaning that the time system will break in the year 2038.) > Anybody got any suggestions for circumventing this problem? I suspect that one or more machines on your network are having problems with their clocks. Make sure that you are using the Network Time Protocol on all of the machines on the network. NTP was explicitly designed to help keep distributed filesystem's timestamps correct between machines sharing files. Consult your System's Administration manuals.... -Chuck Charles William Swiger -- CMU...*crunch*! | "Foosh. Aaughh!!" ------------------------------------------+ "Foosh. Aauuggghh!!" AMS & normal mail: infidel@cmu.edu | "Cold spray deodorant...." Failing that: cs4w+@andrew.cmu.edu | NeXTmail: chuck@mon.slip.andrew.cmu.edu | -- Opus, Bloom County [RIP]
From: hama@potenza.fjh.snet.or.jp (HAMADA) Newsgroups: comp.sys.next.bugs.ctl Subject: newgroup comp.sys.next.bugs Message-ID: <136@potenza.fjh.snet.or.jp> Date: 24 Jan 93 08:49:14 GMT Control: newgroup comp.sys.next.bugs Distribution: comp Organization: FUJITSU HOKURIKU SYSTEMS MASUIZUMI,KANAZAWA,ISHIKAWA,JAPAN
Newsgroups: comp.sys.next.bugs From: James W. McKelvey <fafnir!mckelvey@uu.psi.com> Subject: Code Generation Bug Next-Attachment: .tar.92.Code_Generation_Bug.attach, 28915, 1/1, 39798, 0 Message-ID: <9301270810.AA11317@fafnir.fafnir.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Wed, 27 Jan 1993 08:10:51 GMT begin 666 .tar.92.Code_Generation_Bug.attach M'YV0:=R0*8/'A1PZ9@`H7,BPH<.'$"-*G$BQ(HB+-FC0``'@XD49&SMZ!`&# MXTB/,VS<J'&QQHP9,33&L`%CAL<:*F-TK,BSI\^?0(,*'4JTJ-&C2),J5;J' MRT$S,+B$<3,G35,S;]S0H2.>H7]N\&2C'#8@A;^K(25-&SHZO,;[.N9-F MSAP02,JPL5.&3IHQ87;T4<"E31@Y9]C$D`&#L&'$<A8W;CKF#9LW![N^E5.& M3-0SG,NXB=JU3AD8@@G#.4R&"QT\,69H=(U'A@T<.&C3J"F#MLL;-FBK9$P; M!XR,M'/LSOTZ!HR5O9LSSA&\N>S=7TE'31.U#ALW6<O(E9';S)BH8Z""F%)' MC)PP=,*`F%/GS)DR<^AT!I$$!%8Y=[`&`AUHE`&"&/6!P!D<F-'!WWRB.4C' M&R`X40865.C`!6$*"%'&&0'YAUF`<@Q$!@AMX#='&/=IR*%JK-$6&W:OV6:< M;C`0]YI+-`EW@XYX&-?C:\K!$%5SS]$(&V,W6J=15&;$)<97YW&17DE$P%>& M#H2!)0:48\257@P746$:"R#8!H(24X&0@TTRS*`##3;H<!P(4$Q!Q89=2@F" M$7*\T0:7V7'QY45MC+&&7GSE`0(*5+RAPX'U?>$&070`<2D>=+A061LI\`F7 MH2!$2JB7)5%ZAJ68:HJIIX**&B6I[(FA1AECT'&J=A>A-1`(1XC6%GQI9`6" M$/7)ZJ<49<#!1AXMF-HEKR@JRF@9>0!A1AAFN)&&'+"VL:&L55X)PA4&HA$& M7R+6(1`(\E7VZWV7ON>7L0B>`4*(!!HX@PLE#0'"IW"DP49;+ER4A(-OC#&& M6G?112`(;\#A5QMIZ$&LL4\8843"I1;(&;PCTW$'A7/DT8;%@L[A8@PIG#66 M@6WDD5];Q=8QQ[,@W$?'77-\F*)6&YNUK7<,RT%QOTH'S=?(;04*[H8RQ!P$ M"'8<ED8871E8%P@B8&Q70&>(L*]91Q`AA`OC*@!%H%VW`0*97Y-AVH`4]CNP MH`4?K#2!`:U!]H#J.MCFU9R!>+/2E1$=T%UZOV&&&4$[^!^\9H51QX2&^34& MUEIS?3`(=Q1H5AH.DO$&?B"`UVF9!8*^ENA>0Q[[BBE2?"F\=X4(!]P'RRT# MVQQ2$3M6;%Q&EQOZNG"7&08;.)!85-%A+^MZ__Y&W"`[\8;G!A(('^%MU=XZ MA0/EJR"NF)UXN>:<$_MYUK-W#79`;KB!AP@@]Z<NNX]9U(DDA@:P8>&`9LL> M&\(P!M:5KGS[6EF@X-"6O?WJ:_0:EG[(`#)F*4X_2@N"#/9UE]_AIPXI.M&$ M]N8XLT1N<I43D=)$@#_]F0U!#D(="%3'.M<1CT][:`I\XC,&-`R-#CG(T0[M M]H4PD($,7XA/U[X`/;UP<`X*&`P75E,B&<GF2+6Y#1AW`R0>@7$X8!02&(L$ M1N?<0$F2:1)LK@,E[7"!.USP#G@N-9[RE$L];<O+R+XF'_WD9V!A"!H(\I`6 M1)I%9P:Z7/KJ`R+FN0B(0MP*`XT8H1S48(3Y^4)^YC`0,[@`#5E40!"EHLDB M'M&3(WQB'9KXQ"C2CHIL80,'QY#*50YQDZ_\Y'SH\(6!S&$,:X'#*7N922)R M4BNP7.(LXV`:TRQS,+YLY3.1*$S#!,13S&2E,X,92R:6,B!EN*8JFPG,3@I3 MEE](D5U8E$Y4:I&+K7$2&&TTQAR!T8P^`I(:D[.;-B:IC4QJ(QT+]:4[=N<[ MX>DCE=`#2`X9`3/^40O3SO:?SA7+#6AJ'!%UU38EI$%N31C#$J[E*$19:R_8 MTA:WO`6N3WD$HXE:%$RS11_4I?-3&XI!"VX0DQ:\"0<MP$T-HJ*`5"[EJ1"! MIQ.A*,6#X=**SH.J5GWBD8R$9"0@,<E(2B*2D<R@!C7834M>$I.4V"`V%W%. M#):ZDZW:]:YXS:M>]\K7OA)E!$YXPA>"``4H*,!Y4KR+"*1:RZJ6X:JZ])0( M6.`<&+#`LI9E@GXXU1@7;)8."M`LIF#`)<1R3;$!T14=8M"#R.R@!3*(R1MQ MD!(:X&`'L:7!;&M[@QU,=C$XN*QP.YN?TX*MB(?1`1UDT%H9[``UB^GM;\DS MW,L>MKAB4.QEF'<VU<Z@M3%X;6YWFY';CM>VO/4M9:F+6>N:-KM@<U=5SG"I M$Z56N30`[W-?&][I!K>]Q$TLV+:K+_FF@;[[N2\=:J!?U+2@O^O][W"O*V`1 MS`$-#>JN<FV@WQ;,0`;`N>V'@2/="%<WP,85`8%!L&(%WZ#!_%4O<$],X11? M.,,&1K!]M:)<',#8!F@]JXS9.^'W:C<K^EIQCNNK823"^,%#EC!F:PQ?"Q^8 MR<B5@W*=TV'@[B"Z4::QD>-+E2OO)\M;9JUK42,#M(89P%16K!DN`Y\M,]>U M--BO?\6,7<6JKCT'V_)W77M;U.P9SF-6,9)WF):N;3F_A-:SB1'=9[#9X0UI M(,.6&3S70Q>YTB*P6(G2,+DH*G<&GIYRHJNBA\>JM@:I=B^H[Y!E4],!![%& M<96S]@4VU$6U,^F!"F20ZSB#[0O%0I`9=$"%&+QX#HSYPAAX'`/A?G@'7Q"U M#F;"@@]WV[E?$$,BM\3MC'P;V\IF];990)US4W&!9]"!#4P\$VQ#+]`QJ'9, M:,""VXK2,,E3M@YD0-E/]GL';_YTA<W`H#G8&M635GB*]22%)#CA"%OV\6'" MR^:TXK;8B0X(5B"KZ6;GH`=SV$T5=3EP&/1@X_O-P9?G;=D/HR;AJ@:UQ8W0 ML204@0E$V'(.0`[J3-,A#Q1D=FQ1CH-,1XC4;-$RMVONW(,QCT#K]O:U<2[K M"B>!"%3(`A2*H-S8$KW";@A#BKY0(">V1>G,G<-B_D,B3<=)V#(@.-5W(&Y% MU7W@,_CVN9W.8Y#@O=KFWOK94^R$(#2A"%]`0A&"0(0B2*'LQ(YXSA>>2S(< M/>E4:#/*\[YRJB*=W#0_=^E[+9HS8+W:6G?NZM.>HFW/._&R[_P7LL8&TXB2 M@F.`>F?6W>YKKY[WIADXAU4P5Q-?^W<G/4P>!DY9FAR<Z[I6K!%\#O2PC[WL ML-9\UU,\$#M\?DNA?_9BRB_*C)';VLYE/^T?:W77HR'K!+^V_-6.>A;@'N&+ M5V659P7>1W;+=0,!J%C45`:^UW9DH70^!FTP0'>L,7`G-VP2=FU]MP9_)P,Y M('C71GBZ`A+^QV_&UWDZ,`,N-VPUP&[SIG]E8`<IR%K#=@/.YUS89VPB$`55 M4`0]&'F35WF7MURX)G[9!S9D<$S%A!_(E`8L(P?%Q&P?QG1).`9?<`>L=W7W M)V'U5H7A5DSGMVWZ-F^WY85B(&T+9!=BR`(D6(9*&`;9AFE:\7;<%GL`:(0Z MZ(7&U(1/*$I2^%US@`->B(7UAW5<:`,[8(9@>'IKR&UN:(5G.`9IZ#+YQH;\ M]HA-%(>I18<OF'\XF(!(J(1[F$P3HF54(!M4J(2$V'J&*%Q=J(1GZ'F,6(F. MF(BPB(:)1(G5UH:V:(5PR"";*'6=>&XYF&A"\`1/P`1E!W)CD!]KP3RA-02D MI0`)X`*)%'QI`#9<``,B<%B?18W6Z&OT-3>'E1C;PP8@4$Q,-%6V-$6EIP#J M.$OLZ%@DQR75B%TG<ANNV!@)X&MNL`;P,F\CT`(V8`,I@0/4*!9\@8YA0'!S M``=`T`+@B(\@H(^6%1.A)01"(`/V^%[Y>(@)^08+"0(C$!MH]5:Q98-A8`-` M@`*P!0,I,)%2])'"Q1C4*(ERP)`LZ9(I$9/W.),5*6$>&)(C.9`%>5;\%@;\ MB),,"0,MN9(ZP`8^Z9%!:6W\J)!Z09($:9`V0'!*>9-LD)/PXI0H`)52*9-< M0Y,U-P-$F94KR0*>08U3A8Y&B58X0'!QF0!8B8Z>P0(/&9'4J`;9I33QB&S> M`EH_F99567,TT)8Z"9?\.)=:"60?-F]YN9<[9%E_*9$)()AS0)CP%!"H@Y9B MH);=5@..&9"0*9=/1)<$60.T%5R7*9)9V9>;&9B#F8ZA>9BD:9HID9IOF9>2 M:90&N52KJ9>TR9>:"9&<Z9F@R42BB9A4:9'==@/`:9F1V9J321/*<9R8:9O, MB9N?J9O0R9N)69J+V6TY<)W'.9Q;>5:R>97)F9E^&9[(.9+!F9UDX)I`!IMX M*9\C"9Z`F0`4)!_%=(;SYW"^EA_B^9RSY#/(-A!:(7QRD)I]N9(M^9)329'4 MN1MM*08!F:'.E@+M*1`@6I+M^41W0)+>]I#4^#-<D9F!N3M,(`.]F9XT$`.I M69)GA1/486ZW2:!E(!\QH`-WT*#I.$IL-Q7[V18WVJ$V>I]9R:,G^:/\EI=J M\!X@P`0Z>IZF20-L6:,=R:$2IA'L*9S:29SPZ9WS*:"<B9GYR9K[.9DN81ML M&J#+.:`%FH[TF*"]]FM(6I@0*H(4:J&6A:$NR1@;"I0=BII2"J*(^F`W0*+" M::(D66UH2@8K6I(."0<OFA\@BJ4TRI9>BJ,VL*,FZ:,%R6]!NJ=%>J2=F9NB M]#-+*A!^\Z1E:IU22I>I:@-6>IQ9*A]<BJO"95NA)2?$:EDUT*58&0=:N1LI M.J=QNJOTV:I#JIOP`8=VP!98*(FY&*AM,`>K@DQ#JA]+"#V'^5&&R@*1JJC) MR@*?)*=QL*G!Y:()`*,ZF:'NVIDTVIBE2ITN@:H]ZJNV5:_VZ:I&&JA*JBZV MZJ3_*F%I);!56K#`JJ7#^K#"A5:A10-CRJ@0>ZKWFA_YZI*[X9-J0*...IT0 MJZN82:4^JEN=JJ?7^JH*2ZL,VZ05BK'*BI#4ZK($^R,5*ZQ=JK(9NYY,4`,= MJYC4V95GJI_\6:?8N:Y!"J<[V0(E6Y\#ZISI&*ZK4CGG&IT?92G\MZ[M"I/O M:I#R2J_U^:ET,+(:.J,&P@0@2[26E1$2ZZ,Y0%W62J0)&ZOC.:L.=[.WJK/^ MYZ@MVZLYD+?Q&9@6.[1D*EP%&5IU<K:ZNJ?%E*V[QZVX:!?7F:%7&Z1:&T_B M*DI]\;7IFA6;BT74>J%5NZ]TZW\\.U7S2I(&R[9NNZ\G&[>Z^KJ^>K>^BAL& M*[-\"ZNAN[!,.KBO>P,`.J6("[Q!NZ6.Z['"152A=0-)BY[4^2/4:+ED,`-? MP!D+="]N$$6GU[DD"Y-8VYRRRK6D2TSG=+KC>WYDV[IF2[@W0*JRJ[;VBJ\A MFJCUF[M;RK/)ZZ^'.[")JQ'IN[TSV[?%:[/'Z[#):[CSZ;,'?*7\&*S0^Z[` M$5HX<+VF>0,L.Y_3ZIY':9RSB:<)3*V1^KGV&;KLZ[7OBSIANX?SJZ_UF[RQ M^T2S.P*U&[)MV[]ORZ]QNY[V2\0%7*5)Q*H'N\#$*ZO&V[`Y^[K&X;N)"P,6 MS+A"^ZXXH*-,D`,>G)ZT9;NZ^1?$Y*<+BIBWXJQ<RH]2W)A<6J1B7$QD++;R M]*<,VIEEH,:+H<6HR00:B:PJ'+4)0,+%N;BKFZ=O*L*"3,)+I;QW.J7OV97L MNKQ-^91U(I5PR:S>L3,DV68M2`9=*IF@_,C*F<+<.\=F#*A^ZZ!?,*@2ZA=5 ME+/?>:CTNZA**V&X\:%`/*(E2@8GBJGZN:F+L;8^'*H73*.QH<5&/,&(NZJF MS,0U&[@/',6/:UE)1,7.C*6-^ZXYP,6Q\<74F;=-*Z=/^QR6=<*0?)22_)4J M3)9F2:F:S`:</`*>G,GDO$/`++7V2;6>B[Z@N[ZC"\-E@*Z=89@U[+^VC+T2 MIAQI2[O$S+]E:[+([*^OZTE4G+>(K,##&\VUBK/<#+)'C+>,\;P76]'6R:4< MR\T\V[*1[)64O)TIX951"J<R8,E12:DO?:&AK,/"S*[,*L+5II0V?9:$F[@9 M602`C,%O'%J+`<Y<:"18#+U][!Q.O8]<K)$I+:3"*HT:G8Y$\-5?T'B/5[-R M,`8KT]`\_-`B*Z-XK,=S6\V451,[&LGSQLXT/=0D*M1E><DQ<T!8\*[.X:^6 M^]5$$-:.5P3C/,AI^IKF3,KP4M-[?=.9#`)^G9";3)?U#,KW/,KH7,K__+?. M.`9PD`>`#0..FK\.;:]*[6Q,/;F$ZQPKS<P&#!W/O-&KG*0.#,6E3<2HG=8N MRJ76"]C.D=B,W-B=_=AX/=F5K9>7W<EH9<^*/:><G=,9K<*0_<[I>Q'+[<(! M7;KO6]`!<=!!_+KYMM-DL,,]#-&U#+?0*\!PG6\S+=M(/-)[.S<,[,2Y[=&O MG<P]B[@83=+1>\NNB)%<VL'"?=HZS*OL>MU\[1'+3=XS<=0IS:5(*]RQC9^+ MO-C]>9>.[:;F:[7^W,(`W;7>/=#H!$7A?<CLNM[[S=L\G=J>ZL.W^[_(3,3D M;9/];<#_7=\T>]N`V]'(^]ZQ==$CK<U9_-IYQ]1>3(W@$>,W?N'IS)5UO91A M6<F1?98U.HT]2]>3#)SNS-<^?9,KPY`,+MEA0";:?4"!>1];.N1(OLP8'JW\ M6<@=7MU4*^=T:MPOK:8%V>7MG-R:S=SRC-G/'>BBG,\J7M]RG"MTC!]VC,:R MRE@H_LJ%JN(1#=@?ILN1RLN5ZLN7&JW"W*EL:\SL7:.D2MXO0<5/4MOVW<2@ MG=]!+KT7*1NJ#JU&GL&O?5:A%2=5/>MO+>L1MNL<.=?J/.7RJN!Z_<Y1G>6[ M#L<)T.2`G:.[GM7\VXY6M7IG')AYW.8I^]XT8*-^+`05'LAXSN<V>-P>3NZ9 M.N=(6><IW*Q:F1)X/JVS_.Z*7.ZO&9ON/K6TZ:P#28+KKIKH;N?W'O#$"0/= M.?"L;NV/A>VJ'+KP!*'O:.DL3MXT`.=E`*E5R^F1::DHFJF;JK<Q#J.D+L1M M_NL"?I%+A<TXP>H]WL#2K-NOO:PL+\BK#=CQ6J.N3=XXT9;^[F'_><_TWJ8$ MSQ<_3UT!/_0HS.]Q;O!;B?!7G.@B_K<1[]W>2P;9*K\4;\,);9IS%<)\H?$B M.JF]_,N@?JFB7LQL#<`UNKOO#9M4_$G!RYD(Z^J$^<3ZS?,8SZL&+/<`#M@T ML>O!_=IHN^7JK)++"ZE?+MEYB2E=P=;U'J1,.1_[W._/&O62J?2U6?11WNYV M+<*+C\E,[Y95VY/VCN$9^F$D.OJW2P.K7_FHGZ@X\/K"BZWQ@6SZT0:I"YQE M+OJP3_KX#F1ZKL\#>N=.7\+G3-VG#_S'#YLS8,B1/_6LS([=2W+B[;KOO<&/ MNLMDW^EF#_(D:7#["ZIK3Z/D`=@@K.HP,/==_?+X'?-YG_U0SO<3N_Y__]HW M8+3DT>N4=9<?KA&T+Y'%/A`7`,%5=R,FG,$,A)L\$*&NWPU[;V$LNJ$WM?;# M+IW)JU$V#@(2,/F&MX89C[MOKP[^Q;J4U_\D6%'R;\/LUI4T"'BJ,"#_BP$X MX*0)@2(P[I0:2-AUXXZ\73/#![7"7'1[3&<..`6UJN7Z`%LWTV4E*2E5+1O0 M]=)3#!!GVP]%1:H:T`2ITQ,D5<;/:=&IRK3OI!]NRTFBH2VM*-:5H6:?$=R` M18G+L;/)E^SX&EI38L=*R[&T8N?GX%3HRVM=RJPY"^1VY?!@FOMKG8G-,0&8 M8`3[F$9R;5GPGM6EX6?XI%P=!'V`+I[-L\SVTU!8$/R!^,P'PJDA***JAA$$ M:;3IERE!$54$7UL.`'L9[]-MNAE0!;E0#IA_`D\++D)'=MSXG+'[<WT0NK6! MYD;/"IUYXTN(+OIE+?P&!MV`&*1/FXX)&L&]%PL5(6.CA7M.#;ZT-LCX)"&A M^V2_,!,J/-8'Q%C8,/Q;+VQ=/)88)KX<8"NL25`MNH4Z"CCC)%K<2G6$BS%4 MPFTWD+A-TA-DPE``,C]^1!_$0!R@2]"/Z+$ZC@2K0MKO,@ZLCF/9/=PF`B$8 M7&,,6)`#'L3DM^P"G$)#A_Y*J=&Z00C(7A>3\'E:R1YJ0<WGV7[?8\I+_?`? M.K1MN,3D`T'T774J\-0WA<C1!)=#!':Q)9AULN5TK+*:1\PW1VWGO3>&-@B' M72-T*WY.=B&[.[CL5-"Q<G8_,2(.0$5U_WZB&YN!1M$J,C5[-`>Y4G#Y?`,P M*9$63-:0J)&[8`,`:251HX.`11*-I&-X]<@(Q(#)(N_:2V&:1[?D'24:>;(B M[D,,6$+')`71@%B35'007Z0G,@`PC@'!2!@CW*J)#W1`9VP;PFB3$@VV8T7W M9RZ*']C2F-PBY@I#FG'&M!?8@I!6#6#`-X0QTR4:Y/-8'A*N$#Z:)C02&<OB M84Y5HA$IF$('A!_1*%Q`G`Y:/;B14^A&PFBF$@T]JG[S)P4Q&)C3<?(,#"2, MPD]YZ:`^Q7^V31'BC;21,MF&Z7A+?%FCLX[0T2[9*,NH>Y+C<]R,0"8V_4;= M`QJOXVQD`4^/H;E%)I1,P"-Z+$Y'2`0<GS#0>U`/="Q.G1'4%*+,^!_ADPY* M`Y/Q/KH5'01]'L/TR3S8$3X>/BIC%IZ@WA$NX6X\8I<*F;>J2[AC2^_%+.0= M?M->_!A2HY!SHT.62*R&(D>DAYR!`5**<,@+F5DT$FH*D6F"!*U(&H@B+>2+ M%`*V<4/F2!*)(6=@D)210_)%%@$-66$(52Q[=#K@"'@@ND@CW<NGD!L1ZND\ MR3/&`F)DBODE1:31*:A?$R6GI.!!-%<R'8%)-"`F'UV7U$'>RBY`22DY=,YD MD4F3J<M->LDJTSB@B,/+#V6R3GH;-"DHL&2?)#EZ,@_%HA@T)\WDH+R3A5(W M+4H[D"C=(H+B/V,24-))*BEF\*1W3&7YX4W.1P7X)W7%IK23JL93#C1$R27W MI)]9E;.'_S1*05DE.XNJ))7E45:VRCRT*D&EJ724M?*PW,HVN2O=(BK#E+.2 M4Q+*-H`ER<"QK&/%$M0XRUSA<'QEH%26D))9CC%J22Q_C:@$-??FL5C+4_DH M4V6DQ"56Q5=^RPK#OL8EL.R4Y[)=(LMHR2Y'5ZF\EJC22L9+>ZE[Z&6*85^Q M<NV<,7Q9+O6EMA1=JR)@TA]OZ2K!Q@MK#^<G6>9+6[DONQ;$/#V5$M2PKX@Y M,,EEL`P76')C8DQ_6656@YQTE[027AY,DUDMYR7#9)"!P@JA3&QI+E=FS.R6 MH;)A6IBV\#1PYJ],F<L22P8-.=`SU:7.[(=Q,E/^3)II,(5F>TB:&;/"I(Q8 MA+G83\=\ET$S2>4!JGG[K.;+7#5;L]>\@57A-34EULR60C-L7@:RR2A)IF*A M1_?28ZI,+`DW^^77!#5UTXHT.HE9,"GFP<R;NF1ONDVP01VA)9F4FUD3+AK- MZ?@>J(+@/)QGLV;2S<;9,@7FW:PPUF,J.!S6R#<_)I[,G(Z3<PY.$0!)H-#, MG)C"\ER63I^Y+E/,MBH#6"AN1LZFF8Y>9^RTFSE3!]E.GTDP/>>YW)V+LQI] MH_.D6'Z+D?"0?0%3A);D*1P5P!,$A6'O4HU"EQ03SN%%.H5(D!-23U9H!&'A M"--PC8PB'L4^!Q;=TAW4A;R0$BJ`4#)*2LDI\2OPDR%T%8T@5C["5QDK]=,C M.`>T@N:6U;?;#??+)725O',#ZDK\/*`(-($JT`7*0(_""U`!3:57/`$HD`4J MSA%``E3@48R!F/$$D\BQR`(*X"),@2J0!*A`$5@/'8,*7($@(`6*`)H(`HU' M\B2!)H`FAD`08`))@.=(`2>0!(*`"PBA(."&,@$08$$QZ!0@HD5@"E@>*U`$ MB,`/!:)40/+PA\H3!(YH"X4"+52).@$32@3X@Q,H%4@@"1S1*9!"5V@+#:)G M-(J>4"PZ02U/V`$!'6,]D%`3"D3)J!%0H2S4A0;1LW!#<^@3V*$]]"S\42CP M1X,`%4@"3\`)/-$R$4;':!G-H_SAB!J!*@!(IX#DZ:)5P`D(H3V*0X=`$7`" M2C2(:E(0T`2"0!8X%B>T"BC1+@I#N^@0F*`^AP@`T43*!$ZIQ;D(06`(O%(I M0`1@J"<]%R44"3#2$UHJ+$\3.*)PU`B(T"HP!(3IEDH"GA24$M-6"DRC*!A% M;`K#"0P!)J!*$>D7C:-J-(@>(R8J2"EH$<V@@>60>M*$$47%*`J]HV;TA/Y1 M7.H$3BD<5:-2`(B^4BC@<XXH/BT"<;24GM)TBDKQA!1X`E;@ZS11."H%[JGD MD0)70(R>T%):>8*H%0@"28`)!`$AP`1.*!5X`J(4A%Z$)X!/\83E(:.*]"($ MEE)10CMJJ0"IU/0)7`$G<%+#*!1(IL-4A#[2,^I.\8+E*0)"0*26"BD`0Z>` M$;`\+=2)1M!KZD5U:"D]I(G4BX)1=VI'\2A/1:8C5`@H@2(P!#+H1ST+2`"& M'H$3&E&CZ!.H`NB4I$)3;"I*N^AZ0`)G%>BH5))Z$6:@!`6E5$`*]"!6BDRO MQBMM`DV@A#Z>+?I!Y6@)/:%5-9X^417P`A3`""`U`F&@K0<J\`7TQ!2H/$8@ M\CC6[S99*RL5N*Q%(+,B@:;R6,U`*0$!5N"8DM:`((GJP*_@`>USH)T2'T!: M.:ME_0)-H+(6`29:<=ZH924!N56)\M82"D)'@&W]K+CU"[Q23?H%OLYDG0+` M5;DFTL+V=3:KB=L=MS6W+M>\FHR>:W3=K@=UB%I7@O9<D^L(U:5)=(Q^UNAZ M7CWI%#BBXQ6=E-?<>DR/0%@#J;^5ODX!^[I2XRMV1:[:]0E4U/PJ72NJ?S40 MV?4+&(%<6EEO*H%=L$.@P8*``SM?E:LJI:R8U;MJUPO[!3(LA4VP%B>,"@'< MNE\U;'-U`B*6Q!Z!"6M<KRN"!;`*EL&NU`<K8T%J;76QY74)?`$SRD,OCG?5 ML3S6XJS8#XM<=6Q[3:\_MK(R4_<*7ULL>;VM.A:ISM/?&F6EP$%MJ$1V"GP! M';MS\.MZY;).@.>PV..J98WL;OVI25:XHMD;^V21JY9]I0-VO;Y9`8O8G*Q\ MO:USEKDZU]^:9ZEK%\VRE56Z;M'PZEWS+'=E`H!6RT+8^UIH8VR$#2R)5L4R M6CXK:?NKG?VO6C;3GMG>>DHIK9KEM!,6T%J!`HM8U^NHA;,G5-0*6C_K74_M M=&VN?_;2OE@MZVH';7?]K;7VT*K:(<!A,RRNM;!ZHL.&UC&+8V_KJ%VT,];4 M.EI&JVH7[11PL,K6V=Y451MBD\"(K:^M]L2F6&RK:NOKI%6VWA;2RMKR.FH_ M+7'-MN;VC:K:(_M>T>V21;)L]LZZ62I@2%4IJ44!\J(,Q`P2D%HS+;VE`O86 MU>+;F1$S$NV_#;BO]NL,W(&P;_NM9ZVW?1;6+EQ]2VS;K+^%N(+VT$[<AGM, M'R[`-;3A=>-67'E[<3\NL,6PH57D\MN.JR<P+J\-MIA5Y!I<C(MLGX#*=;@M MU^367)D[;O'LP56T#/;9WER6^W.7K=#-MS&IY\Y;C%MMK^U^';HEU]XV716[ M<6>NR0VW-A?I@H"5&W6U+-;EN636X]K;=)L%H*[8U;3#->R`W6*[=$TNNYT" M9C?GCMUW^UZK+H554OBAJM0%-(`";*(<B!F^1@R0@+PK.O9NWYUF,<FQB@9G M:09`P`-%K<<4!##6U6I:):N[2`-X(-YB6O.Z:<\M#,`#1B+T&@GKI'2][JJ% MM20!]$Z.U6L&$%+I3:[@M;M^7J@@>J':ZPVN=/>(?M[::R1TU.VMM"!U]X;> MU>MZPVZ`K:C"=_1.CN++=DVOM$V]HM<XV%[C>W*%K3+=O=)7]`+:W#IUL>WG MK2RU=_LNVY6Z>U>O]@6T0):%]M@5VQA>KY'-O>3(_7X!*=M0;93\[;(@@"W) MWT][0AM3I$6U)^$B[-XA\'\3KEL5P+!A/;U>D-M=/<+N;4P+>/R"U+$">@EP M!,:Z)V'WRM((3'8=,.AE2[N6S@9@Z*M_PVZM9;4>6*Z]WEQ+:#UP#+#`)KCZ M9E@$'`.$0+.ML1D8-K3?&/Q\*3`,,`+4%L5:6Q5+@V%P\_T"HQ8#^^`2?(3+ M;>=]HPAX^C9A)8M>VVT*;@Q6%^%6'I&+`KJPUC4!%1;5Q@P?X`/(+:E-NF%7 M[D9<A:MUN[#(!<,)=KGZV9@!`LBP&9;#L!8-'V$UG'%#;AOVP@07!,!A&!M[ MF0`=ML/&M@\G(SUL<<^N#$ZY?_@-A^%>&UK'<!E.Q"\7Y1H!1DQR'?'.C<1? MN,+6W$-\B9'KL:VQG'CS\F%IRX4!,<,5Q*(XZ$(!4FR&I6TJGK6.N/L^7:W[ M*%PQQ1W$IE<7'P%+;(:%\2TNKWSXZX+B0`R,DRO6I<6)^!EG8;1K>8AK*Y;$ M"9;L$N-$K(VG,15FLM<X%&?CW+N-3?$W3J^)=P0L7E+C>%4`"+B\>$#R-E9` MVTRYJI']N1LW`;!C%``"0@`R+KIO5^3&C#T``C9$`CC("6`DX#V_P8^KC04Q M$"E@!R1DCV"0$?)(X`R1L2SPXWP+`B+R01X)%?D@#P;%*Q!(345\5[GEJ&7% MATC@G%U7/$B/,"Q.,C,7I<YB6@19;#$/K:/&DA=1D%RDBZG'LA2FMP@X=TD# M50KS\W[:S_Q)$ICRW$@)]VNMP(09H+QP`DN8&S`@-M@``WJ4N[)7_LI@.2PC MA<?J!EKKKU@LYL3%GI)N1);-LH%`R].DFM03MLQ:>\]9AB>'\3ZL9<=:EUWK M6PXEH\@)[>6V;)?_,C&)K:8$#70C!7!^3FN6.1;(B`GL@*:26J0)+:$J/=F* MH(!+DVEBTA[@RY77#"2`RB,$JL`1H$;%Q7.`@,HS!89`Q8$"'U4*)`!\JQ8X M@U:()V/!.XQ#)H0FX/)EAHNE1P1$9)+,>%&S]:@#N6*'B"+ZZ(1*D5Y2$?3D M+^XA%D"-$')>+@.)<0]-YI&@C[6"1T#(J5EG@(`>L)J)0!6XKU*@E#(!ZOR1 MOW,"N(Q:B#U/Y(MPD+'>[3L_\ID[)X!CPA_+`'O6SPF`<[K&X%-%3@1Y7K!, M0(E.Y@2P9/8#@3G(P9$.Y&?W#!RSPF?9SM^9X@A9_7P<O=?\\0A;8+L6`0R! M6X,`%N@""SK+(.3"6082P!;XJ$O@DWZ!CNH$NL!_=L_WJCM>RMQQ$5[T$XC1 M3F!&?U(;38T`M,)TT3!:1M-H(CV?00!"7CWG!TG_:"4]I&\T?:[/S!D.-&DI M#:2%=(W&T/3Y,;MG%;`?^^-XKA!5@`E(9FJ46A#R@$03-00/]`#+LB\6]#%* M1@BY04J?R4R-=A/J0`''<4?KVP7MI^D`H+XE\V<X)X!"C0(4IJ)FU($9#CSJ M\O2GH?3I&<Y`E#Z'Y[M`GJL>,7&240<%F(!/Z3)#Y8#0T8T.39B``:FH]W$( M\+D8-R!OZA0PD*OS0FX+*,`A?[7S/%8/:>79HDA542<`C*P6W``*V-5?H%<K M5&!M917U8.#.F_I,>^HL.:%BF:AFT4H33;!H5<VJ%[2KAM5N-_>B@%E=JP_R MK98#N1I-(&ME_:M-:+->T,.Z+Q3K8VV>DW41\-6?U%U+`6?=5!+`XQT"G$%+ MD(]3;3\2";S8(3/`=B:,R;N?(:-X)L_LBUQI"=,EP[(""K#/VDIS)4U5':%; M=>,=UPZ;4Q]HC,I-6VBY;M@-$5WKZGKMFB</%PVMP7H'3-B#3*PU,K)NV8:T MB<+L=TV-!H,^_M@1>CR39QYJB$'`9S;7L`Y7K^SS+*(Q1,RNSC7;6-_L1&I" MG3;/3@`^.P%0H\<+!3;'P,XR#`2$_)J_<!<DQ\"^SF=C8)/#C[*P&VO#A@\/ M&T4<0(HMOAK=P)U#G$);(VK^X[$?1;3N`2,[HU+2(G"RSW6Z+L_G68E2`<SJ M!![/NY[9\3HC2^UZK;@9M^/>UPO:9_]MN4WBW-=UK=CC*VG>[<^")C!VYH*= MJ:MO@VRX+;+_!,D>W(4[::ML=4VY[S5FW:9*]'%#;7EMLVOWX@ZMN/MI8VWC MS+K/]`LK<>@*=),O"C*Z,47I]E[@JV@L[T&=J9WT/O[;@;MD$VZC;:ME]^%& MUI4[M(:=F`VYHS:]3MRV6WQG`>']K+_SYC[>GCMYU^T]U+SS]@[)TL.9.U_O MD"VT7;?@-MG<&VFG[._MNS$K:_:DNIMF\^[)C;Y_MQ$XX.R[7^_C:5G&2K6# M"`%#.TT7[:.=`$I']'@4%-QG]N,,KJ;+M7[>W-,:5./J4?TL'1V7A-P(N4EW MZ$:W!>*TC8;A%MDCK&K,**Q_]J-XU6UW[E9AN$NN`SAX]MY+VU[C:V:-N:NS M`I?<YUN)+VM]W</;]TDXXL4;8L_M[X;BC'6$1A-./(=?A!D^?VIX_M`?72!_ MX^C]G<6S-^PVX@(<R"EMVMW`,:O%(=]AW'R#[_2]?6CJU1;)3CQ.@X`50)[# M2W7VV7T`!.@%1<+!]W=96QGTNK")-3TJ`H3S""<),6D$:W+NO<D[^4G(1N0Y M3L/KAEW61MNAGB+(D?^8\?R!!VXX83-LCT=86W%/?A$6>1FHWK2<@Z-L.3Z[ M$?=])6P\)Y,2`5F.RZ_T^][B+K:+HX`XK;=1N8=6Y39<C5]I-AX96_>"_=_; M6Y<;[B0>OOLX'D?(>KR`A]8[_L>SMB`GY',C<R]53<V_M7CG-EW@S8M;:.=- ML*T*J4X1TMQZ?VSL[;^U=^P>X-R<CY-SS%V^%W@4[^8#G5]O;7=L/+Y&W0@/ M\R%O%`X4$0;@0,'@+KN0#?@%9Q$^:,?]EM\?969+WNZ"/LC81Y$^I,-@H",Q MX#7&7!DX(F<&,W"&7/$L$D80<`-Y($#D`3">D-TQ2_\/X2-V;`H'<2695&\. MX;[2;-"%Y'$@#$21=L=@BS^Z'[R!IM6T/*9&EV.9GVG4L"]```\X&ZQ<9J>! M%;`"3+C[;N>6F1VI<'3-PAF=NL3A,KP[/O,44<.9=`P?"3M<"PGK;QW$,ZVX M+N+:'(G3\<'RU::X\'[B\QI94QXB8-C+><_6VEJ;.S]>[[&BSH!$#Q3UH8#H MC3-&,1JO8_$/G6<.@(Q@(2&.1^=I'?R'.G-G%>".2P]JSQUO>H=\QM.#)I@4 M:+<B+`8SRA!%ODET1SJ!'6%PM;OCTXB.]`81L`GYH0Y,#JI.9T[$D"HBOIVD MJVV3GA6D3]NF1AY\=)SR:X<[+S@)W^#=.XM;9HG7>:YU9@Z<7+(ZN^<1;*DI MB'QNSR?!!,!G^_/>F[1'4)CU?02;``$-?&!C?@_`\EWWL,;_'M_UM!S(`_G\ M(//UKGN.ZRY@=^+;?+#WZNWS<X0007?BX9Q7WVL*#W0.^T@^R(^7M*OMV4Y! M'D6;$`TH1(/LA_,3,\Q`H)`;>J.]W_*/[([+TL!0%^\A5U009S0X4L!U1]KB M?5I7/XQ]?AHU>Z3M@MATXV=!S)\/QE[_V$"<P<OJD$VKX7B$]^6]^KB#G?&= MP!%[[];PBWL&;/GU[=BK,]?VVC'^M*_MK-`[7$CL..X@0&3\=H;]MJNYM-[) M4*3ZE1XCKS?G3WV?R)'ZSX]QN9ZJP_A$9N:&_B*8;I^B^S9VHI?M]YFV/_KY M3B`$_7PXC?[YT?/WUUB@+7V9/A.//L#KS2_^Z)G`$["AZ_G1&W@$#Z:=M`]' M`5'>$4]YUEWE`SM`'^R+_;AS>`L?D6%XAB?L7VT&['H/WZ_]M3L.`D_D8"?` M\E&6(PE&2?.X?87H#<?RX^M\W(;?WVM5B@$&F&GJMX2^[8&S8[=Z!;_/V[D; M!^"VGI<3\`;>0K</KY?9&-Z@[_&\.L[?_2'OV>O<23_>(,#4H;TNN0O_PT"P M=-$`+Y[(BI?H';UK@(PI(`[WA8,`#W?@QU]GZ<R$7(`7^D5R"(2<:18][B>^ M8K3XMT@6E?B+0)Z)0&O>LAVV`'T!*M#QH;-?_/B**&EZA))_\G7L-JVBF79! M>_P]!/*MT"K20B1_F!3#[2XN^?:"SNJ@G"2(=:\.UO6'6"?KM5H_7^?L7/$O MOB::0TJ#/)-Q:(ZBJ_=!GOHP/^2?GYF_FFM^RA<[D(?E>_WG/$_N`]4_)CW_ M"\G\H&_R(ZS-UZCOM;*.>[;?%[%SV/?Y6<C^!'UG9'6(?C&9ZV6@KB=X:\^I M.;<H$8=T^Z-X^V9NSQ&_XM_[U/QAG_ML#N&]-[_QY5,@"#!1>CL#K2Q/%^-I M0E7O?-XLB,'^'NKU\1Z*@V_1?Z\WJH=7Y/+,0'SFM6_LMU1?&-BE$P2L`8@_ ML#][@``:R+F!V`4SX!UXAA,?Z5.E,U3[7>YW4<#G5^Q?C?2??OB>^EN_RR\# M%/^8J/[O__8W*+Q&Y/D^<A=KOTWE)_-@N.;^W(A?^7F/6;G\A??R#%RW.O#Q M#9BS]/L4RTM!4I8_A15EA0#FE+D1IDU+$!.`!,_!63%"8`14&0S`E0&`%&`% M:`%>@`G4`[54[5,55!)P0650^!8'5<5\4'74'(58[51Z%`P5!,A0-!0?A4/I M4#R4#P5$"5%$E`=H1"%1N1L3I51!45*40E5%(5%85!*5KRU4MU0[Y4C!4Y`4 M)!5.L5%CAQ3P1L51(]1A54>A@"^4"^A'`5)^U2!52$%5BM12902^4U;5">5. M35*5U"4%`@ASE@<GM59)4VT5*65*%52JU$)%3>U3L=0LY0344E)5$*5+_5&] ME#9%5@53.54QE9XA4W&4'-5,702=U"<52K6!994PI48!41;'-M5-157@E!2U M42%4)Q0'>$Y5"&K5.C55'8%A($-%3]E3)-5#I4_!4DG4-=5"!51O($%E5[%1 M"57ET47-4Z6@Y1%1A5(4U0DE^I%L&Y5+]57!4$)5+=A0-8$H5?<`4AU25(`O M"%.-4C*5'V=))0$W52$83H&!\50D=1$\5$"54)57%55'E9751"U23%67E9X9 M4MZ45/4%)E:05$^556U57=5+!5:)57^@6856<8+J%%M%3;U5<95;15`55((4 M7J57R8%]E2`%6`E6^5IA%06:4-1@'K58-5:E5635>,%<291K!@UJ5DJ7L-6: MO6::%6AV6O5;JU595IAY=9':;*5Y&0AU'\H'=J!]7X`64/8-`22`CG42CATI MX<BU.Y2$+J'*5YJMA"WAV0<3WH04%DVH$T(>5T`5@!/6A"AA4"@3DH1FWTL( M>7!30^%/.*,)A3UA4JCR\2!-H5+X!5"%4:'=YQ2>556ARL<59H4FH4W8%:*$ M0L!1N!*&A47A6`@37@%FH4]H%0Y1=5].^!:VA5(A2A@%J(60QUT(%A*%,"%( M%1?RA9"'C35NN84JGS+U%SJ%RM1>Z!1V48>A51AK<5:$(4JX8C6&*M^0-1C6 MA3"A,$49HH3"E&)H%1H!`L-F"!."AG2A5N@8AH:MF5RH\A$!`H-G6!FBABPA M8)BLM8:7H6FH\B$!L*%JR!G2AI`A9@AY#`$E%%ZH7`6'KB%*F$&)AFE?:8@6 MPH0$B'#8'!:',*$3(!0BAV$-5%@;+H>0QQTH'&J'T&%V:.IMA]]A=QC6-%34 MH1.`95V'LJ%YZ!>FANGA0:4<IH<J87FH$HJ'-Q5U.&VAATZA%>`<OH?YX73( M'O:'_*%5J`7T!]3A@!@@JGQ:0!'0!`B'">*"*![6?<(AA"@>"@%0@'Y('5*( M^N&$R`18B/^A5<A198CXH55(>1R'':+*-R*&6OCAS9?W'5&'H8J8:?&'+F*' MU13&B(\ACN43QHA70XN(]VE95X-B&"/6A__ACP@C[HB5%6,8)!*)E]60B/.% M-3;4C$@D:E,]8HI()%H!+.*1N"12B4IBWD<EYHA6HI88^F6)6E85(`14B;&A M<D4DBHE-5F^H%<:(8B*76":RB4)`E*@FHGPQXA10%NJ(2Z*=""9^`6CBG=@E MAHEC8EE(6JUCC=?CU6]==>O3<9:<Z2%96BD2P#5H)\*-<1`H9X#?@+2@08HW MWF%`*7XA(E\9<"F6&3I&IJ@TQ'R32(J&XTE>5A\P@O6Y?WR:,X*<I0[+V3%! MBF`&HL2C""HR&9(BK%@I8D:?XGR!99R*BDBGV"N:&2?"8T8JYB*FHJ:H`J2* M&5];P"HV%:[BHA@K\B&E2#%A*_J*^T&NN"G^?/;'L!@J&HMBW^GA+?Z*FN*Q M:!<DBTK#LOB&7'T@Q+/8$T:$KUELA@+X>3P>\Q`S1(NZHF(T*RH-'QIYM@>P M&F6`Y`"R/2-G0`KP8*AKON'*AS"NB4@BFE`OG@$C66EU6L%C(^%9*!NJA#N8 MC9@PWH3VUYPH&QJ%3!AY%1G"A$PA!.8Q.H54(6H2(GJ%0F&;,2%6).(A6TAZ MG8QS(?,E,B:,=Z$"1C-ZA6G"J;(RHH3*U'#S,\*$7110-#1"'BL6B"$>"E/G MS]&H8`D,>XS3R!H^9>+A$2`P#"5.(VZ8?V6,-Z-MB!("AQD4(>0T9E`8B=/8 M'%)EXJ%T.#>HC#LC2J@=1CA.XQT(4K$V<6-#!0.EC>[A$Y0VJH0VB=-X4PTY M9F/\U38RAT(AK>,T&HA)SN&H(*8)(:-\-3)"'A#B#>0T8HCY5\>H,7J-.^&& MF'\UCC-APG@BZCH^(I(H.%Z.="*1V$59CEUCZ;@D7@TF(^EX]RV)-Q7;Z#J: MB7EB%Z4SSHXQ(I2(E86.5^(1932>C#'BEC@WH(Z.H]D')_J.G"-2R#`NB6WB MW-`ZIHZOXXI8%LX5O&/>UR?.#:<*<<:./5[P&**HCI5D9D!H\<:E>]5?0.?` M)73P'C@G[]5K\,0"8DV@`1C@42``%H`$8`;FE.5?BT'_9!LX&\[&)[$18`3. MA@PP`<:/!&0!:4`>D#Y!8R99X8O[0J<8P*EH*L!:YZUI!?T(KYA%3%$%")^F M0(IF#.3\P78,*61!`-=!RDH.2%N0`)!I(P)KL*"-D&M'"2D'G)`;2-VQH'T= M!0A"%D&^,NY?A7#8`"&4A^6A09X>IQ4#67J,?4>;BG9"RG@3)!UPD%%Z:`"Z M>)"1:;E2BI!$+I$"'G_D>PQHL-&"QIF1`4_DI^<IWB0XWD&VZN60NUX&V51L MD$+?JWB_F1^G1P#7IK$?`.,4N7\(3!9D`@!!SI%?9"(W@)B1C!D0N4`JB@Z" M^_A8N)"?&;0&2((`@N0'Z7:\D"@D@*!">GT,9"+I0L*0#`0'\DAR9S0DVF=# MBB`+6AFI21Z1G<>"MD=^DB=D^9%#\B`^".0A>?204@"?)FLE`),<C7:0N8Q- M!2KY@ZR20L@)N=8].I.9+:E*!B&6QPFY)OE,O60/<DL"D[*9"@`UG3'%9"K) M0^:2RF16X"=U=\WD,<E*EI*3$B^I`/B2SV0PN2P*:DI3-?E+7I/+(JQ$36Z3 MQN0X"4V6$HB2!\E,HI/.)"[I3;*3OI(XV4TV5*5D"Q=.PI/6Y":53W)+]20_ MJ4YZD^$2,2E0WI,GI+QD..4']J0\F4P"3.<D-^E0)I3\4I^G*_T:#24R25%: M3&(`QX11'I03I0H@,E$0VJ1$J5&J`"R306E2DI/:@\QDP664Y.305#2]E"#E M28DT32(E93J)4*H`4]/ETC6U31_E2@E-]I1KTQ)"*;V30J4W693EE/'D2;E4 MNI-!I4X94F)K3&4_J5123H\30UE3DI.@T^9D12Y,6F52F4RN3@&E6'E"[DZE M$DP)30)/-.5N54,^D:;D!2E6NI+?#2QYV,B2"0`MN55"D[LD4BE5GI3#9%GY M5Y*3RV14V502EM(DJV186I7)I'=4-M$!:N4\"4Y6E0-E8VE.ZDU^Y6&Y3EZ6 M@1-4&58.EINE`B18:I9IX#]9P2V4D.5>65HNBV1,Y>3"+9:5)2B9+M&4*Z4W MX09X"@BD43`_@A5*F0%8`'H$C`&=()7%!O<+8V`#W`31Q0"96RZ7S&5S20%B M"FW!Z>"=$67G'1F``F!JC]FRB+GL3C(?>5;)36;99?43O=5M8Y]W*0*`EZ?B M*S?);0&>3!=PIHD`?LUYJ0"$EY@+HY>ZO)?R965FCH@;;,"L(FW,<YS"F6;( M*0"5F6VYS'EGA\$9,`:$%*<B:Y=@V@$,IK*H`H@&VTJ@X`9X9M5;98;"X7G` M67F'J5UD"UPZETI09M,EI6:H;9$8)H@IN8F8(UEE-JWM>2BF*E&]16TL)HF9 M.C`1KHR6%*K%F(6D1T!C&G(M9HDIC]02:]UFAFF0`2FFCQEB`IDV9EI72\"8 M1V:2>1'\F.Y?DRGD73U9S^EA9'9F,J:*N?XQF02FD/F;Z7GE'8\Y8RZ956:8 MZ2!$;`&;N;+F&6MFII=9%M28:N8P,<#%F4KFB@EF%IC('$B'NLP?6R:2V67F MF5]FE<EGOG-O9NH2:$J9ZX.>F68>FNU+Y(>Z%'EXYI2)9HZ8D*9`XV>.+_1; MI=EH%IJ8IG>&[7%QALF!F4,(!!%FNNB,,)I49JBY9LYM"=`"%*$LFH.FI>EH MNIJ.WXK`%TB:<&:466M^FG,FF*DL,HO!B+LX&,*+T*"\2"]:#V3#O7A(,HJR M8G.&4?2+1AO`*#`^C`:CKI8P4@$+H^JX(CJ,S";S$#&2&A.CMY!YB8<8XX38 M,L:,A:-X6#**ARFC>,@5NHR3(\SH-,J,[B8(8#,6CYAC7N@FR)L]HW@8-'*- M_"9V*"/&`/#$=_/_.9<3P6X)7/:63=EOF2/E",<E"/!)O!5$!72`YE0D'X8` MR1$PG"!GR"ER'E"/5UZP2()VHX,D,0-((MG(MA!\^!I'1\]P&?"7'QW8PN9Y MCWW9:Y4?.$MO@$BH$4I6'.&X-:(14DZ@S+D]\$>E!(4U0Q&=&=1^R1\E0(I7 M[D=A#9U_5`:E=#8!3*<BQRE$E]IC^`B:.80)0%[5@QQ88><_F`#X7H3BUUE* M)``(FA)%=K:=14`"T'ZEG3]GXT4?&$/C5HK`+-$FSUFB,-ITG4W%XV4$?`N' M!&>E-V00[T'RX"A`$I&BRB`&6`9WP>3U2I926,#*-T%55DE`@HB0"8UU)>4) MRT$>4T#F&7<>9&(C9Y6D!6EX95ZIL8Q;NYZ7AI!Y,F1GTT9949ZOY^IY7.5L M@Q5EE7HZ!Y(C9Q4$)E1GU12`D!&<NP--A2'$<+X79&A(!0$K'TKH>=)H1T`4 M=9!QGC@6T1;#$9\&PA%@ZL6)3`#R26&9>JC>]4EA>9[-%`N%>6J>I">(<1M\ M6'"5$XA;61P(&=`38K"?56>)A@4@9".&2D!A!54FU/OI!`R?^V<6T'_2GK.D MQJ(`['OR#/J`8Z4,;<#C*<_L=IL:ZW`Y0'S7'>OYNK50J!]]AGUZ4>Y:X\$$ MQ%CTU@=*?]EW3AKQB%WE7AFHDY9\%ENYU_R%!910W)^3]F`HG4[`185#%3:] M%+W5?(X=6QKO28/:H%^'@L5]%#:AI^;)G04V/V@?5=CL>I-<#WH<)*$W:(<% M35%63>@1NOY`H4%H[GC8]*#4)WEE<0"A1*B8J(4^'ERH"6H@!!8=EO?R;'%5 M.50L!=_EE3D"A86&J@T-%AM*X1$!W!\<^CS*5VAHG3AV`(=WZ)9F7,JA@D56 MY8?:H6XH=_81C5MH*`ZE)Z!^>647RH<*%H]';B5$G7IY:&QAAL)16I5RU4+E M;#%61M5$;6DJR";Z!'2BAU5NU4+E53X'$Q6*=E0'F).F@NRAN\,IJ@0@83[' M%?`%K*(5QYGUBBY4WYD*XC-R5E0G3;5%?0&8H!1@!)AZ5\"61D9,G5@`(66, M4E9HH#+*C#JC5@R%E8PNHS)5'KH;#*,X5C7*C3:CV.B^>2DD`%-`%C!BT:$6 M1PGUBFYI2\4FFHZNHR/6%ZJ$1AY/P"/*G2T5M"B"I8YV6/5H#7J/0@$XWQ7` M!\*C3VC/]8^RHP)I%$J'5J'?V5)ACOJC].B)!8:N?+U@=GC81*22*':UD`:D M%VD<*`5\GOMHU&A;?:06Z3V:`UH>)*F3YC^1G?,H0(J,CAU!&OP'C\*-=65, M2H?&@>_H33J1K@<HZ8C&5;VC^VC?>))6I$*`Q;&$DFSP:.)XE,JD:&@+I5%Q M@<)G4=J/_@FBZ$PJHXV@6QI-@(4]7D0`CJ4W,*`.Z//P[,4.19U``#E0""P= M"-!2J@C2GP+`VOT)A.>WIS<L=GL#L\2DN`Q7W;BUV*E8']GTF9516'\IZ8=\ MQJ$'*$,7.QA2]-8AF)8F(7TI9\68YE)(`$;*4<6=KIX\%'@JI@8"!YB7&@9J M:62*8W&``&E@F@"H("LH>46:UE<)&>D9AXY;I"GIUU&UIJ>I"L99D:8)51'0 MC+IZL^AF&C)TII\H0OB81IX,X6T:G$(>;-00@)#M!JDI.I$`X&S]IUXUFQYD MS^@K"9U"'KDI.9H`[`8@V.-E/)"$/\>]=D(-IZ*IR/ASE*:TZ5+1G,Z$YREK MFIZ^II#A>9J<(F1+1?*XFIVGLBEF>I#QH^_B!YJ=OJ>RHWGZ@5Y1G10YF$B] MI]FC=QH[.*(9%'DJ>=:5#"IZ>I#1!.MIW")&S9[[%4)&$VRB$6I^JJ':ICA6 MA/J?3JC9:&+ZF[J!-Y6#6ISB6*44%(">NGK*2X6*HEJFLRF,"I^RJ(0%+JJ; MTJ;*2W?*F9)2`E8.=4JIJ..G@`68TJ;&@8YBHGZG^"!;2*3V7'"5+LJ:NGI) MA(QJ2<E4-*I^F@`D$?)HE*JC[J8'61)AGUZINBB!"DT9J%]4F%JB*J@OEE[5 M#CZIQY6;BF$)`9MG[T5DR:E'ZMEI),BC8F=7E:76J2!JFX6GCJAZZIKZHYIZ M1\`M!:>&J$^`?14'NGIGIUSQ?3:JN!6=M7GZH$OJFS?.T52?Z5Y:GCJ.H=4N M6@1``04J%[AY:J:9JH$PH@6'C^FGNCNLJI357U5*:5+S69B:N*PG?20%X9B% MD>@H*WIQ^&A=VI+V0^JJ`PTIIT:*'&\`R?%`AI%0I!6Q!3@'75\BUV7M>A)9 M4[%T5IWOV*WHH"$9]\K(A!F]&V\`?#"AT6<<T[=JGXU#C<;!,*XZ:>6J%B(V M,0_K:K=*4@Y(H@2&<1#$J^VJ_;$`Z0=\VK5:=!J12A[&)`8$!-+'EK8%V'6+ M&HGW6-`/`VO:<>`9K$R:?A:PDI0#JQ)YD!VL5IJK-[$^%BJ:?H:Q%FGNV<;J M4[9H'FO$&K(JK,5$NDJR?F<?ZY:6L$9Z).6?,46<`1!KQEJ?H:QS1KBJ1):L M-NO+"H=P3-PJSPJR7FDB:Z[HLK:L$BO*NNWI!T>KR4JTHJST0\?*LCJM&BO* MZBX,?V_`'>`&U*Q#:]4*LSX6U`.VT+3VK"+KQ4!XK0Q<:]=ZLWZM\<3WQ[6Z MK"+KF\F=?:QJJQW)YF5.J07N\]+=E_09TNJ>-0[.2!B0MW*7D\C42K9Z+_9E MTC2VUJW5#^&:BRRN<&M]F?ND+I!KTGK[.*YV0>7JGGF0B>LDHKE>:1XDYCH' M?*ZN7GYP*=`!WE'G^KC.K50KTG:Z>D>BZ]MZPM$!IZL_L@;HK8U>X6JXUJVF M:U]@NYYNW8KNVK<2:8RK]R+CD:[UV=7SN9&7IX?L&K)Z+]?9](:\+FK0FUX@ MO45IK.OA>J!,;]^++S.]#GEG2,0T-8T;::O+VE.*K`TK2K>[FJ]AD\C*L!*L M!YZ3YK<:<^VKTHHZ.*_T*]?TLV),4NOPVK.>KRCKN8J_EJ[U*]MJMA87*X/\ MVKKN9P7L_DI2YJSPP0!KKC6PW&O8F@=(L`RL_LJ]MFG9*^]*P8JO`(<_XB`H ML/_K!XLQ^04X767FOWJP&FS$%+6>BBML\9IMO'@6`T(VO[IZ(D5`4`=(;]K# MRF"Q'JS]*M:)K1J1YDB>-'_DH<`JE5:CO7=%;"J7(B"QIZ>7AK`VL8(:%#NE MH9Y5FA-'Q4JO!FL4NZ0QL9<!HA2E=;%7K!0+QB:K"F6+1L8&JUDL0E;$WI2Y M2*.SQBJQ4VP8VU7*L7,L%KO$:K%A;"FA.6ATX0;\R@"Q1G.K%]O&'F1-[$#S MQQ(3[ZO#.LA^E?*K(;O'NK%]K"*+-`2RC.Q7><:>*XOLK$(P[AZ0+,LJR79] M"H"_FD$]:`E`$<M$;K)^["6[B=P'4,B`Q\<FJZTL('NN@K)FFL$J`]2QR>JF M1Z"Q!63`*%?*"K%%I]&:RE:R1P,@FROBLJ9!$)MU0K`6:R*;S!(3T"PS^T4. MLUFGS'HP(++(;">KS2ZLFNPLR\E>LM]L*V/-\FG8++:JH5T<7]_H,B'``2XD M$)O.%IUMFEYB+V$&G8,\:\I.5A4'.VO/K@KQ;->WSZZS1\!!1H^\LP'M9+;/ MUK-P$SX+'^BSQ&P&1=`:M+=$0CO/2K2^:D'[SSH<""T(V19`M%EG/<NU.!P= MQ4,KT$:T_*R0U<Z>`8&+1RL'@+3J;$9+T78-'&W%8-$NM!1DCD;34A$.+1T` MTQ:=$ZU.>S"TM(OD3YM!U;-R@&[&T\H!^>Q)FW4&M4FM56'17K0IK3^[(/27 M'6U1Z]1BJTAML]!?EK0^[58+U,JT\9JSH$B2!?(L56M$YJ?SUQ9%7-FP8BU& MV\]JM!\5/Q5F/0$QW!;@>=:UYV<1$-=6M1JM"E#7.I0*+4I;SZH`3I4A5=@^ MM62M"D#8JK6\J@H0)SH!?MQBBZUBB@\:#D5358=-0.EGV1:=F"VW>A$6`3): M1`5V(`&?;085VC(/Z&C+)J/)4!]@:INM8HLGPH-&^Q%58Y5LN]K2K$IION;, M#K&\JGJYA5ZLHZPK26X6JT%GSFD@P%;AYAD@$JIC4B=AYI=Y=0^C:(<&T%;? M8W%69QINF)J+R6&2F9K9<.;=#IG39'@[F8VWOUF1*=Z*F>P(UW)=GK?L+4_6 M-;RW=29UZ8YXF/#MC2F/:!,,#WV+W@Z3Q]'R*OFMM_JMM,$?/0SARV,A#KT' M]X%_*V8V#D\#,3$AO*MG@(-;X%8&H\V!8N%:9A@N`^3>$K@<;L7``#D6&RX\ MT>$"KR4N$R&QF2LM98H[2ZRXCT546P:XN-)&F[FP<BLT[G>3[3D+?T'1D./J M!<F3A@OB)IQ`KKGRX>:WEME`<#"PN#'3C[ODQKA>;8\KOCBY02Z).^2F94\N MBGOESA)*;I`K2-*X48.ZM^6V,E:/3PD?:)EC+H0"W@9.15Z:6[KL>6[NIT;I M6&MQ+L*4[9D!S%X#0>-"*/3(LF>;Z;EC;G1"XT9UOD:K,>A2!6T!,?'ECKEG M#(U[Q@BY2"X\\>@>N?4M$_'H6KF1KJ7[:VBYF>XL452*!0.!HSLV?0&0!(UK M8(ZYA@$<`.E6NK-$JBM*;$WCQJJ+WKJZXVO0T-^BNA7=K=OIQA.YKNUDZHX% MI(:'*ZZ8NJ-+]3?FLDPG0XE`X[ZE<M)C2N-B9&M!#%+TI0C/;E\0[>Z:'J3H M-N;>&+>3WN3GB@:`[J[+[9ZUN-JVBV%@(<@JC3ON^B,SKKF;M2),ZNZYN^/Z M&H#!E.ON8B&8+JM;K[Z[ONZVV]^9`0Q0%8*N^;N<'@,$YXJ[_ZZPFS#AM_HN M%@GP6K.!K".K[B:\$&_%.O$6O!#ON7KQ$FB/;'^4L@(:[2["B_%R3N5LA4OP M<KP0+S2[\0:[$.\K6^Z*O"@OYV3!KKP/KX!FKQIJ)R_+*Z`UMS1OQ^M['+#Q M`227\]:\7R6R81@TN-LN("GKBIGX(KR;]":UC$Z^B]XVO?VNN`OKVKK>49N[ MZQ((@0(60DU\#R&OOGO'+KJ[;AZ0!QP,>`"-:_9R$4&#VIL'<`:W0J[@]K*] MCT74@!E4NRE#6<8F!01PP.:`]Y8N[.Y_Z2[@O$BNV8OV;KAFK]T[\$YF;.L< M$,*JL"[:`@O`&K`!01Z@PI*P+*R9R[V^L)IB#`NW>B\];`U[L2ZP.:P;L,/6 M;:'O#TO*[K-$;!@+-1VQ>:P9*\XZL6IL(5O&?K&S;Q4;^^*^E"PM&[XVK[OO M(7O,)JO':_`[R7*S:.SH`OO:OFSL\3O\BA)/$TYYQ#*_=.PF>\<NOZ+L[2O\ M2K.=;/IZX%FSU"\6BY=]?POGR"D_803T$_T(<9(5`9A<\6&T!/OCC_`<,`9= M13RJ7)J_^&_^J_\&!8-AO>9H952>%Q:PQ5RER%I@,0X.43``%K#^Q`32EXR* M_1$!WI<"/``G$8&JXUBO+7:D'TD@`<<$%+`-UO^>9Y-I,Y4!)\`+\,63E=FG MBAT5$((R`2/P!FP"UV`](<LV064!&18)/`'G""BP#$Q!1<`E,)?J`4.&+-MQ M*IV>4#8P!UP3Y,!,VW&:G6K`/7!-\`/;B/5:Y='A75\N<!*#!`M;3``/?`/3 M`$^PR!@%_QPM<!,<L'S`;2T1T#%L'RY5$6P"UP!=L`5\GK6H8?`-C+:0P8%% M?17,C5)J<!)A`[3!,V&]EDYM5'=P%Y4'-X`Q\'F63@7"3/`-?`/PP20AY1:E M:L%&<!A#!I.I<7`DS`C[<KF>L#>$(E6)<$P0!(1>A3"PIS9PPFKP)WR%DL&Y MGA`P[%E9G3`-\`EWI(TP"/S5K,(-52F<E?VD#W!C5PN[P)^P40H%R\)@A^W' M"MO"32,0?)Z59AD5$3#)D<*\<$VP@4YM:AI7Q04RPR7P)_S\A,*R)S4\`'_" M'@H9O%Q9`:;>Q:$->\)6#"X<!+ML.ELWN`M7PZ;-,_P%3\'CL"MLVIC#YYD4 M[$O&PY]PX$,&TU^M<#Y,#[^B@UL_;,Z$PEF>F*<"DWGK\#;\'/S#!?$#9_\) MQ%,,&9SE?5;TWC#<#.<R$C$?%[Q9Q.SP-9,1.W!C'CZ,\##$?!P$QQ$KQ)\P M07ROM7L*U4E,#@<!)/'YZ,<EQ.2P$.`.@WF,VV$C$@L!,3%FU>Z)Q%S-1XQ9 MM68=E4LL#P\!/7%H5822MD;Q)TP$W,1*7/V']HG$1$!2;`38H")Q$0`59WD1 M80X:!&3%/7%K=E8Q<0*Q$;`5;WB;<%/\@\7$NG`_7%FHQ%1`??43-\6P#5SL M!`@!WJ`Z[!;#PEC>2KRST<2N<+X1$Y-^<S%@'`1(!G`Q89RF,<6&,7`1"K>C M^II;[!21P1;'%'`/T\77,!G,H.+#,0%4W*+.HZ,P76RLD,&?L3I*^G'&*P\= M+&#-PF.Q(,P+SQ7_L`$L4R'"I?!,`!4'%B(I;?P:KY^J\4JJC+;&;K%TI!H3 MI.^5;EP-.QNQ<?!Y4ED>Q+'N]1I'Q,:P5CH3N\48<72\C3*CG+%''!T3I%*` M$G4%8(/9\3_,'2M12E1'U55QQBDQ&3P><U7'*7HL'N-\%@<4@%9QQC9Q*)R< MIE=_%7W\#Z^BF!5G'!1'QT.P?_P/AWXZX%_L%C_%H3"!7/N5?HUQ59P@:\0) M&F>L%3_((/%!+"$/R"5Q1,@9F\44,DMLX;G%1@"&+!/CPXP!5-S--6XD\BD< M'7=S1#&)S!?3?^J;BRPB8U98<5.<=YC(Z5MK!G:0R,7P+ZS_M7LQ&D+<#\<) M.')>Q0*W4#9R9KPBWVN$<5%L&#.`"7*4^L`5QD/R:+PB_Z&A\9.<&E_)7!5C MA\!IR0/R']H6E\*V@8G\A]97)#)OS"4/`7:P,GRY#<G"\9I\&MO(-\"`7!$_ MR="QC[P-@E*H)(E<'?O(XBAV;"-KQSYR=DHBOT*A\)CWA0[)Z7%TK"CW6/HA MF0P3V\?PL1,@'Q]2->B07!]/PE&J0TDB\\11LDS599'(_[&/#!HGI3P4%0#_ M#<E(<8*LCJH-7RA<I2<,R0CR)/PJH\I60'-,*P_(M_(7*I(V;D/RA&PK"P%J M0QL59I%LP3*O3"R79DK4JDPF<\C#\BB,!7!5SC(O+`.$R*[RLHPJ$P'5<C6L M@IC(M_*C'"GSPBJ(L@PKUZ"D'[!<"L,$X/*R;`#_'/TP3/`/[WI+\:*\+MLD M7/%5?$-EQ6?Q[Z8C+Y]@,5Q<]XW%D;$M#"U'QQ->6FP8K\5P\9C\&DMA7C"8 M)Q<;R+6QBNPC]VIW<5[,&;_(]5JO1B77QG?C1TP84\ROL2\<,2MQBG%'Q1GW MR"<S9,P93\;1<65\&3?&2K*-N/_>S#ASSJPS[\P\<\_L,__,0+-=`0.TRD*Q M4BQZ4L7]LE0\=E#%5G&-O#`+RPASCCP$>,4!LU`L%F=2!7,S?#!?S&AQA2<2 M8\L?L<-\'$/,;K`2-S'KQ15S3)PQ?\RO,<<,YK'-QW'('#6KP!L5W+P-(\8B M<]V\&*_,_[#+C!E#Q3)S.ODR_\.;L6CL&1,6IS)G;"7[R*6Q$#`G-\9;LH\< M6`!SP'%M#/>HQ@?5%6`<W\WZ<'2,&^=NF[,G/!,DQ[YQZ!P8Q\F2\P2%\YG. MAW&='`I_SLNQ%*`KU\9Y\LEL+'/&?[+MS%N-H]DQ5+P>?\<_57A,*7?'0J!Y 6G#4?QXVRC[P>7Z?N,:7\7L7'\S%'#++M ` end
Newsgroups: comp.sys.next.bugs From: colin@agincourt.bsd.uchicago.edu (& Garrett) Subject: Wherefore this group? Message-ID: <1993Jan27.155027.9777@midway.uchicago.edu> Sender: news@uchinews.uchicago.edu (News System) Organization: University of Chicago Computing Organizations Date: Wed, 27 Jan 1993 15:50:27 GMT If someone is not sure that a problem they are encountering is caused by a bug or not, it would seem most appropriate to place the note in a newsgroup dedicated to the subject (c.s.n.sysadmin, programmer, misc). And since 'bug' is a catch-all category for any unexplained unsatisfactory behavior, until the problem is resolved, almost anything is a bug candidate. (What percentage of those "I think there's a bug in gcc..." notes pan out?) On the other hand, if something has been proved to be a bug, the most suitable place for a note explaining it is probably a static locus like a FAQ (preferably archived), rather than a forum for dynamic discussion like a newsgroup. Maybe that's why posts in here have been dwindling? colin
Newsgroups: comp.sys.next.bugs From: gcolello@biosphere.Stanford.EDU (Greg Colello) Subject: Re: Wherefore this group? Message-ID: <1993Jan27.185219.24898@leland.Stanford.EDU> Sender: news@leland.Stanford.EDU (Mr News) Organization: DSO, Stanford University References: <1993Jan27.155027.9777@midway.uchicago.edu> Date: Wed, 27 Jan 93 18:52:19 GMT In article <1993Jan27.155027.9777@midway.uchicago.edu> colin@agincourt.bsd.uchicago.edu (& Garrett) writes: > If someone is not sure that a problem they are encountering is caused by a > bug or not, it would seem most appropriate to place the note in a > newsgroup dedicated to the subject (c.s.n.sysadmin, programmer, misc). > And since 'bug' is a catch-all category for any unexplained unsatisfactory > behavior, until the problem is resolved, almost anything is a bug > candidate. (What percentage of those "I think there's a bug in gcc..." > notes pan out?) > On the other hand, if something has been proved to be a bug, the most > suitable place for a note explaining it is probably a static locus like a > FAQ (preferably archived), rather than a forum for dynamic discussion like > a newsgroup. > Maybe that's why posts in here have been dwindling? > > colin I'm afraid you're right. At the beginning of this group I wrote to the creator of it asking that some posting rules be established. I thought threads should be disallowed and instead a moderator ought to post confirmed bugs (with solutions) as determined by discourse in the other Next newsgroups. Your idea of a FAQ would also be required. Still I would like to see the bug and solution posted once. I think the problem here appears to be the lack of a moderator. ----------------------------------------------------------------- Greg Colello Carnegie Institution, Department of Plant Biology Stanford University gcolello@biosphere.stanford.edu (NeXT mail OK)
From: louie@sayshell.umd.edu (Louis A. Mamakos) Newsgroups: comp.sys.next.bugs Subject: Re: Group Problems w/3.0 Date: 28 Jan 1993 01:59:35 GMT Organization: The University of Maryland, College Park Message-ID: <1k7em7INNn3p@ni.umd.edu> References: <1993Jan27.135357.3950@nic.csu.net> Keywords: Group Netinfo Problems In article <1993Jan27.135357.3950@nic.csu.net> vqueved@nextlab.calstatela.edu writes: >Is there any on that is running into the Limit of: > > A single user cannot belong to more than 14 groups. > >I send out some email about this and responses indicated that it was >a problem with NetInfo only allowing 14 groups per user. It is not a problem with NetInfo. It is a compiled in limit of the UNIX kernel. If you look at /NextDeveloper/Headers/bsd/sys/param.h (otherwise found in /usr/include/sys/param.h on sane computers), you'll find: #define NGROUPS 16 /* max number groups */ If you RTFM 'man 2 getgroups': NAME setgroups - set group access list DESCRIPTION Setgroups sets the group access list of the current user process according to the array gidset. The parameter ngroups indicates the number of entries in the array and must be no more than NGROUPS, as defined in <sys/param.h>. >Is there anyway to modify a file/profile/"sys_rc" file to allow up to 35 >groups? Nope. You'd have to rebuild the UNIX kernel. I'm very surprised that 16 groups isn't enough; perhaps you're not using them very effectively? Louis Mamakos
Newsgroups: comp.sys.next.bugs Subject: Group Problems w/3.0 Message-ID: <1993Jan27.135357.3950@nic.csu.net> From: vqueved@nextlab.calstatela.edu (Operator) Date: 27 Jan 93 13:53:55 PST Keywords: Group Netinfo Problems Is there any on that is running into the Limit of: A single user cannot belong to more than 14 groups. I send out some email about this and responses indicated that it was a problem with NetInfo only allowing 14 groups per user. Is there anyway to modify a file/profile/"sys_rc" file to allow up to 35 groups? -Victor Quevedo vqueved@nextlab.calstatela.edu (NeXTMail acceptable)
Newsgroups: comp.sys.next.bugs From: lemson@ux1.cso.uiuc.edu (David Lemson) Subject: Re: NFS mounts -- time stamp problems? Message-ID: <C1Jxtu.H2q@ux1.cso.uiuc.edu> Organization: University of Illinois at Urbana References: <1993Jan25.191725.23058@imada.ou.dk> Distribution: comp Date: Thu, 28 Jan 1993 06:53:40 GMT ravn@imada.ou.dk (Thorbjoern Ravn Andersen) writes: >We have a Unix network with Sun's and NeXT machines mixed together, with >home directories on the NeXT hard disks. >Apparently, date stamping works incorrectly when setting stamps >explicitly, instead of inheriting the dates from the file system. >Touching a file yields a date of January 1 1970, which is somewhat >incorrect. I *think* this is date Zero, but why should that be? Is this under 2.x or 3.0? I have seen similar time-related problems when 'touch'ing files that are actually on a Sun exported via nfs to a NeXT under 2.1. It was acknowledged as a known bug and fixed in 3.0 when they supposedly re-wrote much of the nfs code. -- David Lemson (217) 244-1205 University of Illinois NeXT Campus Consultant / CCSO NeXT Lab System Admin Internet : lemson@uiuc.edu UUCP :...!uiucuxc!uiucux1!lemson NeXTMail & MIME accepted BITNET : LEMSON@UIUCVMD
Newsgroups: comp.sys.next.bugs From: ravn@imada.ou.dk (Thorbjoern Ravn Andersen) Subject: Re: NFS mounts -- time stamp problems? Message-ID: <1993Jan28.153955.13826@imada.ou.dk> Sender: news@imada.ou.dk (USENET News System) Organization: Dept. of Math. & Computer Science, Odense University, Denmark References: <1993Jan25.191725.23058@imada.ou.dk> <C1Jxtu.H2q@ux1.cso.uiuc.edu> Distribution: comp Date: Thu, 28 Jan 1993 15:39:55 GMT >Is this under 2.x or 3.0? >I have seen similar time-related problems when 'touch'ing files that >are actually on a Sun exported via nfs to a NeXT under 2.1. It was >acknowledged as a known bug and fixed in 3.0 when they supposedly >re-wrote much of the nfs code. This is under 3.0. Apparently, this is both when using Sun and GNU touch but only when updating the date; not when actually creating an empty file. I have been e-mail'ed [will appear in the summary] that this is an internal Sun convention which is not supported by the Mach kernel. The actual setup is: Using: Sun Sparc machine (SunOs 4.1.[23]). Files: Located on NeXT NFS mounted filesystem. (NeXT Cube, OS version 3.0), but without explicitely using the NeXT machine. Further comments are very welcome. I am seriously considering moving my home directory to a hard disk on a Sun server. Regards, -- Thorbj{\o}rn Andersen ravn@imada.ou.dk
From: armitage@milton.u.washington.edu (Armitage) Newsgroups: comp.sys.next.bugs Subject: niload and /passwd under 3.0 Date: 29 Jan 93 04:14:24 GMT Organization: University of Washington Message-ID: <armitage.728280864@milton> Keywords: 3.0 niload Apologies in advance if this thread has already been explored; I've been skipping this group recently (for shame!) I noticed this problem when I tried to update /passwd by niloading it (I had made wholesale textual changes to the password entries). The command I issued was: niload -d passwd / < /tmp/pw I've issued this command many times under 2.1, but I got the following message (actually, tons of messages, since I had a lot of expired accounts): Can not delete named object with children. Or something very similar. It turns out that under 3.0, NeXT added a directory to the /users entries ('info', I believe this directory is called...doesn't seem very useful to me, but whatever). I don't mind one bit that NeXT changed the format of this part of the NI architecture. What I _do_ mind is that they didn't make provisions for using niload in this particular situation. Yes, I know that NeXT would like everyone to use UserManager for all user admin tasks, but this does not take into account off-site administration, nor does it take into account times when one would like to make huge changes to the /passwd database. Ok, you say, I should use NetInfoManager and select the range of users I want to trash, and then delete them. Well, there are two problems with that. First, unless I'm just not clever enough to figure it out, you can not select ranges (in fact, Edit->Select All didn't do a thing; note that it is entriely possible that I'm not clever enough!). Second, I synthesised a password entry from another machine's /etc/passwd, and I wanted to change accounts whose entries had changed, and delete entries that had expired. The command I issued would have worked admirably under 2.1. Oh, and third, I wanted to do it at my Xterminal as opposed to walking to the server room. Anyway, the point is, NeXT really dropped the ball on this one. I feel that NeXT really neglects the fact that people still do things to NeXT machines from the mach shell. Yes, if I had a choice, I would use NeXTStep for everything, but there are plenty of things I need Unix for. The workaround (if anyone is interested) that I used was as follows: I selected the /users directory from NetInfoManager, and deleted it! Ok, it took a long time, but it did disappear. Then I recreated it, and manually created a 'root' user with all of the appropriate user fields. Note that I was changing / /users, and I was logged in as root on the NetInfo master (probably a Good Thing). So, the point is, NeXT ought to remember those few (that is, ALL of those who have to do _any_ administration of NeXTs) people who use the Unix- level NetInfo manipulation commands. Oh, once again, sorry if this has all been covered before. Try to keep your flames to under 2 lines if this is the case. ;) -- [gcc] [g++] /*-------------------------- // --------------------------- Peter (Z-man) Zatloukal // Peter (Z-man) Zatloukal
Newsgroups: comp.sys.next.bugs From: gary@nshade.uah.ualberta.ca (Gary Ritchie) Subject: Re: FontPanel select... Message-ID: <1993Jan28.191732.14260@kakwa.ucs.ualberta.ca> Sender: news@kakwa.ucs.ualberta.ca Organization: University Of Alberta, Edmonton Canada References: <1993Jan26.204432.986@proximus.north.de> Date: Thu, 28 Jan 1993 19:17:32 GMT > It says in the 3.0 release notes: > > o The FontPanel, when key, will select in the Family scrolling list > when you type the first few characters of a family name. > > But when I type letters they appear in the size field. No way to get the > damn thing working. > I believe someone from NeXT acknowledged this as a 3.0 bug a couple of months back. To the best of my knowledge, the fix is the fabled NeXTSTEP 3.1. --- Gary Ritchie
Newsgroups: comp.sys.next.bugs From: nwc (Nick Christopher) Subject: Digital Websters Message-ID: <C1MAwK.1M8@ny.shl.com> Sender: usenet@ny.shl.com (Net News) Organization: SHL Systemhouse Inc. Date: Fri, 29 Jan 1993 13:31:31 GMT Print is broken. I think it uses DPS calls that return values so th postscript breaks outside of the scope of the program. \n Nicholas Christopher Systemhouse nwc@ny.shl.com
Newsgroups: comp.sys.next.bugs From: alberto@parsec.mixcom.com (Manuel Alberto Ricart) Subject: Re: FontPanel select... Message-ID: <1993Jan31.010451.1024@parsec.mixcom.com> Organization: Is the act or process of organizing or of being organized. References: <1993Jan26.204432.986@proximus.north.de> Date: Sun, 31 Jan 1993 01:04:51 GMT In article <1993Jan26.204432.986@proximus.north.de> gemoe@proximus.north.de (Gerhard Moeller) writes: >Hi, > >either I'm stupid or... > >It says in the 3.0 release notes: > > o The FontPanel, when key, will select in the Family scrolling list > when you type the first few characters of a family name. > >But when I type letters they appear in the size field. No way to get the >damn thing working. > > Any hints? > Gerhard. With a little clicking you can get it to work. What you have to do is this: Click on the FontPreview content area to get the insertion point. After that click on a font family name. From now on, typing a few letters of a font name will do what you want: select a different font. I know this is weird, but it sort of works. -- Manuel Alberto Ricart alberto@parsec.mixcom.com <NeXTMail Welcome>
From: eilts@late.e-technik.uni-erlangen.de (Hinrich Eilts) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: gated runs amuck on NS3.0 Date: Sun, 31 Jan 1993 20:28:30 GMT Organization: LATE, Uni-Erlangen, Germany Message-ID: <1khcpeEsr6@uni-erlangen.de> My gated runs ok with NS2.1 but now it's runs amuck: it allocated lots of storage up to an allocation-error (by about 60 MB) which let's him die. I compiled new with the 3.0-NeXT-cc without greater problems (correction of some headerfile-locations) my 2.1-sources, but it's the same error. The computer is a Next-Station mono. Any help will be highly considered. -- Bye | G i b D O S | Hinrich Eilts | k e i n e | (e-mail: eilts@late.e-technik.uni-erlangen.de) | C h a n c e ! |
From: doug@foxtrot.ccmrc.ucsb.edu (Douglas Scott) Newsgroups: comp.sys.next.bugs Subject: IB crash just trashed several hours of work Message-ID: <7495@ucsbcsl.ucsb.edu> Date: 27 Jan 93 19:02:35 GMT Sender: news@ucsbcsl.ucsb.edu Organization: Center for Computer Music Research and Composition, UCSB While working on an App using Interface Builder (OS 3.0), I was editing the text in an Info Panel, when I hit <cmd> :, thinking I could check the spelling of what I had typed in. IB quit instantly, with no error messages, nothing. Just took all my work with it, without a clue. I would consider this a bug. -- Douglas Scott (805)893-8352 Center for Computer Music Research and Composition University of California, Santa Barbara Internet: (NeXTMail ok) <doug@foxtrot.ccmrc.ucsb.edu>
Newsgroups: comp.sys.next.bugs From: kevins@slow.inslab.uky.edu Subject: Re: Digital Websters Message-ID: <C1sEwC.88H@ms.uky.edu> Sender: news@ms.uky.edu (USENET News System) Organization: University Of Kentucky, Dept. of Math Sciences References: <C1MAwK.1M8@ny.shl.com> Date: Mon, 1 Feb 1993 20:43:23 GMT Nick Christopher writes -> -> Print is broken. I think it uses DPS calls that return values so th postscript -> breaks outside of the scope of the program. -> -> \n -> Nicholas Christopher -> Systemhouse -> nwc@ny.shl.com try looking up the number 72 or any other for that matter. -- _______________________________________________________________________________ I'll cross my heart and hope to die but the needle's already in my eye. _______________________________________________________________________________ Kevin Solie
From: stolcke@ICSI.Berkeley.EDU (Andreas Stolcke) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.sysadmin Subject: lseek(bad filedes) crashes NeXT2.1 Date: 2 Feb 1993 18:29:32 GMT Organization: International Computer Science Institute, Berkeley, CA, U.S.A. Message-ID: <1kmeic$p05@agate.berkeley.edu> The little program below tests the file descriptors from 0 ot 31 for open-ness by performing an lseek and checking errno. This happens to consistently crash out NeXT cube running the 2.1 release. Question #1: Is this a known problem? Question #2: Is there a fix for it. The problem doesn't seem to be specific to lseek. A similar program using select(2) to test the file descriptors also crashes the machine. I consider this a very bad bug since it means that any program that is not careful with open's and close's can produce very bad surprises. Please reply via e-mail since I don't read this group. Thanks in advance. -------- cut here ------------ #include <errno.h> extern int errno; extern long lseek(); main() { int i; printf("Open streams:"); for (i = 0; i < 32; i++) { if (lseek(i, 0L, 1) >= 0 || errno == ESPIPE) { printf(" %d", i); } else if (errno != EBADF) { extern char *sys_errlist[]; printf(" error on %d: %s", i, sys_errlist[errno]); } } printf("\n"); exit(0); } ------------------------------- -- Andreas Stolcke stolcke@icsi.berkeley.edu International Computer Science Institute stolcke@ucbicsi.bitnet 1947 Center St., Suite 600, Berkeley, CA 94704 (510) 642-4274 ext. 126
Newsgroups: comp.sys.next.bugs From: liberte@cs.uiuc.edu (Daniel LaLiberte) Subject: Repeating keys in Workspace Message-ID: <LIBERTE.93Feb3103606@ebony.cs.uiuc.edu> Sender: news@cs.uiuc.edu Organization: University of Illinois, Dept. of Comp. Sci., Urbana, IL Date: Wed, 3 Feb 1993 16:36:06 GMT I have a slab with on 8M running NeXTSTEP 3.0. When in the Workspace Manager in a File Viewer, it used to be (in 2.0) that hitting arbitrary characters would bring up the Find window. Now it just beeps. The bug is that the beeping takes increasingly longer as more chars fill up some internal buffer. Recently, the mouse was left on the keyboard for a day (I have a 3 year old), and the beeping didnt stop after 12 hours. I was able to reboot, though it took a couple minutes to shutdown. I would be happy bringing up the Find window in this situation, where presumably repeating chars would be responded to more quickly. Is there a way to do this? Dan LaLiberte liberte@cs.uiuc.edu (Join the League for Programming Freedom: lpf@uunet.uu.net)
Newsgroups: comp.sys.next.bugs From: liberte@cs.uiuc.edu (Daniel LaLiberte) Subject: IB Sounds Message-ID: <LIBERTE.93Feb3104118@ebony.cs.uiuc.edu> Sender: news@cs.uiuc.edu Organization: University of Illinois, Dept. of Comp. Sci., Urbana, IL Date: Wed, 3 Feb 1993 16:41:18 GMT This seems like an obvious bug that someone would have reported previously. Creating new sounds in IB (NeXTSTEP 3.0) has the problem that the names of the sounds want to be generic like "Sound", "Sound1", etc. It lets you rename the sounds, but they don't stick. Dan LaLiberte liberte@cs.uiuc.edu (Join the League for Programming Freedom: lpf@uunet.uu.net)
Newsgroups: comp.sys.next.bugs From: rfschtkt%banruc60.bitnet@ohstvma.acs.ohio-state.edu Subject: sonata.cc.purdue.edu: next/lore/knownbugsin3.0v6.rtf.Z Message-ID: <9302031129.AA00870@flexus> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Wed, 3 Feb 1993 11:29:17 GMT Hello, First of all, sorry for any duplication. I send this to comp.sys.next.(bugs,programmer), NeXT-L@brownvm.brown.edu and NeXT-prog@cpac.washington.edu, plus the people mentioned in this file (so far, and for the version that is in the works). Spread the word. I have only send access to Usenet, but I receive the lists. If you are concerned about NeXTSTEP_3.0 bugs (who isn't), want to check out how I rendered your contribution (you might not have contributed directly, check out how this message reached you), fetch that file, and submit those of your bug reports that aren't covered yet. Raf Schietekat, RfSchtkt@maze.ruca.ac.be (Flanders, Belgium) (my From:'s still state banruc60.bitnet, which is the same site) (real, i.e., with triangle in the Deliver button) NeXT Mail preferred I can't reach sites with ! or % in their address, or ending in ``at''
From: cbarber@bbn.com (Chris Barber) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.next.sysadmin Subject: Re: lseek(bad filedes) crashes NeXT2.1 Date: 4 Feb 1993 21:54:57 GMT Organization: Bolt Beranek and Newman (BBN) Message-ID: <ln345hINN4bg@news.bbn.com> References: <1kmeic$p05@agate.berkeley.edu> In article <1kmeic$p05@agate.berkeley.edu> stolcke@icsi.Berkeley.EDU (Andreas Stolcke) writes: The little program below tests the file descriptors from 0 ot 31 for open-ness by performing an lseek and checking errno. This happens to consistently crash out NeXT cube running the 2.1 release. Question #1: Is this a known problem? Question #2: Is there a fix for it. The problem doesn't seem to be specific to lseek. A similar program using select(2) to test the file descriptors also crashes the machine. I consider this a very bad bug since it means that any program that is not careful with open's and close's can produce very bad surprises. #include <errno.h> extern int errno; extern long lseek(); main() { int i; printf("Open streams:"); for (i = 0; i < 32; i++) { if (lseek(i, 0L, 1) >= 0 || errno == ESPIPE) { printf(" %d", i); } else if (errno != EBADF) { extern char *sys_errlist[]; printf(" error on %d: %s", i, sys_errlist[errno]); } } printf("\n"); exit(0); } I discovered this bug last summer and reported it to bug_next@next.com but didn't hear anything back about it. Basically, the problem occurs for the following set of conditions: - a file access function requiring an open file descriptor is used on a descriptor which is not open. - the file descriptor is >= 20 - the function is repeated in a tight loop You will probably notice that putting a small delay in the loop will make the problem go away. The more correct way to deal with the bug is to use fstat() to see if the file is open before you do anything. Note that a single random access to an unopened file descriptor will not bring about the bug, nor will anything done with a fd less than 20. -- Christopher Barber (cbarber@bbn.com)
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: collis@lele-next2.Stanford.EDU (Samuel Scott Collis) Subject: NFS mounts in NeXTStep 3.0 (they suck?) Message-ID: <1993Feb4.222843.25990@leland.Stanford.EDU> Sender: news@leland.Stanford.EDU (Mr News) Organization: DSO, Stanford University Date: Thu, 4 Feb 93 22:28:43 GMT Help.... We have experienced problems NFS mounting NeXT's running NS 3.0 on a SGI personal Iris. We have several NeXT's still running 2.0 and their mounts work fine. However, the mounts for the NS 3.0 computers are unreliable, sometimes they work, sometimes they don't. Has anyone else had problems with NFS on 3.0? What is really interesting is that mounting NeXT to NeXT seems to allways work, regardless of the version of NeXTStep. I would really appreciate any help on this problem. It is severly troublesome! Thanks, Scott -- Scott Collis Department of Mechanical Engineering Stanford University collis@lele-next2.stanford.edu (415) 725-3318
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: cmumford@iat.holonet.net (Chris Mumford) Subject: Re: NFS mounts in NeXTStep 3.0 (they suck?) Message-ID: <C1y9vw.HJr@iat.holonet.net> Organization: HoloNet National Internet Access BBS: 510-704-1058/modem References: <1993Feb4.222843.25990@leland.Stanford.EDU> Date: Fri, 5 Feb 1993 00:40:43 GMT I'm using a NeXTStation (3.0) to do a mount in the opposite direction with an SGI -- mount a published NFS volume from the SGI to the NeXT, and haven't had any major problems. I say major, because it works, but NFS support definately has problems with it.
From: kiwi@cs.tu-berlin.de (Axel Habermann) Newsgroups: comp.sys.next.bugs Subject: Too many fonts-problem? Date: 7 Feb 1993 14:54:33 GMT Organization: Technical University of Berlin, Germany Message-ID: <1l37r9$fjn@news.cs.tu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 8bit Keywords: fonts Hi folks! Some time ago (maybe one month), there was someone complaining about a limit on the maximum number of fonts under NS3.0. I think I remember a number of 250...and there was a workaround. I would like to ask if someone can repost the solution. Thanks in advance! Axel -- Axel Habermann \\|// FaxFon: +49 30 4543046 Fon:4531772 Fehmarnerstr. 4 )o o( eMail: kiwi@cs.tu-berlin.de 1000 Berlin 65 \ | / "Wenn Du nicht weisst Germany \~/ was du tust, mach's mit Eleganz!"
Newsgroups: comp.sys.next.bugs From: lemson@ux1.cso.uiuc.edu (David Lemson) Subject: Re: Too many fonts-problem? Message-ID: <C23Eu3.KKM@ux1.cso.uiuc.edu> Keywords: fonts Organization: University of Illinois at Urbana References: <1l37r9$fjn@news.cs.tu-berlin.de> Date: Sun, 7 Feb 1993 19:15:25 GMT kiwi@cs.tu-berlin.de (Axel Habermann) writes: >Hi folks! >Some time ago (maybe one month), there was someone complaining about a limit >on the maximum number of fonts under NS3.0. I think I remember a number of >250...and there was a workaround. You need to get a hold of the /usr/bin/builafmdir from a 2.x machine and use it instead of the one supplied with 3.0. That will fix the problem. -- David Lemson (217) 244-1205 University of Illinois NeXT Campus Consultant / CCSO NeXT Lab System Admin Internet : lemson@uiuc.edu UUCP :...!uiucuxc!uiucux1!lemson NeXTMail & MIME accepted BITNET : LEMSON@UIUCVMD
From: rob@polestar.facl.mcgill.ca (Robert Macfarlane) Newsgroups: comp.sys.next.bugs Subject: 3.0 Librarian = Corvair? Message-ID: <1993Feb8.201917.8706@sifon.cc.mcgill.ca> Date: 8 Feb 93 20:19:17 GMT Sender: news@sifon.cc.mcgill.ca Organization: McGill University 3.0 Libarian crashes on me almost every day. Has NeXT released a patch? Rob Macfarlane Faculty of Arts Computer Lab McGill University
Newsgroups: comp.sys.next.bugs From: schmitt@tn.incom.de (Matthias Schmitt) Subject: Diagram won't launch with long system font name Message-ID: <1993Feb9.140343.5540@tn.incom.de> Keywords: Diagram fontname Default Database Sender: news@tn.incom.de Organization: Telenorma - Bosch Telecom Date: Tue, 9 Feb 1993 14:03:43 GMT Having "New Century Schoolbook Roman 12 point" in the GLOBAL default NXSystemFonts prevents Diagram! from launching. -- Matthias Schmitt - OIV11 - Software Engineering Phone: +49-69-266-2634 Telenorma Bosch Telecom Germany Fax: +49-69-266-4406 Mainzer Landstr 128-146 - W-6000 Frankfurt 1 (NeXT)mail: schmitt@tn.incom.de
Newsgroups: comp.sys.next.bugs From: nwc (Nick Christopher) Subject: Does chnk work? Message-ID: <C1rvqy.5Fn@ny.shl.com> Sender: usenet@ny.shl.com (Net News) Organization: SHL Systemhouse Inc. Date: Mon, 1 Feb 1993 13:49:46 GMT I always get pseudo-media errors from floppys with a .chunk file on them. Do chunks work? \n Nicholas Christopher Systemhouse nwc@ny.shl.com
Newsgroups: comp.sys.next.bugs From: Jelske Kloppenburg <Jelske.Kloppenburg@gmd.de> Subject: Symbolic Link in LocalApps Message-ID: <1993Feb10.150852.1481@gmd.de> Keywords: Link Apps Makeservices Sender: news@gmd.de (USENET News) Organization: GMD, Sankt Augustin, Germany Date: Wed, 10 Feb 1993 15:08:52 GMT From tests with X on NeXT I had a symbolic link to the X-Demo on OD in /LocalApps. The OD was not mounted. In NextStep 3.0 that sometimes caused system errors on logoff. It always caused an error if I logged off a short time after login. I assume, the 'make services' didn't like the unresolved link. -- Jelske Kloppenburg, +49 2241 14-2433, <Jelske.Kloppenburg@gmd.de> GMD (German National Research Center for Computer Science) "Don't kill the Winners!" Dennis Tsichritzis
Newsgroups: comp.sys.next.bugs From: mikhe@ida.liu.se (Mike Henry) Subject: atoh() and atoo() missing from NeXT libraries? Message-ID: <1993Feb10.114823.5815@ida.liu.se> Sender: news@ida.liu.se Organization: CIS Dept, Univ of Linkoping, Sweden Date: Wed, 10 Feb 1993 11:48:23 GMT Are the atoh() and atoo() functions missing from the NeXT libraries? They're included in the header files but cannot be found for linking. I've grep'ed through all the libraries in /lib and /usr/lib but to no avail. Is this a known "bug" or am I totally beclouded? -Mike -- Mike Henry INET : mikhe@ida.liu.se /// August Wahlstromsv. 4 /// S-182 31 Danderyd \\\/// SWEDEN TEL : +46 8 755-8687 \XX/
From: hugh@ear.mit.edu (Hugh Secker-Walker) Newsgroups: comp.sys.next.software,comp.sys.next.bugs Subject: link and registerDoc problems Followup-To: comp.sys.next.software Date: 12 Feb 93 13:12:33 Organization: Massachusetts Institute of Technology Message-ID: <HUGH.93Feb12131233@ear.mit.edu> I have a problem with link-using applications that has recently graduated from simply annoying to truly counterproductive. When using Edit or Draw, almost every time I save I get an Alert Panel with the message: Error while writing link information to <pathname>. [OK] This occurs in documents with no link material. Indeed, in trying to track down the problem I've found I cannot create links. When I export them things go smoothly, but when I try to paste them in I get an Alert panel telling me I have to save the source document first. BTW, except for not being able to paste in links, the documents are fine. Edit Help-links work fine, and nothing gets lost or trashed. Just lots of Alert panels, and no links. Also, the following two kinds of error messages appear on the Console 1) Feb 11 22:24:52 localhost Edit[308]: Assertion failed: _unregisterDoc: appkitServer didn't recognize our doc 20 Feb 11 22:35:24 localhost Edit[308]: Assertion failed: _registerDocAt: appkitServer didn't recognize our doc What do these mean? What's the appkitServer? I'm baffled. I can't find anything in NeXTanswers regarding this. I suspect that something important was left out when I upgraded to 3.0. Anyone seen this sort of problem before? Using Mono NeXTstation 20/340 with NeXT printer, ZyXEL modem, NXFax demo, AppsoftDraw demo, NS 3.0 with Developer kit, .... Any suggestions will be appreciated. Thanks for reading. Hugh -- Hugh Secker-Walker INTERNET: hugh@mit.edu MIT, Research Lab of Electronics UUCP: mit-eddie!mit-athena!hugh
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: JPEG compression bug in NXBitmapImageRep (NeXTMail) Next-Attachment: .tar.142.JPEG_compression_bug_.attach, 5548, 1/1, 7656, 0 Message-ID: <9302122330.AA14251@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Fri, 12 Feb 1993 23:30:58 GMT begin 666 .tar.142.JPEG_compression_bug_.attach M'YV0:=R0*8/'A1PZ9@`H7,BPH<.'$"-*G$BQ(HB+-FC0``'@XL4:,CAZO`A# MY,B+,6C$"`FBQHP9-V#,2%FCAL<:-FS,Z%BQI\^?0(,*'4JTJ-&C2),J7;IT M#Y>#9F!P">-F3AJG9MZXH4-'#!LN4<&Z2<,&!)4T;<K,:2'E31NJ._HHX/)6 MSADV*V',K7M&3MZY<,+((<.%#AZ<4@W'R%L83PP;.!+CD2&#AF09D26_M*$Y MAPS)&B,WK@'#\FC.4L.*D9I&:ATV;K26`3L',]@Q4L=$!<%%`14T94!D9</F MS9V`9T#0N?-&.%FU(,9HI1,F((@P(*RV@<,F.!PY;_J&::.\>1@XW/-<C_ZF MN!SG;.B4>4_'O-DD1HR`0!OF3!D7%UT!G!L@R%$'@5IEYU9P2D!1Q!$M2+>= M'&K-4089^[WEWQP@H)`$"&B$84=PV,TQ'G=EL`""&'70L9^+:7`H1QAT`/<> M&X+YEX)RP($@&PAU6.5&<G.EL=]8=!@!7AM&/*=#D26!\-T;9-0Q!G0U!C=< M<<<-R5Y5[04W'WARZ-!;;R`DH6%P%LHQXGMWE+'5=7>$D<<.T5%U@HL2HB@? M"`T^6&`9<=2A%AUG*D`5AEGZV"(<+<(77(SLR6$@'/*1H>(=:#S'8W!O9/G> MCQ:6T0:']8'`G!QK..?&?XDJL(=3-%(W!AII;05324Z4@0450[3WAAQ-QC>? M"VVXT!X9"L@UZU1<A7%KKG3L"D*OOP;K7K'RR8%LLX`)1IAAB#6VV&?F0G99 M99=EUMAFG:%K6&B2D68:N:B!)=5J7+3&Q6NQO4J;;6;@QH5N)27:`@A3U"'& MC-0QG(8<8PC6VQQ642P8$!4'[(*$X#(E\L@DEVSRR45A"ZRPQ)+5+;+*LD$& MRC0[Y%%&&W5T4T@Z>U12SR/-@!-++LV04PQ&X^31#4S7P%/-4$<M]=145VVU M42.@!<>P+O)P'AQKI$''"U^'38<+:/B@0-;;<9T`#W/00>%X<[P0]]RGHJTV MVUL?]#8;:8@QAMX*K*TUUR"(,$?'TKEA!]HB%#["0&8$%-P40P3AQ!=-!`$L M$E],D8061220@%5ZE/&&&2@L3M47;]%Q:PJ%V_%&&AB^%1`*`;F8XQ@JWBH8 M""KD:,<67="^AP().(&%$&*_!8>:_94A11EP)*""=M6#T,.U53#!Q`[,.S^% MW&6,1_S=Z9/WO1/AC\]\'5\(+X?V9-`8!A1R>P^^^.1+0.^R@QX5B0$.<U"1 M&<@00->Y`78TNM7I.M:$"*)!"'4P@QGF0SX%C"0-9NC0[T`0@N_18$?+,]UW M>L<Z$53!1/[1P;5\M;)MN6P^['%/Y;HCJ1A(2@9<<$-AI*-#3U&J4?F+V`ZU M-*P<-C$,9%!#D.A`K2`69HD^I!1VTC('X&"H,BT0@]A`((4C",&)[^&/?ZR( MD.>$Y(@]&DC<`D*C-"1HB:KJU*U`0(8W0"<V+LI#&5Q$D!B=+8@B2$$`$U!( M.J`@!HKTH$?D\L$0MDZ-P?G>%K;@/.A1\3S4\\_UX'`=XKQA#%TXDMB4Y)9B ME4$'QMN"#+J0/!`L;R0).""'-,D]_ZQ(;'.`PGRF<*+N=(%\N)P#>ORWA5ZR MJ9AJ$:8<H)`&/)2!#<>4Y$4$:$D4Z)*$W\/!COC`APXIDY0E!,$,QEG.9F)R M/\'$D1L$4\L4JE`.+$2!"'Y#*4H!<CU<]"((P"A&%Y7QC$1L(B9=@,A(FHZ1 M>!`;"M:)3(\D0"X7!<$U+61+YB5@A5MIH1&JPX8Z4*@\/H*#G$!0`@XM+`QB MX!IR4CJ?.FJ%H6X0`0MB.4N'0E2B,G"H7)@'0A&.X4H80T$L8]`%%4GA"T]8 MPH[2V0)(=M1T[%/?^[!003BX4JEV.1Y35>2\+TBA"$$@PA.<P(0L^+2H*'#> M]:#8.O2-1T4HL!WNB)<"$SBP@K*[8`8W*`<584YSG//<$$`G.M+MJ`??.^SF M.O>YT(VN""CT:`+:H!6Q#0N"V5,!9]W`O_<DH`<3I`I@;X5!#1YKM)Z5PR)/ MAP:W:4]ZF+K?:1.`@M&6M@4^D%X*D#5;^MG/="K``V%SN]O>:N6W/E#N?%(` M7.G*5K/.&T)Q+%17O/F4FQU*XO[ZQ\MWBK>>FDU`X^R04#F@0+RE90%\Y<8" M=ZY)2M7<Z!5P5X8NJ,"^W8-#?MDP!R2DX0QH:"INZ6O=W'ZW#QHE<'#L><]\ MBB!8K\'0/Z6%5`Z95W\X3>1L?^I(&SS8HQ#>Z(0U"U*$Z#,);I".I<HP!A>Y MH0YM$`,.5[>B/,C'PP1JJ0MTNM3DS3:[VRU#=]OWW4:BH`8^Q6B*)7S5C^(S MI/K$L,Q`L.&C5HBE<QCR3L.Z!:;ZU,DG#&`?$E"X2G8(!0OTWQCFYLA8SJ"I M,,B)B7?$`Q#`(+,5QK((6AH=.FO44L-"Y)C/<+P[GSFBCKQ!)$="28]DU7W7 MPL(35.J&KRZ0!64]:UJO((4D4`&S%;T(@'UY!WS*APKX,8(.+@TD(9TA6!.J MD!W=H(.RPCH_7QC"$YH`A;-.071K_8(3UEJ$ZPADI#4>E@Y@<$P0:#/3VGT# M=R\]:8^,(<EP)D.W5>U,X5"HOP%T\I_))Y>KN?O=`%"9MH;%K=?"VV0WTXA) M/L*SD_SL)!>A066(]I*84.8S+`'!#6)P`QD\[=X0C[C$)TYQJ/$-<5Y#C]G( MIG&Q$>[B?H.;74]EMY&'.6V&:UO(`2<XPJ6\;RY2'..T\C@T1&YME+,<PS(W M6<4R]K)834/J5M>ZCD$PL+13@%YS5QTW\&Y.OPM>B-Y3/#(G3U;E>U[T0+FF M46JOW.^+7P#--_+UE3WL`)Q?_::./_V5UG_P2[L`YW1.`R)0@0QDG@./+L&_ M6K"UA.V@F\%ZAC&`$P0GK'*+6_C"ZLE0WBRK]WO:BT<L_I"-E#<BJN*H/TD) MIXGMO4X4IUA%(;:Q.UGDT!;5(E""CO&@:,Q0]=BXQ#=N/CARI`,==7]'3W$J M#7OLXQ]#!0)!$C*B<0OQHR4*R52#`*-PY7"%"'^\63H5JE+UWO>J"F@K6[@* M\_1*<%(U-PP)^2)0Z$X8.'HK&K>J42J50QMB9!6MA)G(9.[I(IT<5#4S+\[? M\P:<1GVRU%1/8%9JQ59NU4!GEVF;)B>>IBF99E9HE8!MY5!P16LF8`+F]$Z: MQ$E:]TG3TW784TK%@4JJE"1+<CYX,VLCESS=EX$CMR-(IFU*QFV+!%<+M"/? M9H/A-FX@D$L(Q$SE5E#!-$S0E$W)M$SE=5\FLAW=<833-&!*:%%PY4U#F$[B M!`+D9$[+E$[KQ(7M5&XQDGY404_=YWV"QD\>QB$;!@(!=2$#10-A]'IF%'L+ MU5"SA69`>%',,V4<94\RZ%W8!FXXZ%$Z*&[1`6X[N$B+IT\C118F-7[-(8`K M16@O%5,',5.6"#&[%F*+5GU&YE%.1E%_6#C@I51>EE1%=GU1-57;9U7V1&M; MU55?U8H3*&H6N("(:$ERE3YDL&1WU2%+QU=^14%_-UCS85@\EUB5U5B8I7T[ MAUB4M5B6Y5A5MEF=51]R`%JW]5S]LUM^%UB`]UK;."RSU46V)5I?(S>F@UK. M15IR`US"15R:95QLIST-YH[PZ%OS&%V$15T`R4'8A079MFTS.%MP-5_OT83= M<UYIJ%XTUU[OY7;TQ9#U56X"9DT$ME\#X5^KYAT#5F`'EF`LL&"%M8]R\&`1 M%HCI]8@7]@89QF7$)WT8(WN^)%[*-V).)FFSA5$9I6+9"),P)F,44F-<AF,Z M-BHA)`8^!AW6(63XQVAE-HH/58,(Z5U["&DH8&*+)&4MN6*!YF(Q.9-=UF%@ M)F9%MGR.!&5JQF;7=H4`6&CI4V=D=F<LD&<YD0)\YF?=!Y.$-F=U>6ADHFAV M9I4DA@+BY'R5=A&TZ("<%H&@A@44.&JE=FI`&)*JXFIE\&NR1FM!@ARX]AVZ MIA6]1IF>&6S#5FQ%<&Q)D&S+Y@3-MBC0QHW35FW7AI4WF)#7UH/<U8C7IIEF M<&[9E)CK%C(5EYS*N9S,V9S.^9S0&9W2.9W469W6>9W8F9W:N9W/"7DV9"S> MT@;<^1#YEC,C`1+[1A+IZ1$")W`?47`P<'#QN30,YW`<,9[XF9_Z*34@US5E M(S8<!S8>AW+]^3:75C<'^G&'LW*!,S@H]W*((W-4L5Z0(SDY]RK3V'//"'2G M(W2JPSI[%SNS4SNWPW2[,T!1%QU31SRQ='4IU$E;-X+5XW7;XX'_(S_-@P4L MV#YFAS=P)W9J=UPJP)`_*G<#5'<K<G?"D7>I]4`BB@9-NEJ"Y5K7%9>6-$+I ME'CV!).-%T,SE"V1=T.3QS*5]QP^5'N81Z::]RE\U'F5!WHL(WI2%#>E=T5F M"D\`Q7IRZ'H&=8>AATFTYT9XBD2'LGN[YGF_%WQ^Y(;$9WP:A7R'E%-L^4A` M"'U7NHIS0(#61T;8]UBQ^)=71I;@!U,\1'[`F);HIW[L!QQC\'X]$G_SAS&[ M=G^A6("3VG^GF`!S:8E.QU,&B(!K=8$,Z*-;]8"=]ASA1E:4J8O!RHNIJ($< M>$GW]8$P*H*A9#TE&`:FA((!L4HK.'(NB#<P6&6#R&2%Z(.':#J)R(.,*&[. M)X2[!`*::8321$Q0B&[7=CI,**_E]H0H(H74Q)%5N$U7^$U:R$Y>B$[?$X9= MJ)EE*$]H.)2AVD)LB*=O&(=?1(<%149^&J=Y**E;*5%IEJ]@*92":$G<=JY9 MR60Y:$D[N(@^")QC*5(D-8DHQ:M@!@*9*%->THDVY0:@Z*N3:HI^2%27VF$$ M.%:<^HJ'QWU5]IC.8XO(BHNA5H'-^E:^B`5S%8R7AE?%J`)]-8ZLI8R%E:'. M:(W0Z*EF6XT_AXT4!EO<Z(VB!8ZFA5IB.Z6$A2SG>%T/I8Y^<UOMJ%O]2+?T M>![#U0;%M7:"@5PJV5S^*`?5%9"12Y!7:9"&R)L/M9`6V9#\^F'4@5X/)9&. M0Y$8B9&:N9'ZQ5\@J9$C:6`(IF"!RP(JR9(F^Y(3FV4RN65G^66>&P8[F5X] M>6*F`XAB.;,O%F/#<I0VII0[UI1/"61I.95BA9@Y>I"[J96:Y61>B6)_&):* M=[MEJ;LUB:G16ZMFMG]<Z99_")>#%VYR1F<$B)=ZN6<@T&=_]KT6%IB&-B:) MEE.UZFCH*U&+>6V-F1T-Z#S&*IE5:YFFAFK!66ZM)C:=&6OARJ.A.22C22&R M:IJ^%FNJ26S&AFR;$YNS^6S28IO4AB>Y:;GHRILCX9M*!IPC(9S$F6Y<>9SM 9MI\ZO,,\W,,^_,-`',1"/,1$7,1&?,02![GH ` end
Newsgroups: comp.sys.next.bugs From: kira.uucp!death@netcom.com (David John Burrowes) Subject: OD booting with 3.0 Message-ID: <1993Feb14.043053.407@kira.net.netcom.com> Sender: death@kira.net.netcom.com Organization: No organization at this time. Date: Sun, 14 Feb 1993 04:30:53 GMT Is there a particular reason (do you know) that I can't set my machine to boot with the OD drive under NS 3.0 on my NeXTcube? I could do this in 2.1, but under 3.0, the preferences panel merely says 'not applicable' \david john burrowes death@kira.net.netcom.com
Newsgroups: comp.sys.next.bugs From: system@arsenal.com (SYSTEM 0PERATOR) Subject: Test of Cnews! Message-ID: <esayyB1w165w@arsenal.com> Sender: bbs@arsenal.com (MadNuG BBS General Account) Organization: The Arsenal BBS - MadNuG - NeXT Users of Madison, WI Date: Sun, 14 Feb 1993 07:32:13 GMT Howdy, This is a test of Cnews. If it works, praise the lord. I hate Cnews. Agh! Later - C --------------------------------------------------------------------- Andrew T. Foster - (608) 251-5522 - FAX:(608) 251-5727 NeXT Campus Consultant - University of Wisconsin System Captain@Arsenal.com - Audix #5585 - NeXTMail Yes! Call the MadNuG BBS! - (608) 251-5565 - v.32bis - 8N1 - 24hours ---------------------------------------------------------------------
Newsgroups: comp.sys.next.bugs From: dan@ennex1.eng.utsa.edu (Daniel Shelton) Subject: Re: OD booting with 3.0 Message-ID: <1993Feb14.200047.11709@ringer.cs.utsa.edu> Sender: news@ringer.cs.utsa.edu Organization: Univ of Texas at San Antonio References: <1993Feb14.043053.407@kira.net.netcom.com> Date: Sun, 14 Feb 1993 20:00:47 GMT In article <1993Feb14.043053.407@kira.net.netcom.com> kira.uucp!death@netcom.com (David John Burrowes) writes: > Is there a particular reason (do you know) that I can't set my machine to > boot with the OD drive under NS 3.0 on my NeXTcube? I could do this in > 2.1, but unde r 3.0, the preferences panel merely says 'not applicable' > > \david john burrowes > death@kira.net.netcom.com The only reason I can think of is that NeXT did not ship 3.0 on an optical so therefore did not enable the option. The fix I would suggest is to do the parameter change yourself from the NeXT monitor by changing the boot option from bsd to bod. Dan
From: hacker@access.digex.com (Dark Hacker) Newsgroups: comp.sys.next.bugs Subject: vi -r is broken under 3.0! Date: 14 Feb 1993 18:07:56 -0500 Organization: Express Access Online Communications, Greenbelt, MD USA Distribution: world Message-ID: <1lmjccINN2dg@access.digex.com> Looks like the recover feature of "vi" is broken under NS 3.0. Everytime I use it I get a file that looks something like this: Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Mf8~~j Pretty strange. I'm amazed that NeXT found a way to break this. Anybody else had the same problem? - Hacker -- Dark Hacker @ Black Silicon, Fortress Of Computation hacker@black-silicon.mclean.va.us "Life itself is... COMPUTATION!"
From: lusty@lusty.tamu.edu (Lusty Wench) Newsgroups: comp.sys.next.bugs Subject: Re: vi -r is broken under 3.0! Date: 14 Feb 1993 23:56:10 GMT Organization: Me Distribution: world Message-ID: <1lmm6qINNcpv@tamsun.tamu.edu> References: <1lmjccINN2dg@access.digex.com> In article <1lmjccINN2dg@access.digex.com> hacker@access.digex.com (Dark Hacker) writes: >Looks like the recover feature of "vi" is broken under NS 3.0. Everytime >I use it I get a file that looks something like this: > >Mf8~~j ...[munch]... >Mf8~~j > I had the exact same thing happen to me the other day. Since I think I only ever tried to vi -r once before upgrading to 3.0 I just kind of ignored it and started over with my original file. Diana
Newsgroups: comp.sys.next.bugs From: verket@venice.sedd.trw.com (Paul Verket) Subject: cc -fno-asm broken under 3.0 Message-ID: <1993Feb15.024827.18335@venice.sedd.trw.com> Originator: verket@verket-home Sender: news@venice.sedd.trw.com (USENET News) Organization: TRW Systems Engineering & Development Division, Carson, CA Date: Mon, 15 Feb 1993 02:48:27 GMT In 3.0 the "-fno-asm" flag to cc does not allow one to use the name "inline" as a variable name as the man page states. The option performs as documented under 2.0. Paul Verket (NeXTmail ok)
From: root@answers-r-us.tamu.edu (Operator) Newsgroups: comp.sys.next.bugs Subject: builder error Date: 15 Feb 1993 19:33:09 GMT Organization: Texas A&M University, College Station, TX Message-ID: <1lor5lINNnjg@tamsun.tamu.edu> Hello all, I am trying to build BackSpace using the projectbuilder application and I get the following error: --------------- cc -g -O -Wall -ObjC -sectcreate __ICON __header BackSpace.iconheader -segprot __ICON r r -sectcreate __ICON app backicon.tiff -sectcreate __ICON backo backo.tiff -sectcreate __ICON backmodule backmodule.tiff -o BackSpace.app/BackSpace obj/BackView.o obj/BackWindow.o obj/BlackView.o obj/BoinkSpaceView.o obj/BoinkViewPart.o obj/FadingImageView.o obj/ModuleList.o obj/MySlider.o obj/PanelImageView.o obj/Password.o obj/SlidingImageView.o obj/SpaceView.o obj/Thinker.o obj/UserPath.o obj/BackSpace_main.o obj/ioctls.o obj/ThinkMore.o obj/Thinspector.o obj/BoinkWraps.o obj/psfuncts.o -lMedia_s -lNeXT_s /bin/ld: Can't locate file for: -lMedia_s *** Exit 1 Stop. *** Exit 1 Stop. ---------------- Can someone tell me what this error means and how I can fix it ? Thank you very much for your help. Sincerely Ravi Ramachandran ravir@answers-r-us.tamu.edu (NeXTmail)
From: seidl@next1.lbl.gov (Peter Seidl) Newsgroups: comp.sys.next.bugs Subject: clock loses time Date: 16 Feb 1993 18:28:09 GMT Organization: Lawrence Berkeley Laboratory, Berkeley CA Distribution: world Message-ID: <29042@dog.ee.lbl.gov> Keywords: clock My Preferences clock loses 10-20 minutes/month. Any suggestions on how to fix this? I have an 040 cube. thanks, -- Peter Seidl Lawrence Berkeley Laboratory Berkeley, CA 94720 seidl@next1.lbl.gov
Newsgroups: comp.music,comp.sys.next,comp.sys.next.bugs,comp.sys.next.programmercompcomp. From: dany@lim.sm.dsi.unimi.it (Daniele Rizzi) Subject: help with midi.... Keywords: midi Sender: news-mail@ghost.dsi.unimi.it (News mail user) Organization: Computer Science Dep. - Milan University Date: Wed, 17 Feb 1993 16:09:55 GMT Message-ID: <1993Feb17.160955.5351@ghost.dsi.unimi.it> Hello all, I work on a 68040 Next computer running the latest 3.0 Release of NextStep. I would like to play a .midi file over the serial port, but when I use playmidifile or Presto, I receive on the Console the error messages: - (using port A) -> MIDIClaimUnit failed: can't init serial port. - (using port B) -> MIDI driver hardware framing error and, obviously, no message come to the Midi Expander; I remember the fair times ('030 machine with release 2.0) when everything worked well... I thank you in advance for any help you may provide. dany@lim.sm.dsi.unimi.it ps. the Next serial port is used to be connected with an Apple Midi interface.
From: heading@signal.dra.hmg.gb (Anthony J.R. Heading) Newsgroups: comp.sys.next.bugs Subject: bug with bash 1.12.1 on NS3.0 Message-ID: <1ltv9rINN9lv@signal.dra.hmg.gb> Date: 17 Feb 93 18:14:19 GMT Organization: Defence Research Agency Sorry if you've seen this before, but I've learnt our news was broken when I first posted it, so I never saw any feedback. If you type: cd /etc; pwd 10 times, the pwd stops returning anything after a couple of times, and often bombs out with a malloc/free fault. Does anyone know why? I now use tcsh. Anthony
Newsgroups: comp.sys.next.bugs From: cedman@princeton.edu (Carl Edman) Subject: Re: bug with bash 1.12.1 on NS3.0 In-Reply-To: heading@signal.dra.hmg.gb's message of 17 Feb 93 18:14:19 GMT To: heading@signal.dra.hmg.gb (Anthony J.R. Heading) Message-ID: <CEDMAN.93Feb17143757@capitalist.princeton.edu> Originator: news@nimaster Sender: news@Princeton.EDU (USENET News System) Organization: Princeton University References: <1ltv9rINN9lv@signal.dra.hmg.gb> Date: Wed, 17 Feb 1993 18:37:57 GMT In article <1ltv9rINN9lv@signal.dra.hmg.gb> heading@signal.dra.hmg.gb (Anthony J.R. Heading) writes: Sorry if you've seen this before, but I've learnt our news was broken when I first posted it, so I never saw any feedback. If you type: cd /etc; pwd 10 times, the pwd stops returning anything after a couple of times, and often bombs out with a malloc/free fault. Does anyone know why? I now use tcsh. Even on danger of repeating myself, I have to point out that if you are used to bash, you should switch to zsh instead. The interactive features are comparable and in all programming respects zsh is much more similar to bash than tcsh is. Zsh also doesn't have this bash bug. :-) The latest version of zsh is available from cs.ucsd.edu in the directory pub/zsh and configures and installs automatically and without any errors or warnings on NeXTs running NS 3.0 (and NS 2.x too, last time I checked). Carl Edman
Newsgroups: comp.sys.next.bugs,gnu.bash.bug From: gad@eclipse.its.rpi.edu (Garance A. Drosehn) Subject: Re: bug with bash 1.12.1 on NS3.0 Message-ID: <w++4srl@rpi.edu> References: <CEDMAN.93Feb17143757@capitalist.princeton.edu> Date: Thu, 18 Feb 1993 02:35:30 GMT Apparently-To: gnu-bash-bug@cis.ohio-state.edu Distribution: world Followup-To: comp.sys.next.bugs,gnu.bash.bug cedman@princeton.edu (Carl Edman) writes: > heading@signal.dra.hmg.gb (Anthony J.R. Heading) writes: > Sorry if you've seen this before, but I've learnt our news was > broken when I first posted it, so I never saw any feedback. > > If you type: > cd /etc; pwd > 10 times, the pwd stops returning anything after a couple of > times, and often bombs out with a malloc/free fault. > > Does anyone know why? > I now use tcsh. > > Even on danger of repeating myself, I have to point out that if you > are used to bash, you should switch to zsh instead. While Carl is repeating himself :-), I'd note that this is a known problem in the current version of bash. It has something to do with the malloc calls in interrupt processing routines (or something like that). The word at one time was that the newer version of bash (which is not released yet) would avoid this problem. One thing you could do to avoid the problem yourself is disable the pwd command of bash (bash has it's own pwd processing). In your .bashrc file, add the command: enable -n pwd (you can do a "help enable" in bash if you want to see what that command is doing to you). I'm hoping to see the newer version of bash, as we (at RPI) are starting to have malloc-related problems with bash on version 3.2 of aix for RS6000's (though on that platform it happens when xterm windows are resized). -- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
Newsgroups: comp.sys.next.bugs From: kinch@valve.heart.rri.uwo.ca (Dave Kinchlea) Subject: NFS bug, overwritten file's contents not seen. Organization: Heart Valve Group, John P. Robarts Research Institiute, London, Ont. Canada Date: Thu, 18 Feb 1993 21:07:55 GMT Message-ID: <1993Feb18.210755.17466@julian.uwo.ca> Followup-To: comp.sys.next.bugs Keywords: NFS, NS 2.1 Sender: news@julian.uwo.ca (USENET News System) Howdy Folk; Perhaps others have seen this, this may even be a FAQ, but here is the situtation. I have a NextSTation running NS 2.1 as a NFS client to a Personal Iris (IRIX 4.0.1). I created a colour TIFF file that the NeXT could not understand so I decided to convert it to a GIF file. To do this I issued, on the iris: #> fromtiff image.tif image.rgb #> togif image.rgb image.gif and then moved the file to /u2/tmp (a NFS shared directory) and read it on the NeXT only to find that the fromtiff had misread the file and switched black for white. No problem, as soon as I looked at the file I knew what had happened (I had done this before) so I, instead, displayed the file via imgview and the captured the image via snapshot, issued the togif command and then copied it back to /u2/tmp UNDER THE SAME NAME. So I looked at the new one again and it was STILL screwed up. I was pretty sure that this file was correct, I could display it on the iris perfectly. As it happens I have two next machines on this network so I looked at the file via the other next box and voila it displayed perfectly. At this point I figured that for some reason the next was looking at a copy of the file from swap (it DIDN'T reside on the disk anymore) even though I DID issue a `Update Viewer' command. SO, I figured that what I needed to do was to somehow convince the Next to ACTUALLY go and read the file and I thought I could do this by coping the file. At a terminal prompt on the next, I issued a: #> cp image.gif newimage.gif and then looked at newimage.gif and it TOO was screwed up! So then, from the iris, I issued a: #> diff image.gif newimage.gif They SHOULD be the same, right? Wrong! newimage.gif was a copy of a file that did not exist anywhere EXCEPT in the next box's swapfile (or biod buffers I guess)! A few points: /u2/tmp resides on the iris and is NFS mounted onto the next; the time between putting the new copy of image.gif and issuing the cp image.gif newimage.gif is over 1 hour (I went to lunch after issuing the second togif command); I logged out (twice) before I finally recopied the file (but this time on the iris), obviously to no avail; all of the machines on the network use NTP so this isn't a datestamp problem (not that it is likely to have an hour difference in time ;-(). Now, this seems to me to be a NFS bug on NeXT's part. I could understand (sort of) the confusion as the name was the same but even after I told the next to re-read the directory (`Update Viewer')?! So, is there some option on the mount(2) command and/or biod(8) to ensure that this doesn't happen? If not, will upgrading to NS 3.0 fix this (not that I really want to do that ;-()? Sorry for the long-winded explanation, but I wanted to make sure that I explained the behaviour correctly. cheerskinch Dave Kinchlea Snail-mail: System Administrator/Research Programmer John P. Robarts Research Institute. ".... 2 to the power 759 million to 1 and falling ..." HHGTG
From: hardy@golem.ps.uci.edu (Meinhard E. Mayer (Hardy)) Subject: Re: NFS bug, overwritten file's contents not seen. Message-ID: <HARDY.93Feb18141004@golem.ps.uci.edu> In-reply-to: kinch@valve.heart.rri.uwo.ca's message of Thu, 18 Feb 1993 21:07:55 GMT Newsgroups: comp.sys.next.bugs Organization: Department of Physics, UC Irvine, CA 92717-4575, USA Date: 18 Feb 93 22:08:37 GMT References: <1993Feb18.210755.17466@julian.uwo.ca> Followup-To: comp.sys.next.bugs The same thing was true, e.g., for TeXview under 2.1 -- If I modified the dvi-file on a remote host, and TeXviewed the NFS-mounted dvi-file, the updates were never displayed. Somehow, NFS "cached" the old file and never displayed the update, unless I saved it under a different name, and reopened it. This "feature" has disappered under NS 3.0 -- and things behave normally. So: upgrade to 3.0. -- Hardy ----- Meinhard E. Mayer, Department of Physics, UC Irvine e-mail: hardy@golem.ps.uci.edu (preferred) or MMAYER@UCI.BITNET !!!! NO NEXTMAIL TO THESE ADDRESSES, PLEASE !!!!!
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: Documentation Error Message-ID: <9302190101.AA17797@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Fri, 19 Feb 1993 01:01:54 GMT Nothing serious, but I just noticed a typo in GeneralRef/08_InterfaceBuilder/Protocols. At the bottom of the file, the method description for unregisterDocumentController: seems correct, but the method declaration is that of registerDocumentController: (looks like someone used cut and paste and didn't finish the job :<) --- Subrata Sircar|ssircar@canon.com (NextMail ok)|Prophet & SPAMIT Charter Member Canon Information Systems and I do not share the same views on everything. "I'm just mad that I missed the sexual revolution." - me "Yes, but you dress much better as a result." - Mike
From: andrewd@tamsun.tamu.edu (Andrew T Duchowski) Newsgroups: comp.sys.next.bugs Subject: getty under 3.0 - bug? Date: 18 Feb 1993 20:35:19 -0600 Organization: Texas A&M University, College Station, TX Distribution: usa Message-ID: <1m1h17INNr3u@tamsun.tamu.edu> I recently tried setting up my dovefax modem to be able to dial in. If you're familiar with this issue, you'll know that among other things there are two steps: 1. set the ttyda line to 'on' in /etc/ttys, 2. install a kill file: -------------------- kill file --------------------- #! /bin/sh -u PIDS=`ps -ax | bm getty | grep -v bm | awk '{print $1}'` kill -TERM $PIDS kill -HUP 1 ----------------------------------------------------- and run every 15 mintues to get rid of a phantom lock that would lock up /dev/cua. Essentially, the kill file would permit one to share the serial port between dialing in and dialing out. This seems to work under 2.0 since I have it set up this way with an abaton modem. But under 3.0, as soon as I set up the ttyda to 'on', the getty process starts up (after kill -HUP 1) and then I can no longer dial out (using kermit -- kermit says /dev/cua is busy). The kill file doesn't help, nor does kill -9 of the getty process (after resetting the /etc/ttys file to off). This latter point is somewhat alarming since I can't seem to unfreeze the serial port even if I reset the ttys file and kill -HUP 1. The only way I can free up /dev/cua is to reboot. Is this some kind of bug in 3.0? Please reply by email to andrewd@cs.tamu.edu. Thanks in advance, Puzzled.... -- -- Andrew Duchowski | -- -- Texas A&M University | This page intentionally blank -- -- andrewd@cs.tamu.edu (non-NeXT mail) | -- -- andrewd@visual2.cs.tamu.edu (NeXT mail) | --
Newsgroups: comp.sys.next.bugs From: mikhe@ida.liu.se (Mike Henry) Subject: chunkPackage under 3.0 munged... Message-ID: <1993Feb18.231854.13297@ida.liu.se> Sender: news@ida.liu.se Organization: CIS Dept, Univ of Linkoping, Sweden Date: Thu, 18 Feb 1993 23:18:54 GMT I don't know if this is a known bug or not but chunkPackage (under /NextAdmin/Installer.app) dies when the destination directory is in the current working directory as where the command is being run from. What happens is that chunkPackage cd's into the package to be chunk'ed and promptly forgets that the "chunks" directory is "above" it. Anyway symptoms are that chunkPackage dies with the message: localhost> chunkPackage MyApp.pkg 2400 -d ./MyApp.chunks Chunking MyApp.pkg ... creating archive chunks ... done. destPkg: Undefined variable. Of course using a full path solves the problem but is annoying most of the time. Also it's too bad the example from the manual doesn't work... -Mike -- Mike Henry INET : mikhe@ida.liu.se /// August Wahlstromsv. 4 /// S-182 31 Danderyd \\\/// SWEDEN TEL : +46 8 755-8687 \XX/
From: shuque@microsupport.sas.upenn.edu (Shumon Huque) Newsgroups: comp.sys.next.bugs Subject: Re: bug with bash 1.12.1 on NS3.0 Message-ID: <110377@netnews.upenn.edu> Date: 19 Feb 93 17:41:56 GMT References: <CEDMAN.93Feb17143757@capitalist.princeton.edu> Sender: news@netnews.upenn.edu > If you type: > cd /etc; pwd > 10 times, the pwd stops returning anything after a couple of > times, and often bombs out with a malloc/free fault. > Even on danger of repeating myself, I have to point out that if you > are used to bash, you should switch to zsh instead. The interactive > features are comparable and in all programming respects zsh is much > more similar to bash than tcsh is. Zsh also doesn't have this bash > bug. :-) > > > Carl Edman Yes, this is a rather annoying bug. I still like bash however. Carl, you did not point out *why* zsh is better than bash, merely that it has comparable features .. Convince me and I'll switch :) Cheers, --Shumon School of Engineering & Applied Science, University of Pennsylvania. shuque@spectre.sas.upenn.edu
From: shuque@microsupport.sas.upenn.edu (Shumon Huque) Newsgroups: comp.sys.next.bugs Subject: Re: bug with bash 1.12.1 on NS3.0 Message-ID: <110378@netnews.upenn.edu> Date: 19 Feb 93 17:42:57 GMT References: <CEDMAN.93Feb17143757@capitalist.princeton.edu> Sender: news@netnews.upenn.edu > Even on danger of repeating myself, I have to point out that if you > are used to bash, you should switch to zsh instead. The interactive > features are comparable and in all programming respects zsh is much > more similar to bash than tcsh is. Zsh also doesn't have this bash > bug. :-) > > > Carl Edman Yes, this is a rather annoying bug. I still like bash however. Carl, you did not point out *why* zsh is better than bash, merely that it has comparable features .. Convince me and I'll switch :) Cheers, --Shumon School of Engineering & Applied Science, University of Pennsylvania. shuque@spectre.sas.upenn.edu
From: heading@signal.dra.hmg.gb (Anthony J.R. Heading) Newsgroups: comp.sys.next.bugs Subject: Preview buggers %%Page... (3.0) Message-ID: <1m40d6INN11v@liszt.dra.hmg.gb> Date: 20 Feb 93 01:09:58 GMT Organization: Defence Research Agency Here's a section of context diff after a file has been printed from Preview.... 1c1,7 < %!PS-Adobe-2.1 --- > %!PS-Adobe-2.0 > %%Title: 5003.PPD_Spec.ps - /mnt/data2/heading/unix/printers/ps > %%Creator: Preview > %%CreationDate: Fri Feb 19 16:34:53 1993 > %%For: heading > %%DocumentFonts: Helvetica-Bold Helvetica Times-Italic Times-Roman Symbol Courier Courier-Bold Helvetica-Oblique Courier-Oblique Helvetica-BoldOblique Times-Bold > %%Pages: 158 1 3,4c9,10 < %%Pages: (atend) < %%DocumentFonts: (atend) --- > %%DocumentPaperSizes: Letter > %%Orientation: Portrait 5a12,16 > > %%Feature: *OutputOrder Normal > %%Feature: *OutputBin Upper > %%Feature: *Duplex DuplexNoTumble > %%Feature: *Resolution 600dpi 975c986,989 < %%Page: "i" 1 --- > > %%Page: 1 "i" > userdict /_NXPreviewSaveObject0 systemdict /save get exec put > %Page: "i" 1 2547c2561,2567 < %%Page: "ii" 2 --- > clear userdict /_NXPreviewSaveObject0 known { > _NXPreviewSaveObject0 systemdict /restore get exec > } if > > %%Page: 2 "ii" > userdict /_NXPreviewSaveObject0 systemdict /save get exec put > %Page: "ii" 2 Entirely contrary to both normal practice and the official spec, it's reversed the label and the page numbers on all the %%Page lines. This is a kind of trivial bug, but do NeXT know about it? More importantly, does it do it reliably such that I can put in a kludge into my PS filters? Anthony
Newsgroups: comp.sys.next.bugs,comp.sys.next.software,comp.sys.next.sysadmin From: cedman@princeton.edu (Carl Edman) Subject: A brief comparative study of UN*X shells (was: bug with bash 1.12.1 on NS3.0) In-Reply-To: shuque@microsupport.sas.upenn.edu's message of 19 Feb 93 17:42:57 GMT To: shuque@microsupport.sas.upenn.edu (Shumon Huque) Message-ID: <CEDMAN.93Feb19183209@capitalist.princeton.edu> Originator: news@nimaster Sender: news@Princeton.EDU (USENET News System) Organization: Princeton University References: <CEDMAN.93Feb17143757@capitalist.princeton.edu> <110378@netnews.upenn.edu> Date: Fri, 19 Feb 1993 22:32:09 GMT In article <110378@netnews.upenn.edu> shuque@microsupport.sas.upenn.edu (Shumon Huque) writes: > Even on danger of repeating myself, I have to point out that if you > are used to bash, you should switch to zsh instead. The interactive > features are comparable and in all programming respects zsh is much > more similar to bash than tcsh is. Zsh also doesn't have this bash > bug. :-) Yes, this is a rather annoying bug. I still like bash however. Carl, you did not point out *why* zsh is better than bash, merely that it has comparable features .. Convince me and I'll switch :) You are quite right of course. A long writeup of why zsh is the best of all possible shells has been on my TODO list for a long time. But as I haven't gotten around to it yet, let me give you the short version. In the beginning there was sh - the Bourne shell. Sh has a decent programming language and today exists probably on every single UNIX system in the universe. These two reasons alone dictate that virtually all system shell scripts on all UNIX systems (including the NeXT) as well as almost all portable shell scripts have been written in it (and those few which haven't been, should have been). Unfortunately sh has almost no features which would make it pleasant to use interactively. Basically it reads programs from the stdin and writes their output to stdout and that's that. So csh was born. Csh has a prompt. And aliases. And history substitution. And many other little things which make life for interactive users simpler -- on the NeXT it can even do limited command line editing. If the author of csh simply had put all these things on top of the basic programming language of sh, probably sh would no longer exist and all other shells would have descended from csh. Instead he got the idea that giving csh a programming language a little bit like a cross between C and sh would be a neat idea. What resulted is arguably the worst most convulted bug-ridden inconsistent unpredictable language ever to be created unintentionally (Intercal was intentionally designed to be like that, though some would argue that csh still beats it). For details on this, I refer you to Tom Christiansens writeup "Csh Programming Considered Harmful". From these two shells almost all other shells descend. Foremost among the csh descendants is tcsh (available via anon ftp from tesla.ee.cornell.edu). Tcsh adds many more nice interactive features to csh, among them real command line and history editing, filename and command name completion, spelling checking aso. As csh used to be the standard shell for interactive use very many users have upgraded to tcsh now for those reasons. It even fixes up some of the worst errors in the csh scripting language, but as long as its author, Christos Zoulas, insists on keeping it csh compatible there really isn't much one can do. The latest version of tcsh compiles on a NeXT running 3.0 without problems. The main descendant of the Bourne shell, is the Korn shell. Ksh is what Csh should have been. The ksh programming language extends sh by niceties like functions and local variables, but most sh programs run under it. Ksh also has advanced interactive features. Unfortunately, AT&T holds the copyright to ksh and NeXT has not licensed it from them which means that you probably can't get ksh without paying several thousand dollars for the source to AT&T. To remedy this situation at least two shells have been written which are by and large ksh compatible but publicly available. These are the FSF's bash (available via anon ftp to prep.ai.mit.edu) and Paul Falstad's zsh (available via anon ftp to cs.ucsd.edu) which these is maintained by the a mailing list of developers, zsh-list@cs.uow.edu.au. The latest releases of both should compile on a NeXT running 3.0 just fine (though bash requires a little twiddling). As they both strive to be ksh compatible their programming languages are almost identical. The difference lies in the interactive features. Bash has some very nice ones. Zsh has every single one which any other of the above shells (including bash and tcsh) has. It has coprocesses, spelling correction, functions, aliases, global aliases, variables, arrays, local variables and arrays, sh quoting, csh quoting, named user directories, arbitrarily named directories, arithmetic expansion, process substitution, complete sh history expansion, generalized filename patterns with almost the same expressivity as any find script, variable modifiers with the expressivity of sed, default commands, pre-commands, post-commands, directory change commands, periodic commands, scheduled commands, trapped exception commands, emacs emulating command line editor, vi emulating command line editor, re-bindable command keys, 152 editing commands in all, builtin help, random number generation, cdpath, directory stacks, ignored extensions, hostname completion, customized completion mail check, login/out check, autologout, option completion, left justified prompts, right justified prompts, directory in prompts, trailing directory component in prompt, shortest directory description using names in prompt, time&day in prompt, most of the functionality of time(1), automatic export, implicit 'cd', automatic menus on completion, automatic pushd, automatic removal of trailing slashes, automatic resume, automatic nicing of background jobs, optional chasing of symlinks, to name just a few of the things which come to mind. For a more extensive list I refer you to the man page and info page and introductory text which comes with every zsh (though usually there are even more options which just nobody has gotten around to documenting). In short, there is no reason in the world _not_ to use zsh. For people who worry about software bloat, here are the data on my system. -rwxr-xr-x 1 root wheel 45120 May 30 1992 /usr/local/bin/rc* -rwxr-xr-x 1 root wheel 106496 Jul 21 1992 /bin/csh* -rwxr-xr-x 2 root wheel 122880 Jul 21 1992 /bin/sh* -rwxr-xr-x 1 cedman wheel 204800 Feb 14 11:52 /usr/local/bin/zsh* -rwxrwxr-x 1 cedman wheel 204800 Feb 6 22:45 /usr/local/bin/tcsh* -rwxr-xr-x 1 cedman wheel 221184 Feb 14 13:27 /usr/local/bin/bash* I'll let you know if I ever finish the long writeup. Carl Edman
Newsgroups: comp.sys.next.bugs From: uad1220@dircon.co.uk (Antony Freeman) Subject: HELP- Problem copying to DOS floppy Date: Sun, 21 Feb 1993 12:55:42 GMT Message-ID: <C2suKv.A4u@dircon.co.uk> Summary: Suddenly I can't copy any files to my DOS floppy Keywords: DOS floppy When I try to copy files to a DOS floppy, I get the error 'Invalid Argument' This is the same whether I copy from the workspace, terminal window, or through FloppyWorks. The funny thing is I can write to the Floppy if its UNIX or Mac format. This is most odd as last week I was copying no problem, the only thing I did since then was to try to install SLIP (without success so far) maybe I fouled something up on the way. Any suggestions?? Antony Freeman Data Text (UK) +44 892 861222 uad1220@dircon.co.uk antony@datatext.demon.co.uk
Newsgroups: comp.sys.next.bugs From: stan@netcom.com (Stanley Perkins) Subject: Re: HELP- Problem copying to DOS floppy Message-ID: <1993Feb21.164908.13415@netcom.com> Keywords: DOS floppy Organization: Netcom - Online Communication Services (408 241-9760 guest) References: <C2suKv.A4u@dircon.co.uk> Date: Sun, 21 Feb 1993 16:49:08 GMT I had the same problem a few days ago... It seems that, even after making sure that the file name was a valid DOS name (8chars.3chars) I still got the "Invalid Argument" error. When I changed the name from all uppercase to all lowercase, it just worked! :-) Stan
Newsgroups: comp.sys.next.bugs,comp.sys.next.software,comp.sys.next.sysadmin From: irving@Happy-Man.com (Irving_Wolfe) Subject: Re: A brief comparative study of UN*X shells (was: bug with bash 1.12.1 on NS3.0) Message-ID: <1993Feb21.170707.13545@Happy-Man.com> Organization: Happy Man Corp., Vashon Island, WA 98070-7399 References: <CEDMAN.93Feb17143757@capitalist.princeton.edu> <110378@netnews.upenn.edu> <CEDMAN.93Feb19183209@capitalist.princeton.edu> Date: Sun, 21 Feb 1993 17:07:07 GMT Just so Carl Edman isn't alone in this, let me add that I've been using zsh since he introduced me to it some months ago, and have made it the standard interactive shell on our systems. I think only one person still uses csh. Try it, you'll like it! Regards, - Irving -- Irving_Wolfe@Happy-Man.com Happy Man Corp. 206/463-9399 x101 4410 SW Pt. Robinson Rd., Vashon Island, WA 98070-7399 fax x108 We publish SOLID VALUE for the intelligent investor. NextMail OK Info free; sample $20. Send POSTAL addrs: Solid-Value@Happy-Man.com
Newsgroups: comp.sys.next.bugs,comp.sys.next.software,comp.sys.next.sysadmin From: davisre@sage.cc.purdue.edu (Robert Davis) Subject: Re: A brief comparative study of UN*X shells (was: bug with bash 1.12.1 on NS3.0) Message-ID: <C2ttyB.6Bt@mentor.cc.purdue.edu> Sender: news@mentor.cc.purdue.edu (USENET News) Organization: Purdue University Computing Center References: <110378@netnews.upenn.edu> <CEDMAN.93Feb19183209@capitalist.princeton.edu> <1993Feb21.170707.13545@Happy-Man.com> Date: Mon, 22 Feb 1993 01:39:46 GMT In article <1993Feb21.170707.13545@Happy-Man.com> Irving_Wolfe@Happy-Man.com writes: >Just so Carl Edman isn't alone in this, let me add that I've been >using zsh since he introduced me to it some months ago, and have >made it the standard interactive shell on our systems. I think >only one person still uses csh. > >Try it, you'll like it! I'll add a me-too to that. I've been using zsh for almost a year and a half and it really is great. One thing that I've heard some people (mostly csh/tcsh users) say is that if you don't like csh's programming syntax then you can just use Bourne's shell to program in but use something like tcsh interactively. But I frequently write mini-scripts interactively -- using for and while from the command line -- and it sure is nice to have a shell like zsh that gives you a great interactive environment without compromising the programming syntax. Plus the author has always made an effort to make it easy to compile on a NeXT. Many months ago there was a document at convex.com about why csh syntax is not very good to program in -- if it's still there, check it out. Rob -- | Robert Davis davis@sonata.cc.purdue.edu | "Look up, Hannah. Look up." NeXT Mail accepted --
From: maurices@spock.dis.cccd.edu (Maurice Shihadi) Newsgroups: comp.sys.next.bugs Subject: Calendar is wrong. Date: 22 Feb 1993 02:47:39 -0800 Organization: Coast Community College District, Costa Mesa, CA Message-ID: <1mab0bINNbrf@spock.dis.cccd.edu> When I first ubgraded from 2.1 everything was great but now that I gotten a new HD and installed 3.0 the dates on my months for 1993 are incorrect. Can this be fixed or do I have to remain in 1971 to keep my dates correct? maurices
Newsgroups: comp.sys.next.bugs From: sawada@etlst1.etl Subject: Re: atoh() and atoo() missing from NeXT libraries? Content-Type: text/plain Message-ID: <sawada.93Feb2316304@etlst1.etl> Sender: news@etl.go.jp (News System) Organization: Electrotechnical Laboratory, Tsukuba Science City References: <1993Feb10.114823.5815@ida.liu.se> Mime-Version: 1.0 Distribution: comp Date: Tue, 23 Feb 1993 07:30:04 GMT Sorry, but this is not the answer to the original poster. I have similar problem. > Are the atoh() and atoo() functions missing from the NeXT > libraries? They're included in the header files but cannot > be found for linking. I've grep'ed through all the > libraries in /lib and /usr/lib but to no avail. Is this > a known "bug" or am I totally beclouded? And, does 'fileno()' exist in the libraies? I also 'nm'ed /usr/lib/*.a and /lib/*.a , but could not find it. It seems a nightnamare.... -- sawada@etl.go.jp Computer Language Section Electrotechnical Laboratory, on leave from Software Research Associates Inc.
Newsgroups: comp.sys.next.bugs From: sawada@etl.go.jp Subject: Re: atoh() and atoo() missing from NeXT libraries? Content-Type: text/plain Message-ID: <sawada.93Feb23182230@etlst1.etl> Sender: news@etl.go.jp (News System) Organization: Electrotechnical Laboratory, Tsukuba Science City References: <1993Feb10.114823.5815@ida.liu.se> <sawada.93Feb2316304@etlst1.etl> Mime-Version: 1.0 Distribution: comp Date: Tue, 23 Feb 1993 09:22:30 GMT I article <sawada.93Feb2316304@etlst1.etl> I said: > > Are the atoh() and atoo() > > functions missing from the NeXT > libraries? They're > > included in the header files but cannot > > be found for > > linking. I've grep'ed through all the > > libraries in > > /lib and /usr/lib but to no avail. Is this > a known > > "bug" or am I totally beclouded? > > And, does 'fileno()' exist in the libraies? I also 'nm'ed > /usr/lib/*.a and /lib/*.a , but could not find it. It > seems a nightnamare.... Sorry for my ignorance. My problem was caused by the -ansi and -bsd combination of command arguments of cc (see stdio.h). -- sawada@etl.go.jp Computer Language Section Electrotechnical Laboratory, on leave from Software Research Associates Inc.
From: cjs@po.CWRU.Edu (Christopher J. Seline) Newsgroups: comp.sys.next.bugs,comp.sys.next.hardware Subject: 3.0 Freezes at "Checking Files" Date: 23 Feb 1993 11:53:16 GMT Organization: Case Western Reserve University, Cleveland, OH (USA) Message-ID: <1md37cINNh3g@usenet.INS.CWRU.Edu> My slab has frozen twice; it freezes several seconds into "checking disks"; the animated icon freezes, the harddrive stops making noise and I can not get it to go into the monitor. Any clues? (help) cjs@occs.cs.oberlin.edu thanks in advance!
From: tobi@pcmp.caltech.edu (Tobi Delbruck) Newsgroups: comp.sys.next.bugs Subject: what is analagous to rpc.statd? Date: 24 Feb 1993 00:55:34 GMT Organization: California Institute of Technology, Pasadena Message-ID: <1meh26INNa5g@gap.caltech.edu> our hp hp-ux machines are having a hard time rpc.stad'ing with our next machine running 2.x mach. is there a program on the next machine that is supposed to talk to or interact with rpc.statd? any help appreciated! -tobi delbruck
Newsgroups: comp.sys.next.bugs Subject: Re: NFS bug, overwritten file's contents not seen. Message-ID: <1993Feb22.152459.15649@urz.unibas.ch> From: frank@ifi.unibas.ch (Robert Frank) Date: Mon, 22 Feb 1993 15:24:59 GMT Sender: news@urz.unibas.ch (USENET News System) References: <1993Feb18.210755.17466@julian.uwo.ca> Organization: Institut fuer Informatik In article <1993Feb18.210755.17466@julian.uwo.ca> writes: > Howdy Folk; > > Perhaps others have seen this, this may even be a FAQ, but > here is the situtation. I have a NextSTation running NS 2.1 as a NFS > client to a Personal Iris (IRIX 4.0.1). I created a colour TIFF file > that the NeXT could not understand so I decided to convert it to a GIF > file. To do this I issued, on the iris: > > #> fromtiff image.tif image.rgb > #> togif image.rgb image.gif > > and then moved the file to /u2/tmp (a NFS shared directory) and read > it on the NeXT only to find that the fromtiff had misread the file and > switched black for white. No problem, as soon as I looked at the file > I knew what had happened (I had done this before) so I, instead, > displayed the file via imgview and the captured the image via > snapshot, issued the togif command and then copied it back to /u2/tmp > UNDER THE SAME NAME. > > So I looked at the new one again and it was STILL screwed up. > I was pretty sure that this file was correct, I could display it on > the iris perfectly. As it happens I have two next machines on this > network so I looked at the file via the other next box and voila it > displayed perfectly. At this point I figured that for some reason the > next was looking at a copy of the file from swap (it DIDN'T reside on > the disk anymore) even though I DID issue a `Update Viewer' command. > SO, I figured that what I needed to do was to somehow convince the > Next to ACTUALLY go and read the file and I thought I could do this by > coping the file. > > At a terminal prompt on the next, I issued a: > #> cp image.gif newimage.gif > and then looked at newimage.gif and it TOO was screwed up! So then, > from the iris, I issued a: > #> diff image.gif newimage.gif > They SHOULD be the same, right? Wrong! newimage.gif was a copy of a > file that did not exist anywhere EXCEPT in the next box's swapfile (or > biod buffers I guess)! > > A few points: /u2/tmp resides on the iris and is NFS > mounted onto the next; the time between putting the new copy of > image.gif and issuing the cp image.gif newimage.gif is over 1 hour (I > went to lunch after issuing the second togif command); I logged out > (twice) before I finally recopied the file (but this time on the > iris), obviously to no avail; all of the machines on the network use > NTP so this isn't a datestamp problem (not that it is likely to have > an hour difference in time ;-(). > > Now, this seems to me to be a NFS bug on NeXT's part. I could > understand (sort of) the confusion as the name was the same but even > after I told the next to re-read the directory (`Update Viewer')?! > > So, is there some option on the mount(2) command and/or biod(8) to > ensure that this doesn't happen? If not, will upgrading to NS 3.0 fix > this (not that I really want to do that ;-()? > > Sorry for the long-winded explanation, but I wanted to make sure that > I explained the behaviour correctly. > > cheerskinch > Dave Kinchlea > Snail-mail: System Administrator/Research Programmer > John P. Robarts Research Institute. > > ".... 2 to the power 759 million to 1 and falling ..." HHGTG > > I'm getting simular results with PB and IB over the network. I work on a client accessing files on the nfs server (all NeXTs). Updating a file in edit doesn't seem to clear out the image stored somewhere on my client. If I then say 'build' or 'run', PB simply ignores the changes and runs the old version of the program. If I explicitly 'touch' the files, everything works out fine almost all the time. I, too, could use some info on the automounter! -Robert -- Robert Frank tel. + (061) 321 99 67 Institut fuer Informatik fax + (061) 321 99 15 University of Basel, Switzerland Mittlere Strasse 142 rfc822: frank@ifi.unibas.ch (NeXT mail accepted) CH-4056 Basel X400: S=frank;OU=ifi;O=unibas;P=switch;A=arcom;C=ch ( if all fails try frank@urz.unibas.ch )
Newsgroups: comp.sys.next.bugs Subject: Re: NFS bug, overwritten file's contents not seen. Message-ID: <1993Feb23.084017.29682@urz.unibas.ch> From: frank@ifi.unibas.ch (Robert Frank) Date: Tue, 23 Feb 1993 08:40:17 GMT Sender: news@urz.unibas.ch (USENET News System) References: <HARDY.93Feb18141004@golem.ps.uci.edu> Organization: Institut fuer Informatik In article <HARDY.93Feb18141004@golem.ps.uci.edu> writes: > The same thing was true, e.g., for TeXview under 2.1 -- If I modified > the dvi-file on a remote host, and TeXviewed the NFS-mounted dvi-file, > the updates were never displayed. Somehow, NFS "cached" the old file > and never displayed the update, unless I saved it under a different > name, and reopened it. > > This "feature" has disappered under NS 3.0 -- and things behave > normally. So: upgrade to 3.0. *I* AM using 3.0!!! It DOESN't work for me! > -- > Hardy > ----- > Meinhard E. Mayer, Department of Physics, UC Irvine > e-mail: hardy@golem.ps.uci.edu (preferred) or MMAYER@UCI.BITNET > !!!! NO NEXTMAIL TO THESE ADDRESSES, PLEASE !!!!! -- Robert Frank tel. + (061) 321 99 67 Institut fuer Informatik fax + (061) 321 99 15 University of Basel, Switzerland Mittlere Strasse 142 rfc822: frank@ifi.unibas.ch (NeXT mail accepted) CH-4056 Basel X400: S=frank;OU=ifi;O=unibas;P=switch;A=arcom;C=ch ( if all fails try frank@urz.unibas.ch )
From: robert@steffi.demon.co.uk (Robert David Nicholson) Newsgroups: comp.sys.next.bugs Subject: Re: atoh() and atoo() missing from NeXT libraries? Date: 23 Feb 1993 15:52:30 -0000 Organization: me organized? That's a joke! Distribution: comp Message-ID: <1mdh7u$4mo@steffi.demon.co.uk> References: <sawada.93Feb2316304@etlst1.etl> sawada@etlst1.etl wrote: : Sorry, but this is not the answer to the original poster. : I have similar problem. : > Are the atoh() and atoo() functions missing from the NeXT : > libraries? They're included in the header files but cannot : > be found for linking. I've grep'ed through all the : > libraries in /lib and /usr/lib but to no avail. Is this : > a known "bug" or am I totally beclouded? : And, does 'fileno()' exist in the libraies? I also 'nm'ed : /usr/lib/*.a and /lib/*.a , but could not find it. : It seems a nightnamare.... : -- : sawada@etl.go.jp Computer Language Section : Electrotechnical Laboratory, : on leave from Software Research Associates Inc. Well in NS3.0 atoh and atoo and fileno are present. Have you a bad MANPATH environment variable set or something?
Newsgroups: comp.sys.next.bugs From: dan@ennex1.eng.utsa.edu (Daniel Shelton) Subject: Re: 3.0 Freezes at "Checking Files" Message-ID: <1993Feb24.191224.18375@ringer.cs.utsa.edu> Sender: news@ringer.cs.utsa.edu Organization: Univ of Texas at San Antonio References: <1md37cINNh3g@usenet.INS.CWRU.Edu> Date: Wed, 24 Feb 1993 19:12:24 GMT In article <1md37cINNh3g@usenet.INS.CWRU.Edu> cjs@po.CWRU.Edu (Christopher J. Seline) writes: > > My slab has frozen twice; it freezes several seconds into "checking disks"; > the animated icon freezes, the harddrive stops making noise and I can not get > it to go into the monitor. > > Any clues? (help) > > cjs@occs.cs.oberlin.edu > > thanks in advance! You need to make the boot sequence verbose in order to narrow the problem down a little more. You do this by typing holding the command keys down and at the same time type the ~ on the numeric keypad. This will put you in the NMI monitor, type halt which will put you into the NeXT monitor. Type p and then answer yes when the boot verbose question comes up. Then reboot the machine and find out where exactly the machines hangs. Dan this point you should be in a blem down a little more. You do this by
Newsgroups: comp.sys.next.bugs From: jjfeiler@relief.com (John Jay Feiler) Subject: Re: A brief comparative study of UN*X shells (was: bug with bash 1.12.1 on NS3.0) Message-ID: <1993Feb23.071024.1494@relief.com> Sender: jjfeiler@relief.com Organization: relief consulting References: <1993Feb21.170707.13545@Happy-Man.com> Date: Tue, 23 Feb 1993 07:10:24 GMT In article <1993Feb21.170707.13545@Happy-Man.com> irving@Happy-Man.com (Irving_Wolfe) writes: > Just so Carl Edman isn't alone in this, let me add that I've been > using zsh since he introduced me to it some months ago, and have > made it the standard interactive shell on our systems. I think > only one person still uses csh. > > Try it, you'll like it! > > Regards, > - Irving > -- > Irving_Wolfe@Happy-Man.com Happy Man Corp. 206/463-9399 x101 > 4410 SW Pt. Robinson Rd., Vashon Island, WA 98070-7399 fax x108 > We publish SOLID VALUE for the intelligent investor. NextMail OK > Info free; sample $20. Send POSTAL addrs: Solid-Value@Happy-Man.com OK, I saw the light, and downloaded zsh. Looks good. But I've got one question: does anyone know of a way to get the path into the titlebar of the Terminal.app window? I know you can set it with the preferences panel, I'm just wondering if there's a public, semi-public, or even private API for doing such a thing...... Thanks, John -- John Feiler jjfeiler@relief.com 4926 152nd St. SW NeXTmail Welcome!!! Edmonds, WA 98026-3344 Independent NeXTSTEP Developer
Newsgroups: comp.sys.next.bugs From: lmccullo@nyx.cs.du.edu (Michael McCulloch) Subject: Re: NFS bug, overwritten file's contents not seen. Message-ID: <1993Feb24.213204.15755@mnemosyne.cs.du.edu> Sender: usenet@mnemosyne.cs.du.edu (netnews admin account) Organization: University of Denver, Dept. of Math & Comp. Sci. Date: Wed, 24 Feb 93 21:32:04 GMT In message <1993Feb22.152459.15649@urz.unibas.ch> frank@ifi.unibas.ch (Robert Frank) writes: >In article <1993Feb18.210755.17466@julian.uwo.ca> writes: >> >> Perhaps others have seen this, this may even be a FAQ, but >> here is the situtation. I have a NextSTation running NS 2.1 as a NFS >> client to a Personal Iris (IRIX 4.0.1). I created a colour TIFF file [... deleted] >> So I looked at the new one again and it was STILL screwed up. >> I was pretty sure that this file was correct, I could display it on >> the iris perfectly. As it happens I have two next machines on this >> network so I looked at the file via the other next box and voila it >> displayed perfectly. At this point I figured that for some reason the >> next was looking at a copy of the file from swap (it DIDN'T reside on >> the disk anymore) even though I DID issue a `Update Viewer' command. >> SO, I figured that what I needed to do was to somehow convince the >> Next to ACTUALLY go and read the file and I thought I could do this by >> coping the file. >> >> A few points: /u2/tmp resides on the iris and is NFS >> mounted onto the next; the time between putting the new copy of >> image.gif and issuing the cp image.gif newimage.gif is over 1 hour (I >> went to lunch after issuing the second togif command); I logged out >> (twice) before I finally recopied the file (but this time on the >> iris), obviously to no avail; all of the machines on the network use >> NTP so this isn't a datestamp problem (not that it is likely to have >> an hour difference in time ;-(). >> >> Now, this seems to me to be a NFS bug on NeXT's part. I could >> understand (sort of) the confusion as the name was the same but even >> after I told the next to re-read the directory (`Update Viewer')?! >> >> So, is there some option on the mount(2) command and/or biod(8) to >> ensure that this doesn't happen? If not, will upgrading to NS 3.0 fix >> this (not that I really want to do that ;-()? >> >> Sorry for the long-winded explanation, but I wanted to make sure that >> I explained the behaviour correctly. >> > >I'm getting simular results with PB and IB over the network. I work on a client >accessing files on the nfs server (all NeXTs). Updating a file in edit doesn't >seem to clear out the image stored somewhere on my client. If I then say >'build' or 'run', PB simply ignores the changes and runs the old version of the >program. If I explicitly 'touch' the files, everything works out fine almost >all the time. > > >-Robert >-- I had the same problem with NFS'ed directories mounted on my NeXT from an SGI running IRIX 4.0.5. My NeXT is running 3.0. julian.uwo.ca's problems sound like the same inexplicable behavior I was encountering. Please check the /etc/TIMEZONE file on your SGI server. Make sure that that file has the correct description of your local timezone -- 'cause if it doesn't, your NeXT will incorrectly timestamp files from some apps (Edit.app was problematic for me) saved to the NFS'ed directories. Also, setting TZ in a user's shell had no effect. NFS's daemon evidently checks the system default timezone. If you've checked the TIMEZONE file on your SGI, and you still experience problems after an upgrade to NS 3.0 -- then I recommend going into Preferences.app and resetting your timezone using both the map view and pop-up buttons. I don't know for sure if it's a bug -- but my timezone preferences seemed to have been blown away by the upgrade to 3.0. Checking these two things (after lots of hair pulling) solved my NFS timestamp problems. BTW, like I said above, some apps would timestamp the files OK and some wouldn't -- which made tracking down the problem quite confusing. The SGI and NeXT would even report different modification times on the same files!!!! Michael McCulloch Huntsville, AL
Newsgroups: comp.sys.next.bugs,comp.sys.next.software From: lmccullo@nyx.cs.du.edu (Michael McCulloch) Subject: Diagram2! and Custom Palettes Message-ID: <1993Feb24.215348.16535@mnemosyne.cs.du.edu> Sender: usenet@mnemosyne.cs.du.edu (netnews admin account) Organization: Nyx, Public Access Unix @ U. of Denver Math/CS dept. Distribution: na Date: Wed, 24 Feb 93 21:53:48 GMT I'm having problems with creating custom palettes using Diagram2!. I've used several of the graphics primitives to create several grouped graphics, which I then want to use as a palette. I've experienced the following problems: 1. Grid alignment works sporadically when dragging from a palette to a new document. 2. Graphics consisting of grouped graphic primitives don't draw correctly after dragging them from the palette to a new document. 3. When dragging from a palette to a new document, the imaged graphic in the new document is placed with some offset from the point at which I release the mouse. Sometimes the graphic images itself off the page (took me a while to figure out what was happening). The above problems are extremely irritating. Has anyone else noticed the same behavior? My *very* enthusiastic opinion of Diagram2! has been somewhat squelched by these problems. Michael McCulloch Huntsville, AL
From: goldly@u.washington.edu (Lloyd P. Goldwasser) Newsgroups: comp.sys.next.bugs Subject: 3.0 and DOS disks Date: 24 Feb 1993 23:26:56 GMT Organization: University of Washington Message-ID: <1mh080INNjmc@shelley.u.washington.edu> After I upgraded to 3.0, my '040 Cube stopped dealing with DOS disks. When I put them in the drive (a PLI Superfloppy), it does nothing with them: no icon, no mounting, no way to eject them (logging off and rebooting don't even do the trick). Is there something somewhere that I can do to convince the machine that DOS disks are not something to eat? Thanks, Lloyd Goldwasser goldly@u.washington.edu
Newsgroups: comp.sys.next.bugs From: dave@prim.demon.co.uk (Dave Griffiths) Subject: Handling of invalid method calls for DO's Organization: Primitive Software Ltd. Apparently-To: mail2news@dis.demon.co.uk Date: Wed, 24 Feb 1993 14:10:15 +0000 Message-ID: <1993Feb24.141015.3628@prim> Sender: usenet@demon.co.uk Another "feature": if you try to send a distributed object a message it doesn't respond to, your client has an exception raised and the message: Unknown error code 11004 in NXReportError appears on the console. This is inconsistent with the behaviour for sending local objects unknown messages, which more helpfully results in a "blah blah does not respond to blah blah" message on the console and your app dies. BTW, I haven't started using protocols yet. Dave Griffiths
Newsgroups: comp.sys.next.bugs From: mikhe@ida.liu.se (Mike Henry) Subject: Re: atoh() and atoo() missing from NeXT libraries? Message-ID: <1993Feb25.131450.23078@ida.liu.se> Sender: news@ida.liu.se Organization: CIS Dept, Univ of Linkoping, Sweden References: <sawada.93Feb2316304@etlst1.etl> <1mdh7u$4mo@steffi.demon.co.uk> Distribution: comp Date: Thu, 25 Feb 1993 13:14:50 GMT Sorry to have to do this to you... robert@steffi.demon.co.uk (Robert David Nicholson) writes: >Well in NS3.0 atoh and atoo and fileno are present. >Have you a bad MANPATH environment variable set or something? When I originally wrote the article questioning the whereabouts of these functions I wasn't denying their existance in the manpages or header files, on the contrary. What I stated was that these functions are missing from the *LIBRARIES* under 3.0. I.e. have you actually compiled any programs containing these functions under 3.0 to verify your statement? I don't seriously think that (re)setting my MANPATH is going to make any difference for purposes of linking... Have a nice one! B^) -Mike -- Mike Henry INET : mikhe@ida.liu.se /// August Wahlstromsv. 4 /// S-182 31 Danderyd \\\/// SWEDEN TEL : +46 8 755-8687 \XX/
Newsgroups: comp.sys.next.bugs From: kinch@newshost.uwo.ca (Dave Kinchlea) Subject: Re: NFS bug, overwritten file's contents not seen. Organization: The University of Western Ontario, London, Canada. Date: Thu, 25 Feb 1993 23:02:42 GMT Message-ID: <KINCH.93Feb25180242@valve.newshost.uwo.ca> In-Reply-To: lmccullo@nyx.cs.du.edu's message of Wed, 24 Feb 93 21:32:04 GMT References: <1993Feb24.213204.15755@mnemosyne.cs.du.edu> Sender: news@julian.uwo.ca (USENET News System) In article <1993Feb24.213204.15755@mnemosyne.cs.du.edu> lmccullo@nyx.cs.du.edu (Michael McCulloch) writes: julian.uwo.ca's problems sound like the same inexplicable behavior I was Just so there is no confusion (not that it really matters, but) julian.uwo.ca is not a NeXT box nor even on my LAN but is simply the newshost for the campus. Understandable mistake, however. cheerskinch
From: ngervae@next.com (Nik Gervae) Newsgroups: comp.sys.next.bugs,comp.sys.next.software,comp.sys.next.sysadmin Subject: Re: A brief comparative study of UN*X shells (was: bug with bash 1.12.1 on NS3.0) Message-ID: <6890@rosie.NeXT.COM> Date: 26 Feb 93 01:15:43 GMT References: <C2ttyB.6Bt@mentor.cc.purdue.edu> Sender: news@NeXT.COM Followup-To: comp.sys.next.bugs Just adding my "me too" to the list of satisfied users of zsh. If I had my way (which I don't :-), I'd have zsh shipped as the default shell for users. -- Nik "and no, that doesn't mean you should ask NeXT to do it" Gervae Software Publications ## S2/8bg+-l+y++-/:!z-n-++o-+/++:x-+:a++:u-+v+:j+-p(eR) NeXT Computer, Inc. ## B2(go-t & sburns)f-t+c-!g-(--)k--s---mr-p+ ngervae@next.com ## [[[[NXDisclaimer alloc] init] disclaim] free];
Newsgroups: comp.sys.next.bugs From: outland!jmast@ms.uky.edu (John A. Mastrolia) Subject: HELP: Improv "HotLink" fails under 3.0 Message-ID: <C31AGx.AG8@outland.uucp> Sender: jmast@outland.uucp (John A. Mastrolia) Organization: Kentucky NeXT Users Group (KYNUG) Date: Fri, 26 Feb 1993 02:19:44 GMT After upgrading to 3.0, I noticed recently, that Improv's "HotlinkPort" fails when creating a graph from Improv data. PresentationBuilder can't update graphs created with Improv data due to this little incinsistency. Anyone heard of a fix for this problem? Better yet, how about a point of contact for the NeXT version of Improv? I can't seem to find one in my documentation. Here's the exact console message: Workspace: Cannot launch program: 'HotlinkPort' Thanks in advance. -- John A. Mastrolia (Politically, Fashionably, and Aerodynamically Incorrect) outland!jmast@ms.uky.edu
From: mezzino@gauss.cl.uh.edu (Mike Mezzino) Newsgroups: comp.sys.next.bugs Subject: GatorFTP_1.1 Bug Date: 25 Feb 1993 19:08:49 GMT Organization: University of Houston Message-ID: <1mj5g1INN995@menudo.uh.edu> Hello Again (Unfortunately) Faithful GatorFTP Users: GatorFTP_1.1 has been removed from sonata because it contains a bug which only users new to GatorFTP will notice. It is related to the automatic version synchronization scheme embedded in the resource file called ".gatorftprc". Since we welcome new users as well as appreciate those who have helped to make this app better, the bug has been permanently (we hope) eradicated. You may, if you obtained it, temporarily avoid the bug by 1) deleting the hidden file .gatorftprc file in your home directory 2) launching GatorFTP 2) entering something, say anonymous, into the default Login Name in Preferences Because of the inconvenience, I will make GatorFTP_1.2 even faster! With abject apologies, Mike Mezzino mezzino@gauss.cl.uh.edu
Newsgroups: comp.music,comp.sys.next,comp.sys.next.bugs,comp.sys.next.programmercompcomp. From: schaik@cnplss5.cft.philips.nl (Willem van Schaik) Subject: Re: help with midi.... Message-ID: <1993Feb26.092839.3972@cnplss5.cnps.philips.nl> Keywords: midi Sender: news@cnplss5.cnps.philips.nl (USENET News System) Organization: Philips Communications & Processing Services, Eindhoven References: <1993Feb17.160955.5351@ghost.dsi.unimi.it> Date: Fri, 26 Feb 1993 09:28:39 GMT dany@lim.sm.dsi.unimi.it (Daniele Rizzi) writes: >Hello all, >I work on a 68040 Next computer running the latest 3.0 Release of >NextStep. >I would like to play a .midi file over the serial port, but when I use >playmidifile or Presto, I receive on the Console the error messages: >- (using port A) -> MIDIClaimUnit failed: can't init serial port. >- (using port B) -> MIDI driver hardware framing error >and, obviously, no message come to the Midi Expander; >I remember the fair times ('030 machine with release 2.0) when everything >worked well... >I thank you in advance for any help you may provide. >dany@lim.sm.dsi.unimi.it >ps. the Next serial port is used to be connected with an Apple Midi interface. Did you change your cable when you moved from 68030 to 68040?? Pin connections are different. Willem
Newsgroups: comp.sys.next.bugs From: kluge@avalon.physik.unizh.ch (kluge daniel) Subject: Autodiskmount Probleme Message-ID: <1993Feb26.151507.29505@ifi.unizh.ch> Sender: news@ifi.unizh.ch (USENET News Admin) Organization: University of Zurich, Department of Computer Science Date: Fri, 26 Feb 1993 15:15:07 GMT No, no, the problem ist not to eject the Disks, the Problem is, that the Disks get ejected to often! I often work with tar-disk, because of the larger space on the Disks, so I always get the phony "Disk is not readable, eject, format (in German of course...)"-Alert, ok, that I can handle, but it ocurs to often, that the second disk I insert gets ejected, and every other non-NeXT Disk as well. So I have to abort my multivolume tar-extraction, and reboot the machine. Last time I had to do this, there was a DOS-FS hanging, which could not be killed, so I had a 'dirty' reboot, with the occasional fsck - Thanx God I don't have my 1GB HD yet! *waiting, till 500MBs are checked..* Is there a workaround on its way, or is this another NeXT-immanent Feature ?? - daniel -- Daniel G. Kluge @ Swiss Federal Institute of Technology Zuerich E-Mail : kluge@avalon.physik.unizh.ch (NeXT-Mail welcome) study-related stuff : dankluge@iiic.ethz.ch DECnet : EZINFO::CLUESCH
Newsgroups: comp.sys.next.bugs,comp.sys.next.misc From: emurphy@cs.uno.edu (Edward Murphy) Subject: (?) how to stop all but one banner page Message-ID: <1993Feb26.150109.25057@cs.uno.edu> Sender: news@cs.uno.edu Organization: University of New Orleans Date: Fri, 26 Feb 1993 15:01:09 GMT Howdy (yep, my third attempt) Does anyone anywhere have any idea on how to stop the printing of multiple banner pages when multiple copies of a document is printed. I am in a university setting where multiple copies are printed often, and the ectra expense of the multiple banner pages is neither welcome nor needed. Thus far, the answers I have received tell me only things I have learned are that (1) NeXT uses the UNIX lpr command with the appropriate switches to do multiple copies (2) Alot of people want to know how to do the very same thing! One question I have is since PostScript code is generated in order to print the file on the NeXTprinter why not use the PostScript construct '/#copies'? Assuming the reqpresentatives from NeXT look at this list from time to time, I am hoping one of them will get in touch with me to either help me work out a patch for this or inform me of an existing patch for this. Also, any comment on the above problem will be appreciated. Ed Murphy P.S. Our setup: Network of several dozen NeXT machines, 2 NeXT printers, and NeXTstep 2.0 (passe', I know, but the powers that be cannot upgrade quite yet).
Control: newgroup comp.sys.next.bugs Newsgroups: comp.sys.next.bugs From: news@vnp.com Subject: newgroup comp.sys.next.bugs Message-ID: <1993Feb28.170625.26901@vnp.com> Organization: VNP Software Date: Sun, 28 Feb 93 17:06:25 GMT Non-empty. Path: informatik.uni-muenchen.de!lrz-muenchen.de!fauern!ira.uka.de!sol.ctr.columbia.edu!howland.reston.ans.net!bogus.sura.net!darwin.sura.net!haven.umd.edu!umd5!umd5.umd.edu!anderson From: anderson@sapir.cog.jhu.edu (Stephen Anderson) Newsgroups: comp.sys.next.bugs Subject: Re: A brief comparative study of UN*X shells (was: bug with bash 1.12.1 on NS3.0) Message-ID: <ANDERSON.93Mar1070245@sapir.cog.jhu.edu> Date: 1 Mar 93 12:02:45 GMT References: <C2ttyB.6Bt@mentor.cc.purdue.edu> <6890@rosie.NeXT.COM> Sender: news@umd5.umd.edu Followup-To: comp.sys.next.bugs Organization: Dept. of Cognitive Science, The Johns Hopkins University Lines: 17 In-reply-to: ngervae@next.com's message of 26 Feb 93 01:15:43 GMT Yes, zsh is indeed the greatest thing since sliced bread. I've greatly enjoyed itfor some time too. But it has one property that is somewhat annoying. Randomly (as far as I can tell), about 2-3 times per day of regular use, when you try to execute some command at the shell prompt, it comes back with the message that it (e.g. "ls" or something else exotic like that :-) is a "bad executable". All you have to do is hit ^P and run it again, and it always works fine the second time, but I've never been able to track down this problem. Still exists in the very latest version of zsh. Never happens with csh. At one point someone suggested a caching problem related to nfs mounting the directory with executables, but that can't be it in my case (I don't nfs mount anything at all). Do others have the same problem? And does anyone understand what's going on? --Steve Anderson
Newsgroups: comp.sys.next.bugs From: cedman@princeton.edu (Carl Edman) Subject: Re: A brief comparative study of UN*X shells (was: bug with bash 1.12.1 on NS3.0) In-Reply-To: anderson@sapir.cog.jhu.edu's message of 1 Mar 93 12:02:45 GMT To: anderson@sapir.cog.jhu.edu (Stephen Anderson) Message-ID: <CEDMAN.93Mar1074050@capitalist.princeton.edu> Followup-To: comp.sys.next.bugs Originator: news@nimaster Sender: news@Princeton.EDU (USENET News System) Organization: Princeton University References: <C2ttyB.6Bt@mentor.cc.purdue.edu> <6890@rosie.NeXT.COM> <ANDERSON.93Mar1070245@sapir.cog.jhu.edu> Date: Mon, 1 Mar 1993 11:40:50 GMT In article <ANDERSON.93Mar1070245@sapir.cog.jhu.edu> anderson@sapir.cog.jhu.edu (Stephen Anderson) writes: Yes, zsh is indeed the greatest thing since sliced bread. I've greatly enjoyed itfor some time too. But it has one property that is somewhat annoying. Randomly (as far as I can tell), about 2-3 times per day of regular use, when you try to execute some command at the shell prompt, it comes back with the message that it (e.g. "ls" or something else exotic like that :-) is a "bad executable". All you have to do is hit ^P and run it again, and it always works fine the second time, but I've never been able to track down this problem. Still exists in the very latest version of zsh. Never happens with csh. At one point someone suggested a caching problem related to nfs mounting the directory with executables, but that can't be it in my case (I don't nfs mount anything at all). Do others have the same problem? And does anyone understand what's going on? You are not the only one with this problem which seems to be NeXT specific. On the other hand there are people running the latest version of zsh without ever experiencing this problem which seems to indicate a configuration problem. But worry not, the top-notch shell researchers of zsh-list are hot on the trail of this problem and a resolution is expected to be forthcoming in the next few weeks. Carl Edman
Organization: Senior, Chemistry, Carnegie Mellon, Pittsburgh, PA Newsgroups: comp.sys.next.bugs Message-ID: <cfYbLQa00WB781C8BA@andrew.cmu.edu> Date: Mon, 1 Mar 1993 15:45:16 -0500 From: Charles William Swiger <infidel+@CMU.EDU> Subject: Re: A brief comparative study of UN*X shells (was: bug with bash 1.12.1 on NS3.0) In-Reply-To: <ANDERSON.93Mar1070245@sapir.cog.jhu.edu> Excerpts from netnews.comp.sys.next.bugs: 1-Mar-93 Re: A brief comparative stu.. by Stephen Anderson@sapir.c > Yes, zsh is indeed the greatest thing since sliced bread. I've greatly > enjoyed itfor some time too. Yep. > But it has one property that is somewhat annoying. Randomly (as far as > I can tell), about 2-3 times per day of regular use, when you try to > execute some command at the shell prompt, it comes back with the > message that it (e.g. "ls" or something else exotic like that :-) is a > "bad executable". All you have to do is hit ^P and run it again, and > it always works fine the second time, but I've never been able to > track down this problem. I just had "cp" complain about 5 minutes ago :-). Nope, I'm not NFS mounting anything either. -Chuck Charles William Swiger -- CMU...*crunch*! | "Foosh. Aaughh!!" ------------------------------------------+ "Foosh. Aauuggghh!!" AMS & normal mail: infidel@cmu.edu | "Cold spray deodorant...." Failing that: cs4w+@andrew.cmu.edu | NeXTmail: chuck@mon.slip.andrew.cmu.edu | -- Opus, Bloom County [RIP]
Newsgroups: comp.sys.next.bugs From: andylee@oahu.cs.ucla.edu (Andy Lee) Subject: Keyboard.app Bug Message-ID: <1993Mar3.090120.5007@cs.ucla.edu> Sender: usenet@cs.ucla.edu (Mr Usenet) Organization: UCLA, Computer Science Department Date: Wed, 3 Mar 93 09:01:20 GMT I think I've found a bug in the 3.0 Developer demo program Keyboard.app. I used it a few months ago to remap my keyboard to the Dvorak layout, now I've gotten used to the new layout, this happens: the letter 'z' (remapped to the "?/" key) is still lower case when I turn on the CapLock via Command-Shift. The same goes for all the letters that remapped to symbol keys, i.e. 'z' at the "?/" key, 's' at ":;", 'w' at "<,", and 'v' at ">.". Does anyone have a solution to this? I need to find a solution because I'll be typing in caps for the next few weeks (telnet'ing to an old old system which has all commands in caps). # Andy A. Lee # UCLA MS/PhD Graduate Student, AI Neural Networks # # andylee@cs.ucla.edu # Down-to-Earth Idealist & Independent NeXT Developer # # (310)824-7131 # "If you don't watch your figure, who will?" - TV Ad. # # (714)651-0763 # "Trust is earned, not given away." - Star Trek: TNG # -- # Andy A. Lee # UCLA MS/PhD Graduate Student, AI Neural Networks # # andylee@cs.ucla.edu # Down-to-Earth Idealist & Independent NeXT Developer # # (310)824-7131 # "If you don't watch your figure, who will?" - TV Ad. # # (714)651-0763 # "Trust is earned, not given away." - Star Trek: TNG #
From: wwille.hanse.de!wwille (Winfried Wille) Newsgroups: comp.sys.next.bugs,comp.sys.next.misc Subject: Re: zsh bad executable message fixed (was:A brief comparative study of UN*X shells ) Date: 3 Mar 1993 10:09:34 GMT Organization: Terraprolls Message-ID: <1n204u$9vq@wwille.hanse.de> References: <ANDERSON.93Mar1070245@sapir.cog.jhu.edu> In article <ANDERSON.93Mar1070245@sapir.cog.jhu.edu> anderson@sapir.cog.jhu.edu (Stephen Anderson) writes: > Yes, zsh is indeed the greatest thing since sliced bread. I've greatly > enjoyed itfor some time too. > > But it has one property that is somewhat annoying. Randomly (as far as > I can tell), about 2-3 times per day of regular use, when you try to > execute some command at the shell prompt, it comes back with the > message that it (e.g. "ls" or something else exotic like that :-) is a > "bad executable". All you have to do is hit ^P and run it again, and > ... > --Steve Anderson Hello, i posted the solution to the zsh-list (never reaches there because the domain hanse.de has lost it's main provider, but that will hopefully soon be fixed) and de.comp.sys.next- here it comes again: Date: Mon, 1 Mar 93 12:21:11 GMT+0100 From: wwille (To: zsh-list@cs.uow.edu.au) To: zsh-list@cs.uow.edu.au Subject: Re: bad executable or shared library - the Solution Cc: cedman@capitalist.princeton.edu, thunami, trohde dan@phoenix.az.stratus.com (Dan Danz) writes: >> [ Marc Guenther writes ] >> >> Does anyone know, what causes these errormessages ? >> >> it occurs to me all the time, that some commands cause this >> errormsg. When I execute the command a second >> time, it works. >> >> Im sitting in front of a Next, running NextStep 3.0 >> >> zsh version is 2.3.1 >> I got the some problems and mailed them to this list last december. I wrote to Carl Edman and described him an situation, where there error in 2 of 3 cases occured. I was so kind to test it, he had no problems. It couldn't be the binarie because since zsh2.3 it compiles and installs straight through, even under 3.0 - thanks to Carl Edman for that. Remains the runtime enviroment, i checked the changes i made. Failure is here, on my machine its in the /etc/zshrc -- snip --- # use hard limits, except for a smaller stack and no core dumps unlimit limit stack 8192 limit core 0 #limit -s -- snap -- the above works I had outcommented the "limit stack 8192" which causes an unlimited stacksize, with which zsh on an NeXT can't deal. Now it works all fine. Have fun with the famous zsh Winni --- * Winfried Wille * * email: wwille@wwille.hanse.de, NeXT mail accepted * * Voice: 040/456543, Data 040/457826 * * * * Kein Studieren ohne Kopieren (Spruch aus einem Copyshop) *
From: rbp@investor.pgh.pa.us (Bob Peirce #305) Newsgroups: comp.sys.next.bugs Subject: Re: HELP: Improv "HotLink" fails under 3.0 Message-ID: <1993Mar2.163717.25999@investor.pgh.pa.us> Date: 2 Mar 93 16:37:17 GMT References: <C31AGx.AG8@outland.uucp> Organization: Cookson, Peirce & Co., Pittsburgh, PA In article <C31AGx.AG8@outland.uucp> outland!jmast@ms.uky.edu (John A. Mastrolia) writes: >After upgrading to 3.0, I noticed recently, that Improv's "HotlinkPort" fails >when creating a graph from Improv data. PresentationBuilder can't update >graphs created with Improv data due to this little incinsistency. Anyone heard >of a fix for this problem? Better yet, how about a point of contact for the >NeXT version of Improv? I can't seem to find one in my documentation. > I just used PresentationBuilder to run a series of charts. The data was hot linked and I changed pieces of it for different versions of the charts. It worked fine. -- Bob Peirce, Pittsburgh, PA rbp@investor.pgh.pa.us 412-471-5320 venetia@investor.pgh.pa.us [NeXT Mail] !uunet.uu.net!pitt!investor!rbp [UUCP]
From: karuzis@wccf.mit.edu (GLENN HOLM) Newsgroups: comp.sys.next.bugs Subject: workspace preferences inspector color Date: 3 Mar 1993 21:13 EST Organization: MIT - Whitaker College Distribution: world Message-ID: <3MAR199321131944@wccf.mit.edu> News-Software: VAX/VMS VNEWS 1.41 i know it's a little thing, but the inspector in workspace preferrences that shows the dock items and whether they autolaunch at login has a miniature screen that should be the color of the default login panel background color, instead of the darker blue-gray that shows on my NSC. That way users who use Preferences to reset their workspace background color will be able to go back to the default lighter blue-gray using the magnifying glass in the color panel. ------------------------------------------------------------------ |Glenn Holm Internet:karuzis@wccf.mit.edu| |M.I.T Dept. of Brain + Cog. Sci. This VAX doesn't do NeXTmail | |Cambridge, MA 02139 "Real Neuroscientists don't do gels!" | ------------------------------------------------------------------
From: zac@dolphin.com (Zacharias J. Beckman) Newsgroups: comp.sys.next.bugs Subject: Checking Disk EVERY TIME! Help! Message-ID: <1993Mar4.032012.363@dolphin.com> Date: 4 Mar 93 03:20:12 GMT Sender: zac@dolphin.com Distribution: usa Organization: Dolphin Software Our server (NS 3.0) has recently, for no reason which I can imagine, begun to do the "Checking Disk" thing EVERY time it boots! Some research has shown that it apparently cannot unmount either the root device (/dev/sd0a) or the swapdrive (/dev/sd3a). Doing a "shutdown now" works... follow it with a "halt" and I see various error messages which lead me to believe this. The errors are followed by a string of "1 "'s and, eventually, a reboot. Unfortunately, the file system does not appear to be in a clean state... Also curious, our system has recently (the same time, I suspect) begun compaining about our tape drive... morning backups send mail off to the system administrator saying that the tape device (/dev/nrst0) is "Device Busy" and aborts the backup. This is the same message which I see when I try to "umount /private/swapdisk," although I can't find any rational explanation. If you have any ideas on what could be causing this, I would much appreciate your help! We have been running with continual fsck's and no backups for several days now! Thanks in advance. -- Zacharias J. Beckman - Dolphin Software Inc. - zac@dolphin.com - use NeXTMAIL! To be "matter of fact" about the world is to blunder into fantasy.... and dull fantasy at that, as the real world is strange and wonderful. --- R. A. Heinlen Those opinions I express herein are my own, I'm fairly sure. --- Z. J. Beckman
From: ahoag@hue.gac.edu (Andrew J Hoag) Newsgroups: comp.sys.next.bugs Subject: WP and Docs over 100 pgs Date: 3 Mar 1993 22:44:17 GMT Organization: Gustavus Adolphus College Message-ID: <1n3cc1INNpk@news.gac.edu> Keywords: wordperfect I am having problems with WordPerfect (1.0.1 - dated 25 Sept 92). I have been concatenating a series of documents and I have hit the 100-page mark. This is where WP has begun to act strange. Find & Replace will just up and terminate WordPerfect. I seems specific to Find with Replace, but I cannot nail down the circumstances. Anyone had similar problems? Any workaround suggestions? Someone suggested that there is a document-link feature, but I am not familiar with it. Any comments would be appreciated. Andrew Hoag AHOAG@NIC.GAC.EDU
Newsgroups: comp.sys.next.software,comp.sys.next.bugs From: mbparker@netcom4.com (Michael Benjamin Parker) Subject: Digital Librarian hangs, crashes, & lies; does it work for anyone? Message-ID: <MBPARKER.93Mar4080820@netcom4.com> Sender: mbparker@netcom.com (Michael Benjamin Parker) Organization: Netcom Online Communications Service Date: Thu, 4 Mar 1993 16:08:20 GMT Anyone have serious problems with Digital Librarian, especially when indexing and searching directories containing many types of files, not all text? I've had constant problems with bugs in Digital Librarian when ever I've tried to index stuff other than the pre-indexed stuff supplied by NeXT and 3rd party software vendors. For instance, I've tried indexing my entire home directory, ~, entire /NextLibrary/ and /LocalLibrary/, and /usr/spool/news/ (all on-line news articles) --very useful things to do-- but all have failed. Digital Librarian hangs indexing the stuff and crashes looking it up. Ideally, I'd like to index the root directory, ie, index everything, but it appears there's not chance-in-a-lifetime Digital Librarian could succeed at that! Root is much larger than Librarian is reliable, and has way too many non-standard files to trip Librarian up. Digital Librarian from NeXTstep 2.x was never able to index my home directory (about 40 MB) nor /usr/spool/news; inevitably one of Librarian's filters would hang on some character of one of the files, causing the whole indexing process to hang. And updating an index never seemed to work; Librarian either wouldn't make the updates it said it did, or it would permanently corrupt the index. Very disappointingly, Librarian from NeXTstep 3.0 is worse! Librarian of 3.0 not only hangs indexing, it hangs unpredictably and unrepeatably. At least Librarian of 2.x was consistent. Moreover, looking up anything from any of the indexes I have built (~, /LocalLibrary, and /NextLibrary) almost always crashes Librarian of 3.0 (depending on what you search for), or generates unpredictable, incomplete results (I've had Librarian of 3.0 give me a little longer hit list each time I reclicked on the Search button telling it to search again (for the same thing), as if I had to say ``Search!'' several times before it really took me seriously!) Worst of all, with NeXT new Technical ``Support'' wants *me to pay them* before they will explain to me why their software doesn't work work as advertized --due to *their* bugs in code or documentation, not my use of the product-- and how maybe I can work around it. I have just about given up on NeXT's Digital Librarian for doing most anything but looking up canned stuff pre-indexed by NeXT or 3rd parties. I'm just about ready to buy the competitive, non-standard product, Vertex, for no other reason than to get a bug-free product and decent technical support. But I'd keep with the NeXT indexing standard --it sounds from it's spec /NextLibrary/Documentation/NextDev/GeneralRef/07_Preliminary_IndexingKit/ to be pretty good, indexing general Objective-C objects in addition to files. So as a last ditch effort, has anyone encountered problems such as I have with with Digital Librarian? What did you do? -- Michael B. Parker P.O. Box 60951, Palo Alto, CA 94306-0951; voice 415-325-4451 (24 hrs), email MBParker@MIT.Edu (no NeXTmail yet), fax & modem 415-325-8604. -- Michael B. Parker P.O. Box 60951, Palo Alto, CA 94306-0951; voice 415-325-4451 (24 hrs), email MBParker@MIT.Edu (no NeXTmail yet), fax & modem 415-325-8604.
From: wjabi@libra.arch.umich.edu (Wassim M. Jabi) Newsgroups: comp.sys.next.bugs Subject: Diagram2! bug Date: 4 Mar 1993 20:41:53 GMT Organization: University of Michigan Engineering, Ann Arbor Distribution: usa Message-ID: <1n5pihINNni4@srvr1.engin.umich.edu> Bug Report: Here are steps to make Diagram!2 quit: 1) Start Diagram!2 with a new document. 2) Drag in something/anything from the pallette 3) Choose the Sound Inspector 4) Click on RECORD (say something if you want) 5) Click on PAUSE (things already look wrong in the inspector) 6) Click on the small sound icon near the shape in the main document. 7) Boom!! the program quits. -Wassim I got this reply from Lighthouse. > Wassim, > This is a bug that we had not come across yet. We thank you very much > for reporting it to us. We appreciate your time and effort in finding > this. We'll be working hard to correct this problem in a future > release. > Janet Hufnagel > Lighthouse Design I must say, I am still quite impressed with Diagram!2. For $25 dollars it is worth it. Keep up the good work, Lighthouse Design!!! -- Wassim M. Jabi (313) 936-0229 Doctoral Program in Architecture, University of Michigan 2000 Bonisteel Boulevard Ann Arbor Michigan 48105-2313 wjabi@libra.arch.umich.edu NeXTMail-friendly
From: jmcaulif@husc.harvard.edu (Jon McAuliffe) Newsgroups: comp.sys.next.bugs Subject: Anybody used loginwindow -LogoutHook? Keywords: next, loginwindow, logouthook, bugs Message-ID: <1993Mar4.172130.21325@husc3.harvard.edu> Date: 4 Mar 93 22:21:26 GMT Article-I.D.: husc3.1993Mar4.172130.21325 Distribution: world Hi, Has anybody successfully used the -LogoutHook option to loginwindow to execute a script on logout? I'd like to have my printer power off when I logout. I wrote loginwindow LogoutHook /usr/local/bin/logoutscript to the defaults database. The contents of that file are: #! /bin/csh /usr/etc/nppower off But it doesn't work. I log out, and the printer remains on. Ideas? (I don't think it's a permissions problem.)
Newsgroups: comp.sys.next.bugs From: abe@mace.cc.purdue.edu (Vic Abell) Subject: Re: Anybody used loginwindow -LogoutHook? Message-ID: <C3ExJo.KvD@mentor.cc.purdue.edu> Keywords: next, loginwindow, logouthook, bugs Sender: news@mentor.cc.purdue.edu (USENET News) Organization: Purdue University References: <1993Mar4.172130.21325@husc3.harvard.edu> Date: Fri, 5 Mar 1993 11:07:00 GMT In article <1993Mar4.172130.21325@husc3.harvard.edu> jmcaulif@husc.harvard.edu (Jon McAuliffe) writes: > >Hi, > > Has anybody successfully used the -LogoutHook option to >loginwindow to execute a script on logout? I'd like to have my printer >power off when I logout. I wrote > >loginwindow LogoutHook /usr/local/bin/logoutscript > >to the defaults database. The contents of that file are: I believe you have to use root's default database. After doing so you need to reboot or the equivalent.
From: ssircar@canon.com (Subrata Sircar) Newsgroups: comp.sys.next.bugs Subject: Mail.app problem Message-ID: <9303060308.AA29441@alychne-nc.canon.com> Date: 6 Mar 93 03:08:55 GMT Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Bug. Copy a piece of text from somewhere. Launch Mail.app. Paste (into Active.mbox). Boom. When you launch Mail.app again, there is a mail message (which it thinks has been read) from my username containing the piece of text I was attempting to paste, with the title "Appended Message". --- Subrata Sircar|ssircar@canon.com (NextMail ok)|Prophet & SPAMIT Charter Member Canon Information Systems and I do not share the same views on everything. "I'm just mad that I missed the sexual revolution." - me "Yes, but you dress much better as a result." - Mike
From: alex@cs.umd.edu (Alex Blakemore) Newsgroups: comp.sys.next.software,comp.sys.next.bugs Subject: Re: Digital Librarian hangs, crashes, & lies; does it work for anyone? Message-ID: <64854@mimsy.umd.edu> Date: 6 Mar 93 03:41:39 GMT References: <MBPARKER.93Mar4080820@netcom4.com> Sender: news@mimsy.umd.edu Followup-To: comp.sys.next.software Organization: U of Maryland, Dept. of Computer Science, Coll. Pk., MD 20742 In article <MBPARKER.93Mar4080820@netcom4.com> mbparker@mit.edu writes: > has anyone encountered problems such as I have with > with Digital Librarian? What did you do? man ixbuild it helps alot, especially to avoid trying to index files that dont make sense anyway like mach executables. -- --------------------------------------------------- Alex Blakemore alex@cs.umd.edu NeXT mail accepted
Newsgroups: comp.sys.next.bugs From: cedman@princeton.edu (Carl Edman) Subject: Re: Anybody used loginwindow -LogoutHook? In-Reply-To: abe@mace.cc.purdue.edu's message of Fri, 5 Mar 1993 11:07:00 GMT To: abe@mace.cc.purdue.edu (Vic Abell) Message-ID: <CEDMAN.93Mar5090409@capitalist.princeton.edu> Originator: news@nimaster Sender: news@Princeton.EDU (USENET News System) Organization: Princeton University References: <1993Mar4.172130.21325@husc3.harvard.edu> <C3ExJo.KvD@mentor.cc.purdue.edu> Date: Fri, 5 Mar 1993 13:04:09 GMT In article <C3ExJo.KvD@mentor.cc.purdue.edu> abe@mace.cc.purdue.edu (Vic Abell) writes: In article <1993Mar4.172130.21325@husc3.harvard.edu> jmcaulif@husc.harvard.edu (Jon McAuliffe) writes: > Has anybody successfully used the -LogoutHook option to >loginwindow to execute a script on logout? I'd like to have my printer >power off when I logout. I wrote > >loginwindow LogoutHook /usr/local/bin/logoutscript > >to the defaults database. The contents of that file are: I believe you have to use root's default database. After doing so you need to reboot or the equivalent. Just typing 'exit' at the loginwindow username prompt should be enough. Loginwindow should be back after a few seconds and with the new defaults. Carl Edman
From: gordie@cyclesoft.com (Gordie Freedman) Newsgroups: comp.sys.next.bugs Subject: Re: Mail.app problem Message-ID: <1993Mar6.192834.5992@netcom.com> Date: 6 Mar 93 19:28:34 GMT References: <9303060308.AA29441@alychne-nc.canon.com> Sender: gordie@netcom.com Organization: Cyclesoft Media Works In article <9303060308.AA29441@alychne-nc.canon.com> ssircar@canon.com (Subrata Sircar) writes: > > Bug. Copy a piece of text from somewhere. Launch Mail.app. Paste (into > Active.mbox). Boom. When you launch Mail.app again, there is a mail message > (which it thinks has been read) from my username containing the piece of text > I was attempting to paste, with the title "Appended Message". > > --- > Subrata Sircar|ssircar@canon.com (NextMail ok)|Prophet & SPAMIT Charter You don't have to launch Mail.app again, at least I didn't - it stuck it in immediately. In fact, I think they must have done this intentionally, I like it - a fast easy way to save something as a mail message to myself. Unless I'm misunderstanding you here for which I apologize, this doesn't look like a bug to me, maybe an undocumented "trick" ... NeXT people? -- >>> Gordie Freedman -> gordie@cyclesoft.com NeXTMail Yes! >>> Thou shalt not inline functions more complicated than 20
From: glenn@rightbrain.com (Glenn Reid) Newsgroups: comp.sys.next.misc,comp.sys.next.bugs Subject: Re: 'at' command not working under 3.0??? Message-ID: <1131@rtbrain.rightbrain.com> Date: 7 Mar 93 00:58:04 GMT References: <1993Mar5.074223.27029@netcom.com> Sender: glenn@rightbrain.com Followup-To: comp.sys.next.misc I've just been using 'at' a lot on my 2.1 system, and when I read your message, I tried it under 3.0. I can't get it to work, either. I think it's broken. -- Glenn Reid NeXTmail: glenn@rightbrain.com RightBrain Software 415-326-2974 (NeXTfax 326-2977) Palo Alto, California Electronic Frontier Foundation, member #054
Newsgroups: comp.sys.next.bugs Subject: IB Crashing Message-ID: <1993Mar7.122403.1538@yvax.byu.edu> From: feijai@endor.byu.edu (Sean Luke) Date: 7 Mar 93 12:24:02 -0700 Is it my imagination, or is IB crashing a great deal more nowadays than it did in 2.1? -- Sean Luke Brigham Young University MILK: It Comes From Cows sean@digaudio.byu.edu NeXTmail and nifty Mac stuff welcome
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: Mail.app problems ... Message-ID: <9303081948.AA29969@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Mon, 8 Mar 1993 19:48:12 GMT Gordie Freeman writes: >You don't have to launch Mail.app again, at least I didn't >- it stuck it in immediately. In fact, I think they must >have done this intentionally, I like it - a fast easy way >to save something as a mail message to myself. I just tried this again on two other 3.0 machines, and it does the same thing: namely, it *crashes* Mail.app. That sounds like a bug to me, n'est-ce-pas? :<) Judging from your post, it doesn't crash Mail.app for you. (One other person wrote me back to indicate it did crash Mail.app for him, but didn't paste in the message.) The bug report number is #39767. If you have further information, about this, you might want to mention it to NeXT using that number. - Subrata
From: anderson@sapir.cog.jhu.edu (Stephen Anderson) Newsgroups: comp.sys.next.bugs,comp.sys.next.misc Subject: Re: zsh bad executable message fixed (was:A brief comparative study of UN*X shells ) Message-ID: <ANDERSON.93Mar9144243@sapir.cog.jhu.edu> Date: 9 Mar 93 19:42:43 GMT References: <ANDERSON.93Mar1070245@sapir.cog.jhu.edu> <1n204u$9vq@wwille.hanse.de> Sender: news@umd5.umd.edu Organization: Dept. of Cognitive Science, The Johns Hopkins University >>>>> On 3 Mar 93 10:09:34 GMT, wwille (To: zsh-list@cs.uow.edu.au) said: [much deleted...] ww> to Carl Edman for that. Remains the runtime enviroment, i ww> checked the changes i made. Failure is here, on my machine its ww> in the /etc/zshrc ww> -- snip --- ww> # use hard limits, except for a smaller stack and no core dumps ww> unlimit ww> limit stack 8192 ww> limit core 0 ww> #limit -s ww> -- snap -- the above works ww> I had outcommented the "limit stack 8192" which causes an ww> unlimited stacksize, with which zsh on an NeXT can't deal. Now ww> it works all fine. Just to confirm that after I put a limit on stacksize into my zsh init files, the "bad executable" problem has disappeared. So it looks like that's the answer. Thanks for the help, --Steve Anderson
From: goldly@u.washington.edu (Lloyd P. Goldwasser) Newsgroups: comp.sys.next.bugs Subject: Re: Keyboard.app Bug Date: 10 Mar 1993 00:55:04 GMT Organization: University of Washington Message-ID: <1nje98INNak3@shelley.u.washington.edu> References: <1993Mar3.090120.5007@cs.ucla.edu> In article <1993Mar3.090120.5007@cs.ucla.edu> andylee@oahu.cs.ucla.edu (Andy Lee) writes: > I think I've found a bug in the 3.0 Developer demo program > Keyboard.app. I used it a few months ago to remap my keyboard to the > Dvorak layout, now I've gotten used to the new layout, this happens: > the letter 'z' (remapped to the "?/" key) is still lower case when I > turn on the CapLock via Command-Shift. The same goes for all the > letters that remapped to symbol keys, i.e. 'z' at the "?/" key, 's' at > ":;", 'w' at "<,", and 'v' at ">.". > > Does anyone have a solution to this? I had that problem for a while, and then found that it solved itself when I simply repeated what the process, in Keyboard.app, of pulling down the desired caps onto their keys. Go figure. Lloyd Goldwasser goldwalp@zoology.washington.edu
Newsgroups: comp.sys.next.bugs From: death@kira.net.netcom.com (David John Burrowes) Subject: Re: Mail.app problem Message-ID: <1993Mar9.025433.645@kira.net.netcom.com> Sender: death@kira.net.netcom.com Organization: No organization at this time. References: <1993Mar6.192834.5992@netcom.com> Date: Tue, 9 Mar 1993 02:54:33 GMT In article <1993Mar6.192834.5992@netcom.com> gordie@cyclesoft.com (Gordie Freedman) writes: > In article <9303060308.AA29441@alychne-nc.canon.com> ssircar@canon.com (Subrata > Sircar) writes: > > > > Bug. Copy a piece of text from somewhere. Launch Mail.app. Paste > > Active.mbox). Boom. When you launch Mail.app again, there is a mail > > (which it thinks has been read) from my username containing the piece > > I was attempting to paste, with the title "Appended Message". > > You don't have to launch Mail.app again, at least I didn't - it stuck it > immediately. In fact, I think they must have done this intentionally, I > - a fast easy way to save something as a mail message to myself. > > Unless I'm misunderstanding you here for which I apologize, this doesn't > like a bug to me, maybe an undocumented "trick" ... NeXT people? The issue is (if I understand the above) that if you have copied text, and paste it into the .mbox window, rather than in a send window, it gets pasted into the mbox as if you were adding a new message. If there are any stuff before a 'From ' line, it stores those as an 'appended' message, and then stores any following messages as normal mail messages with subjects etc. I find this invaluable when, say, wanting to import some email from another system into a mbox... just copy it and paste it into the .mbox window. The Appended Message stuff at least warns one that one didn't paste a full message. I thought I had read of this feature (and the related one of being able to copy multiple messages out of a mail box by selecting them and choosing copy)? In any case, I don't think it's a bug (unless you, legitimately, define a bug as a thing that's not documented, and this isn't a documented thing...) \david john
Newsgroups: comp.sys.next.bugs From: mmoss@ic.sunysb.edu (Matthew D Moss) Subject: Can't Open DSP Error!!! Message-ID: <C3n3uI.AHp@max.physics.sunysb.edu> Sender: news@max.physics.sunysb.edu (News Administration) Organization: State University of New York at Stony Brook Distribution: usa Date: Tue, 9 Mar 1993 21:03:54 GMT I was using the playscore command to run some *.score files as demos. I created a Terminal Service so that any *.score file could be selected, then the service chosen from the Services menu and played in the background. Soon later, the songs would no longer play. On typing the command directly into a terminal shell, I got the error message "Can't open DSP." Help?!?! -- +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Matthew David Moss | Blessed are the pure in heart, for they | | INTERNET: mmoss@ic.sunysb.edu | will see God. | | BITNET : mmoss@sbccmail | Matthew 5:8 |
Newsgroups: comp.sys.next.bugs From: d89cb@efd.lth.se (Christian Brunschen) Subject: Terminal, Kermit(188), Telnet, Rlogin (or something) in 3.0 Message-ID: <1993Mar10.134849.3085@lth.se> Sender: news@lth.se Organization: Lund Institute of Technology, Sweden Date: Wed, 10 Mar 1993 13:48:49 GMT When I run kermit in a Terminal window, the msb of every character that is displayed on the screen appears to be stripped. Not by the modem, either -- I ca do binary file transfers just fine. And the character remapping feature in Kermit (from ISO-8859-1 to NeXTSTEPencoding) also works, because the 7bit characters I get are the corresponding NeXTSTEPEncoding characters with the msb turned off. If I tell kermit to convert a local file from latin-1 to Nextstepencoding, everything is fine, and the characters are as 8-bit as they should be (and they're correctly remapped). If I try to cat an 8bit file, that works, too -- but less(1):ing or more(1):ing strips the msb. Telnet and rlogin -- even rlogin -8, which is supposed to be 8bit-clean -- do the same thing to me, ie, strip the msb .. and that's when connecting to localhost ... it feels a little sad, knowing that all those cute little 8-bit characters are there, and I just can't see them. (Being in Sweden, I'd very much like to use the national characters we use here, thank you.) anyone care to comment ? is this a bug ? where is the bug ? oh, the "kermit" I'm using is the latest one available, 188 with some extra patches for the NeXT (if I recall correctly) best regards Christian Brunschen -- /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/ // Christian Brunschen // Husmansv. 26 // S - 227 38 Lund // Sweden // // voice# +46 (0)46 139345 // internet d89cb@efd.lth.se // // // irc snooker Senster gcb // ------------------------------------++
Newsgroups: comp.sys.next.bugs From: wisdom@geom.umn.edu (Scott Wisdom) Subject: Re: Can't Open DSP Error!!! Message-ID: <C3nKBE.7D0@news2.cis.umn.edu> Sender: news@news2.cis.umn.edu (Usenet News Administration) Organization: Geometry Center, University of Minnesota References: <C3n3uI.AHp@max.physics.sunysb.edu> Distribution: usa Date: Wed, 10 Mar 1993 02:59:31 GMT In article <C3n3uI.AHp@max.physics.sunysb.edu> mmoss@ic.sunysb.edu (Matthew D Moss) writes: >Soon later, the songs would no longer play. On typing the command directly >into a terminal shell, I got the error message "Can't open DSP." I too have gotten this. Rebooting won't change a thing. I can still run playscore from root, though. Anyone know what this is? SW wisdom@geom.umn.edu
From: jmcaulif@husc.harvard.edu (Jon McAuliffe) Newsgroups: comp.sys.next.bugs Subject: Re: Mail.app problems ... Keywords: mail next problem bug Message-ID: <1993Mar10.133543.21499@husc3.harvard.edu> Date: 10 Mar 93 18:35:38 GMT Article-I.D.: husc3.1993Mar10.133543.21499 References: <9303081948.AA29969@alychne-nc.canon.com> Distribution: world In article <9303081948.AA29969@alychne-nc.canon.com> writes: > > Gordie Freeman writes: > >You don't have to launch Mail.app again, at least I didn't > >- it stuck it in immediately. In fact, I think they must > >have done this intentionally, I like it - a fast easy way > >to save something as a mail message to myself. > > I just tried this again on two other 3.0 machines, and it does the same > thing: > namely, it *crashes* Mail.app. That sounds like a bug to me, n'est-ce-pas? > > Judging from your post, it doesn't crash Mail.app for you. (One other person > wrote me back to indicate it did crash Mail.app for him, but didn't paste in > the message.) The bug report number is #39767. If you have further > information, about this, you might want to mention it to NeXT using that > number. > > - Subrata I think we may be talking about slightly different things. Here's what I did, under 3.0: 1) Quit Mail.app (because it was running before I read this bug post). 2) Select the original bug report paragraph from the posting in NewsGrazer; hit Cmd-C. 3) Launch Mail.app from the Dock. 4) With Active.mbox selected, hit Cmd-V. What happened: A message appeared in the message list called "appended message". Selecting it, I saw "From: Jon", "Subject: Appended Message", followed by exactly the text I had copied. No bug here, as far as I can tell. Have I misinterpreted what the bug is supposed to be? I would call this a feature, because it is easy to take any text and get it into a user mailbox. _________________________________________________________________________ Jon McAuliffe jmcaulif@husc.harvard.edu NeXT ex-Campus Consultant 617.493.3004 (voice/fax) Harvard University
From: jmcaulif@husc.harvard.edu (Jon McAuliffe) Newsgroups: comp.sys.next.bugs Subject: Thanks for the -LogoutHook Help! Keywords: next loginwindow logouthook Message-ID: <1993Mar10.133951.21500@husc3.harvard.edu> Date: 10 Mar 93 18:39:50 GMT Article-I.D.: husc3.1993Mar10.133951.21500 Distribution: world Hi, Thanks to all the people who mailed and posted replies to my problems using the -LogoutHook option to loginwindow. I actually did do the dwrite as root originally; I simply neglected to mention that. The REAL problem was that I wrote the script in my home directory and only set the execute permission for my username, then copied it into /usr/local/bin and expected root to be able to execute it without x set. <Ahem>. chmod a+x fixed all that. Now I'll move on to getting -Log IN Hook to work! _________________________________________________________________________ Jon McAuliffe jmcaulif@husc.harvard.edu NeXT ex-Campus Consultant 617.493.3004 (voice/fax) Harvard University
Newsgroups: comp.sys.next.bugs From: infoman!root (Operator) Subject: Addresses bug Message-ID: <1993Mar10.145809.4955@infoman.com> Sender: root@infoman.com (Operator) Organization: Information Management Inc. Distribution: usa Date: Wed, 10 Mar 1993 14:58:09 GMT I have Customers.addresses file which resides in my LocalLibrary/Addresses directory which is distributed across the network. The fax panel finds and displays these entries for all users. All users on the network need to fax to the same numbers, and the numbers need to be updated in one place since we have so many customers. What seems to happen is that records entered by one user are not visible to other users for a period of time even in the .addresses file itself and sometimes they disappear all together. Has anyone experienced similar problems. Any help would be much appreciated Thanks for your help in advance. - willi please reply via e-mail willi@infoman.com
From: jmcaulif@husc.harvard.edu (Jon McAuliffe) Newsgroups: comp.sys.next.bugs Subject: Workspace crashes related to BackSpace? Keywords: workspace backspace crash bug Message-ID: <1993Mar10.135045.21501@husc3.harvard.edu> Date: 10 Mar 93 18:50:42 GMT Article-I.D.: husc3.1993Mar10.135045.21501 Distribution: world Hi, Has anyone experienced a problem using Backspace where the workspace mysteriously dies after about two hours? I often leave BackSpace on when I go to classes with the screenlock on so I don't have to log in again when I come back. Unfortuantely, if it stays on for more than about two hours, I come back to my room greeted by loginwindow and find the following in /usr/adm/messages: Mar 7 21:26:55 goethe Terminal[344]: DPS client library error: Error while writing to connection, DPSContext 46278, data -102 Mar 7 21:26:55 goethe Edit[234]: DPS client library error: Error while writing to connection, DPSContext 7c1f8, data -102 Mar 7 21:26:55 goethe Edit[234]: Exiting due to Window Server death Mar 7 21:26:55 goethe Terminal[344]: Exiting due to Window Server death Mar 7 21:26:55 goethe Mail[232]: DPS client library error: Error while writing to connection, DPSContext 7a224, data -102 Mar 7 21:26:55 goethe Mail[232]: Exiting due to Window Server death Mar 7 21:26:55 goethe Preferences[231]: DPS client library error: Error while writing to connection, DPSContext 52294, data -102 Mar 7 21:26:55 goethe Preferences[231]: Exiting due to Window Server death Mar 7 21:26:56 goethe WM[230]: DPS client library error: Error while writing to connection, DPSContext c425c, data -102 Mar 7 21:26:55 goethe BackSpace[236]: DPS client library error: Error while writing to connection, DPSContext 15e1e4, data -102 Mar 7 21:26:56 goethe BackSpace[236]: Exiting due to Window Server death Mar 7 21:27:25 goethe -[149]: loginwindow: Workspace exited ts 0 cd 0 rc 0 sv 0 ss 0. Mar 7 21:27:26 goethe -[149]: DPS error dps_err_write Mar 7 21:27:26 goethe last message repeated 20 times Mar 7 21:27:26 goethe -[149]: exiting due to excessive DPS errors Mar 7 21:27:33 goethe mach: audio kernel server unloaded Mar 7 21:27:34 goethe mach: audio kernel server initialized Mar 7 21:29:42 goethe syslog: Workspace logged in I remember reading somewhere in the BackSpace sources that some timer should ideally be a long long instead of just a long, but I sure don't remember anything about crashing the Workspace. I have a lot of backspace modules, if that matters. By the way, the first time my Workspace EVER crashed was because of this, and believe me I give it a workout. _________________________________________________________________________ Jon McAuliffe jmcaulif@husc.harvard.edu NeXT ex-Campus Consultant 617.493.3004 (voice/fax) Harvard University
Newsgroups: comp.sys.next.bugs From: bobcook@slac.stanford.edu (Bob Cook) Subject: Re: Mail.app problems ... Message-ID: <C3pG1q.Mnr@unixhub.SLAC.Stanford.EDU> Sender: news@unixhub.SLAC.Stanford.EDU Organization: Stanford Linear Accelerator Center References: <1993Mar10.133543.21499@husc3.harvard.edu> Date: Thu, 11 Mar 1993 03:22:37 GMT Jon McAuliffe writes > > I think we may be talking about slightly different things. Here's what >I did, under 3.0: > >1) Quit Mail.app (because it was running before I read this bug post). >2) Select the original bug report paragraph from the posting in NewsGrazer; > hit Cmd-C. >3) Launch Mail.app from the Dock. >4) With Active.mbox selected, hit Cmd-V. > >What happened: >A message appeared in the message list called "appended message". Selecting >it, I saw "From: Jon", "Subject: Appended Message", followed by exactly the >text I had copied. No bug here, as far as I can tell. Have I misinterpreted >what the bug is supposed to be? I would call this a feature, because it is >easy to take any text and get it into a user mailbox. I did exactly your four steps above, and Mail crashed. When I brought it back up, the new item was there as you describe. -- Bob Cook bobcook@slac.stanford.edu Stanford Linear Accelerator Center NeXT mail okay
From: nevai@mps.ohio-state.edu (Paul Nevai) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.software,comp.sys.next.bugs Subject: BUG in either "man" or "mf.1" because "man mf" doesn't work Date: 11 Mar 1993 13:10:52 GMT Organization: Department of Mathematics, The Ohio State University Sender: Paul Nevai Distribution: world Message-ID: <1nndos$e7i@zaphod.mps.ohio-state.edu> Originator: nevai@ops.mps.ohio-state.edu Attention mf users on NeXT: Do a "man mf" job. Your mf.1 will not be processed because "man" can't finish the job. This is what happens: > ops> man mf > Reformatting page. Wait... > /usr/man/man1/mf.1: line 124: no specification > tbl quits > done What the heck is going on? The same mf.1 works fine on our SUN so I guess the NeXT has the problem. Any ideas? Take care...Paul Paul Nevai nevai@mps.ohio-state.edu Dept Math - Ohio State University 1-614-292-3317 (Office) Columbus, Ohio 43210-1174, U.S.A. 1-614-292-1479 (Math Dept Fax)
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: Workspace Console hanging ... Message-ID: <9303111825.AA02915@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Thu, 11 Mar 1993 18:25:40 GMT You wrote: >The Workspace console window hangs if you output a very long line to it. Could you elaborate? The following can be repeatedly executed on my mono NeXTStation running 3.0: ***** starblazer> echo "fja;fjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj jjjjjjjjjjjjjjjjjjjjjjddddddddddddddddddddddddaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaeeee eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee eeeeeeeeeeeeeeeeedddddddadsvbadfaerqwerwqerqweafgafgafgsdghdhjfhljghifyuetrdfgsf hvgjcbnmvbjdhsfsgfasdfgsadrerhretsdfgsdhfghdfgfghnfgsdfgadfgsdfgsthesrghsdfhgsdf gwqsdghfghdghdjhjfgjhfgjhfgjfgkjkghkghksdfgwetwertwergsgsgtsegsdfgasfgsfgh" > /dev/console starblazer> echo "fja;fjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj jjjjjjjjjjjjjjjjjjjjjjddddddddddddddddddddddddaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaeeee eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee eeeeeeeeeeeeeeeeedddddddadsvbadfaerqwerwqerqweafgafgafgsdghdhjfhljghifyuetrdfgsf hvgjcbnmvbjdhsfsgfasdfgsadrerhretsdfgsdhfghdfgfghnfgsdfgadfgsdfgsthesrghsdfhgsdf gwqsdghfghdghdjhjfgjhfgjhfgjfgkjkghkghksdfgwetwertwergsgsgtsegsdfgasfgsfgh" > /dev/console starblazer | /home/alychne/eng/ssircar> ***** Do the lines need to be longer, or is there something else going on that I didn't take into account? --- Subrata Sircar|ssircar@canon.com (NextMail ok)|Prophet & SPAMIT Charter Member Canon Information Systems and I do not share the same views on everything. "I'm just mad that I missed the sexual revolution." - me "Yes, but you dress much better as a result." - Mike
Newsgroups: comp.sys.next.bugs From: marcel@nice.usergroup.ethz.ch (Marcel Waldvogel) Subject: telnetd/rlogind Message-ID: <1993Mar12.022914.14821@bernina.ethz.ch> Sender: news@bernina.ethz.ch (USENET News System) Organization: NiCE - NeXT User Group, Zuerich Date: Fri, 12 Mar 1993 02:29:14 GMT Does anyone know whether the old and ugly bugs in telnetd (larger outputs often miss some carriage returns) and rlogind (login shell is not connected to its tty) will be fixed anytime soon? It would also be nice if there were an option to disable telnet's automatic sending of the USER environment variable. -Marcel
Newsgroups: comp.sys.next.bugs,comp.sys.next.misc From: Zacharias J. Beckman <zac@dolphin.com> Subject: Cross posts in NewsGrazer--Do they work? Message-ID: <1993Mar11.214325.2152@dolphin.com> Sender: zac@dolphin.com Organization: Dolphin Software Distribution: usa Date: Thu, 11 Mar 1993 21:43:25 GMT Whenever I try to crosspost an article using NewsGrazer the article does not show up at our local site. For instance, this article is being crossposted and I've no idea if it will go out on the net, becuase I won't get to see it locally. Does anyone have a fix for the problem? This message is also serving as a bit of a test... if I receive any replies, I'll at least know that the article goes out! Thanks in advance. -- Zacharias J. Beckman - Dolphin Software Inc. - zac@dolphin.com - use NeXTMAIL! To be "matter of fact" about the world is to blunder into fantasy.... and dull fantasy at that, as the real world is strange and wonderful. --- R. A. Heinlen Those opinions I express herein are my own, I'm fairly sure. --- Z. J. Beckman
From: zac@dolphin.com (Zacharias J. Beckman) Newsgroups: comp.sys.next.bugs Subject: Can't Open DSP??? Help! Message-ID: <1993Mar11.213925.1980@dolphin.com> Date: 11 Mar 93 21:39:25 GMT Sender: zac@dolphin.com Distribution: usa Organization: Dolphin Software Has anyone an explanation why, fairly recently, our server lost its ability to use 'playscore'? This is what happens when I try: /Users/zac/Apps/hours% playscore ./chimes/five playscore reading ./chimes/five... ...done Can't open DSP. /Users/zac/Apps/hours% ll /dev/dsp crw-rw-rw- 1 root wheel 3, 3 Jul 30 1992 /dev/dsp /Users/zac/Apps/hours% On another machine it plays the score, rather than saying "Can't open DSP." We haven't done anything recently that made a large-scale change to the machine (such as upgrading to 3.0; it already is a 3.0 machine). Any help would be appreciated! -- Zacharias J. Beckman - Dolphin Software Inc. - zac@dolphin.com - use NeXTMAIL! To be "matter of fact" about the world is to blunder into fantasy.... and dull fantasy at that, as the real world is strange and wonderful. --- R. A. Heinlen Those opinions I express herein are my own, I'm fairly sure. --- Z. J. Beckman
Newsgroups: comp.sys.next.bugs From: vramaswa@newstand.syr.edu (Vijayalakshmi Ramaswami) Subject: Openwindows - calentool Organization: Academic Computing Services, Syracuse University, Syracuse, NY. Message-ID: <1993Mar11.180712.22579@newstand.syr.edu> Keywords: calentool Date: Thu, 11 Mar 93 18:07:12 EST We have calentool version 2.1 for our openwindows. We have the 'calentool' on our menus. But whenever we click the menu there is the following error seen on the console window /bin/sh: 6645 Memory fault. Could someone please help us? Please e-mail to vramaswa@rodan.syr.edu Thanks, Academic Computing Services, Syracuse University.
From: eilts@late.e-technik.uni-erlangen.de (Hinrich Eilts) Newsgroups: comp.sys.next.bugs Subject: Re: telnetd/rlogind Date: Fri, 12 Mar 1993 12:53:53 GMT Organization: LATE, Uni-Erlangen, Germany Message-ID: <1nq151E930@uni-erlangen.de> References: <1993Mar12.022914.14821@bernina.ethz.ch> marcel@nice.usergroup.ethz.ch (Marcel Waldvogel) writes: >Does anyone know whether the old and ugly bugs in telnetd (larger >outputs often miss some carriage returns) and rlogind (login shell is >not connected to its tty) will be fixed anytime soon? The rlogind-bug is still here: Do a rlogin on your own and start kermit: Error opening /dev/tty -- Bye | G i b D O S | Hinrich Eilts | k e i n e | (e-mail: eilts@late.e-technik.uni-erlangen.de) | C h a n c e ! |
From: wade@hobbes.ucsd.edu (Wade Blomgren) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: transcript and text files Followup-To: comp.sys.next.sysadmin Date: 12 Mar 1993 20:38:33 GMT Organization: University of California, San Diego Message-ID: <1nqsc9INNol0@network.ucsd.edu> Since upgrading to 3.0, lpr will no longer "automagically" print text files to a Postscript printer. More specifically, the transcript print filters no longer detect that a file is "not already postscript" and enscript it. There are references to the "pstext" filter in the psint.sh script (the script that the transcript filters are linked to), that is to say the environment variable specifying the path for pstext is set and exported, but the pstext program itself does not exist. I am certain that lpr of a plain text file worked prior to the upgrade. Has anyone else run into this? Does /usr/lib/transcript/pstext exist in 2.x? I have access to transcript 2.0 sources (the NeXT appears to use 3.0) and I thought I might try compiling and installing pstext from transcript 2.0 but I doubt that will fix things since running strings on the NeXT's pscomm shows no reference to the pstext variable. Any ideas would be appreciated. Wade Blomgren wade@hobbes.ucsd.edu
From: pjoe@cubenet.sub.org (Peter Eybert) Newsgroups: comp.sys.next.bugs Subject: Renderman Bug ? Message-ID: <8871@salyko.cubenet.sub.org> Date: 12 Mar 93 17:40:02 GMT Sender: root@cubenet.sub.org Distribution: comp A question concerning the photoreal Renderman: Has anybody managed to write a transformation shader that works ? If you check a transformation shader with [aShader shaderType] you get UNKNOWN as a result and the shader does nothing. Also, if you call such a shader via RiDeformation (although I think that this shouldn't be necessary), there are no results. As I've tested them, all the other shader types work properly. Any suggestions are welcome. Peter. -- Peter Eybert pjoe@cubenet.sub.org Appenzeller Str. 123 voice +49-89-7593734 8000 Muenchen 71
From: heading@signal.dra.hmg.gb (Anthony J.R. Heading) Newsgroups: comp.sys.next.bugs Subject: Problem with Ohlfs10 screenfont Message-ID: <1nnf14INNkkn@milne.dra.hmg.gb> Date: 11 Mar 93 13:32:20 GMT Organization: Defence Research Agency There's something weird with Ohlfs 10pt, and maybe other fonts, whereby the pitch of some characters is variable. If you have this posting in a terminal window, and then bring up the font panel and switch between 10 and 11 pt, the exclamation marks all bunch up. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ...................................... This is not terribly useful for editors. Interestingly, 10.01 pt looks very similar to 10pt but solves the problem, presumably it isn't invoking the screen font. The line spacing changes completely though, for no good reason. As implemented, screen fonts seem to be very unpleasant things what are they for anyway? I might just delete mine. Anthony
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs,comp.sys.next.software From: uunet!sci34hub!tybrin4!holli!me (Trey McClendon) Subject: Mathematica installation gotcha Message-ID: <1993Mar13.013827.1704@holli.uucp> Sender: me@holli.uucp (My Account) Organization: TYBRIN Corporation -- Huntsville, AL Date: Sat, 13 Mar 1993 01:38:27 GMT Hi, Just wanted to alert everyone to a potential problem with Mathematica that I ran into recently. A pre-2.1 version of Mathematica was installed on a file server. mathremote, math, and mcc were put into /usr/local/bin. The problem manifested itself in that math from the command line worked fine, but the Notebook front end would not work. A simple operation like '1+1' put the 'running...' up in the title bar and no answer ever came out. After talking with Wolfram, who really tried hard to help, over the course of several phone calls, they sent me v2.1. Same thing happened. Well I got busy and dropped it until today. Another guy was installing a copy on another machine and had the same problem. Well, the problem ended up being the shell script 'mathremote', which the front-end uses to talk to the kernel, I think. There is a long command including many pipes that just happens to use a command called 'control' -- without a pathname. We had a command called control that is a script used for some other purpose that was being used instead of the one nestled within the heirarchy of the Mathematica.app. After modifying the script to put the absolute pathname in front of control, everything worked fine. Trey McClendon tybrin4!trey@sci34hub.sci.com -- Trey McClendon, Madison, AL
Newsgroups: comp.sys.next.bugs From: monty@its.com (Montgomery Zukowski) Subject: distributed objects trap and fix Message-ID: <9303151749.AA01201@its.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Mon, 15 Mar 1993 17:48:19 GMT Please send any responses to monty@its.com since I have no news feed. I have an app which monitors activity on another process via distributed objects. Since the other process is independent, my app has a timed entry function which tries to connect to the process every 0.5 seconds. The problem I ran into is if the process breaks its connection, the next time the loop comes around I hang in [NXConnection connectToName:"other_process"]; if it is called too soon (<0.5 sec). Strangely enough this has only happened when the process was launched before the app. If it is launched afterwards I have not had it hang, but I have no idea why this could affect things. Everything I described happened every time I tried, so things seemed to be consistent. In my timed event function I originally called a method named connect: to try to reconnect. Now I use the perform:with:afterDelay:cancelPrevious method with a delay of 5 seconds to call connect:, and everything works. I believe the problem has something to do with trying to reconnect too soon after recieving invalidation notification. Perhaps the processes connection is broken but the net-name is not yet unregistered. Here is some sample code: -appDidInit:sender { .... DPSAddTimedEntry(0.5,timeFun,self, NX_BASETHRESHOLD); .... } - connect:sender { if (server) return self; //next line will hang if called too soon after invalidation notification server=[NXConnection connectToName:"other_process"]; if (server) { [[server connectionForProxy] registerForInvalidationNotification:self]; } return self; } - senderIsInvalid:sender { server=nil; return self; } - server { return server; } void timeFun(DPSTimedEntry timedEntry, double now, void *userData ) //userData is really self { if ([(id) userData server]) { [[(id) userData statusField] setStringValue:"Connected!"]; } else { [[(id) userData statusField] setStringValue:"Unconnected"]; //originally [(id) userData connect:] which caused a hang [(id) userData perform:@selector(connect:) with:userData afterDelay:5000 cancelPrevious:NO]; } return; } THaNKS, MoNTY PS. Please send any responses to monty@its.com since I have no news feed.
Newsgroups: comp.sys.next.bugs From: gemoe@proximus.north.de (Gerhard Moeller) Subject: Re: Mail.app problems ... References: <C3pG1q.Mnr@unixhub.SLAC.Stanford.EDU> Sender: gemoe@proximus.north.de Organization: Gerhard Moeller, German NeXT User Group, Oldenburg. Date: Tue, 16 Mar 1993 14:12:00 GMT Message-ID: <1993Mar16.141200.1244@proximus.north.de> In article <C3pG1q.Mnr@unixhub.SLAC.Stanford.EDU> bobcook@slac.stanford.edu (Bob Cook) writes: > I did exactly your four steps above, and Mail crashed. When I brought it back up, the new item was there as you describe. So there are different versions. Mine didn't crash. The message was there. Am I lucky? :-) -- +---------------------------< principiis obsta! >---------------------------+ N Gerhard Moeller, Teichstrasse 12, 2900 Oldenburg (FRG) [*: 02/21/1968] N e Private: gemoe@proximus.north.de Phone (voice): +49-441-75520 e X Uni: Gerhard.Moeller@arbi.Informatik.Uni-Oldenburg.DE NeXTmail X T Z-Net: Gerhard.Moeller@uniol.zer encouraged! T +-> NoGeNUG - Northern German NeXT User Group: NoGeNUG@proximus.north.DE <-+
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: Mail Reply-To: bug ... Message-ID: <9303170050.AA06442@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Wed, 17 Mar 1993 00:50:48 GMT You wrote: >In Mail.app, under NS3.0, changing the Reply-To: field and >clicking on Set does not modify the field on the next >outgoing piece of mail. However, if another setting in the >Preferences dialog _is_ changed along with the Reply-To >field (e.g. checking a check box), then the change to the >Reply-To field takes effect. I did the following: 1. Went to preferences and changed the Reply-To: field, clicking on Set afterwards. (I changed it to joseph@technicolor.dreamcoat.) 2. Sent a mail message with no body to myself. 3. Read the message - the Reply-To: field did specify joseph. Am I missing something? --- Subrata Sircar|ssircar@canon.com (NextMail ok)|Prophet & SPAMIT Charter Member Canon Information Systems and I do not share the same views on everything. "I'm just mad that I missed the sexual revolution." - me "Yes, but you dress much better as a result." - Mike
Newsgroups: comp.sys.next.bugs From: seung@guardian.cs.psu.edu (Gueesang Lee) Subject: Problems in login panel etc. Message-ID: <C40JMo.Joy@cs.psu.edu> Sender: news@cs.psu.edu (Usenet) Organization: Penn State Computer Science Distribution: usa Date: Wed, 17 Mar 1993 03:13:35 GMT I am using NeXT station with NeXT 2.0. I hope someone can help me with the following problems. 1. Login problem. When I try to login, it accepts the password and then within a second, it returns to the login panel right away. This happens to all normal accounts except "root." With "root", everything seems to be working normally. Does anybody have an idea what's wrong with this? 2. I have to use ROM monitor, after once I used it, to log in. And I have to type "p" to set the booting device every time I login. Is there anyway to have the login panel automatically? 3. I am trying to use a XT PC as a terminal by a null modem cable. The cable is ready, but I don't know what should be done in NeXT, for example setting device files etc. I would appreciate any help from you.
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: SetParaStyle bug ... Message-ID: <9303170449.AA06698@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Wed, 17 Mar 1993 04:49:48 GMT I could not reproduce your bug with the following code: ***** #import <appkit/appkit.h> @interface Delegate:Object { Window *window; } - openMainWindow; @end @implementation Delegate - openMainWindow { NXRect contentRect = {{0.0f, 0.0f}, {200.0f, 200.0f}}; window = [[Window alloc] initContent:&contentRect style:NX_TITLEDSTYLE backing:NX_BUFFERED buttonMask:(NX_CLOSEBUTTONMASK | NX_MINIATURIZEBUTTONMASK) defer :YES]; [window setFreeWhenClosed:YES]; [window center]; [window makeKeyAndOrderFront:self]; return self; } - appDidInit:(const id)sender /* app delegate */ { Text *text; static NXTabStop ts = {NX_LEFTTAB,5.0}; static NXTextStyle style = {0.0,0.0,0.0,3.0,0,1,&ts}; int timeout = 30000; NXRect contentRect = {{0.0f, 0.0f}, {200.0f, 200.0f}}; [self openMainWindow]; text = [window getFieldEditor:YES for:window]; // doing some stuff to show text's presence: [[window contentView] addSubview:text]; [text setFrame:&contentRect]; [text setText:"Hello, world."]; [text setParaStyle:(void *)&style]; // [text setParaStyle:[text defaultParaStyle]]; [NXApp perform:@selector(terminate:) with :self afterDelay:timeout cancelPrevious:NO]; return self; } @end /* * This is correct, but gcc will give a warning because methods can't be * qualified. */ volatile int main(void) { Delegate *const delegate = [[Delegate alloc] init]; NXApp = [Application new]; [NXApp setDelegate:delegate]; [NXApp run]; /* NOTREACHED */ } ***** I also looked at it in the debugger and manually set the paraStyle to various styles, including the default style; none of them crashed. Could you elaborate or correct this example? --- Subrata Sircar|ssircar@canon.com (NextMail ok)|Prophet & SPAMIT Charter Member Canon Information Systems and I do not share the same views on everything. "I'm just mad that I missed the sexual revolution." - me "Yes, but you dress much better as a result." - Mike
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: Mail "appended message" bug #39767 - Followup Message-ID: <9303170529.AA06730@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Wed, 17 Mar 1993 05:29:41 GMT Thanks to those who sent me information about their systems. I can tell some of what it is not, but I was unable to narrow the bug down and am open to suggestions. Originally, I reported the following to NeXT as bug #39767: 1. Select ascii or RTF text from somewhere; Mail will just beep at other pasteboard types. 2. Make the upper portion of a *.mbox window's NXSplitView the first responder. This can be done by quitting mail and restarting, clicking on the title bar of a *.mbox window, etc. 3. Paste. Boom - for some of us Mail will crash. 4. Restart Mail if it crashed. For some of us there will be a message in the window, from the user who attempted to paste, titled "appended message", marked unread, containing the text which would have been pasted. For several people, Mail doesn't crash. For others, Mail crashes but doesn't paste in the message. Whether or not the message is pasted doesn't appear to be hardware dependant; the TurboColor (ADB) doesn't get a message pasted with one user logged in and does when I log in. The crash doesn't appear to be hardware or network dependent; if I log in as root and execute the above procedure on my machine, Mail doesn't crash and does paste the message. Neither does it appear to be preference or default-related, since if I do starblazer> dread -o Mail |dremove and run Mail again, executing the above, it still crashes. I don't think it is preference-related unless it is the preferences on some system file: I do own all of the pieces of the mailbox I'm attempting to paste to. Even more strange, gdb doesn't like the resulting core file: **** starblazer> gdb /NextApps/Mail.app/Mail Reading symbol data from /NextApps/Mail.app/Mail... (no debugging symbols found)...done. Reading symbol data from /usr/shlib/libIndexing_s.A.shlib...done. Reading symbol data from /usr/shlib/libMedia_s.A.shlib...done. Reading symbol data from /usr/shlib/libNeXT_s.C.shlib...done. Reading symbol data from /usr/shlib/libsys_s.B.shlib...done. (gdb) core-file /cores/core.865 0x22042 in setField () (gdb) bt #0 0x22042 in setField () Cannot access memory: address 0xa0a0018 out of bounds. (gdb) quit **** Every time I generated a core file, it crashed in the same place and didn't allow any backtraces of the stack. [Can anyone who does get a crash verify this? You can get Mail to dump core by doing "dwrite Workspace CoreLimit XXXXXXX" and then logging out and logging back in.] --- Subrata Sircar|ssircar@canon.com (NextMail ok)|Prophet & SPAMIT Charter Member Canon Information Systems and I do not share the same views on everything. "I'm just mad that I missed the sexual revolution." - me "Yes, but you dress much better as a result." - Mike
Newsgroups: comp.sys.next.bugs From: marcel@nice.usergroup.ethz.ch (Marcel Waldvogel) Subject: HD disks with (additional) ED hole Message-ID: <1993Mar17.152534.18961@bernina.ethz.ch> Keywords: Processes hanging in kernel mode accessing floppy, unable to eject Sender: news@bernina.ethz.ch (USENET News System) Organization: NiCE - NeXT User Group, Zuerich Date: Wed, 17 Mar 1993 15:25:34 GMT Yesterday a user inserted a HD disk with additional ED hole he found lying around in one of the disk boxes into one of our NeXTs (running 3.0). The disk inserted didn't spin up, it just stayed in the drive. So far so good. But now to the problems that arised from this disk just silently sitting in the drive: - The user started a tar to that disk. This tar didn't do any disk access and nobody was able to kill that process (sleeping uninteruptibly in kernel mode). - So he wanted to eject that disk. He did a 'disk -e /dev/rfd0b' and the same thing happened to this process. - So we did a reboot (Cmd-~ r). Nothing happened for about a minute. - The next Cmd-~ h led to an immediate halt without sync()ing (as we discovered from the fsck that was done on the next boot). - So we tried to eject the disk the hard way: "ef" from the ROM monitor. The drive shortly came up (shorter than the normal eject sound) and then became silent again, the disk still inside the drive. - Another "ef" didn't even start the eject mechanism. It is not a hardware problem, the disk may be inserted and ejected from Non-NeXT-systems (say "Mac") without problems. So this seems to be a problem with the floppy drivers and since it may generate unkillable processes und unejectable floppies it's quite a hassle. -Marcel
From: fischedj@NeXTwork.Rose-Hulman.Edu (David J. Fischer) Newsgroups: comp.sys.next.bugs Subject: 3.0 --> 2.x nib conversion Date: 17 Mar 1993 17:16:54 GMT Organization: Computer Science Department at Rose-Hulman Distribution: world Message-ID: <1o7me6INNnal@master.cs.rose-hulman.edu> O.K., this isn't actually a bug, but a question about backward compatibility: Nonetheless, I am hoping one or more of you people can help me out with this: What I want: I am looking for a way to convert .nib files created by ProjectBuilder under NS3.x to .nib files that can be loaded by InterfaceBuilder under NS2.x. Why I want it: I am a part-time developer at Rose-Hulman Institute of Technology working on Fluid Science simulations. We have a few hundred NeXT machines running NS2.1 (or maybe 2.2, I'm not sure) and five running 3.0. By this summer, the school's NeXT network will hopefully be ungraded to 3.0 (or maybe even 3.1). This is good for us as it will let us use the newest and neatest 3.0 toys/tools/tricks. However, we are out of luck if we want to distribute our code to other universities that still use 2.x. While the code should remain compatible to 2.x (and what problems there are, can be fixed), the .nib files generated by ProjectBuilder are not readable by InterfaceBuilder on 2.x. (This leads to a Catch 22: If I could load the nib files under 2.x, I could fix them them to work under 2.x. But if I could load them I would be running 3.0 and wouldn't need to fix them.) I am hoping there is a means to convert 3.0 nib files to 2.x format allowing other sites to utilize our code and freeing us from the "shackles" of 2.x. If there is such a process please tell me, show me, push me in the right direction. Thanks: Thank you for you help and consideration. -- David J. Fischer : fischedj@nextwork.rose-hulman.edu "Don't ever put your brain on a runaway train 'cause it won't be coming back!" - Mad At The World.
Newsgroups: comp.sys.next.bugs From: ssircar@canon.com (Subrata Sircar) Subject: Re: SetParaStyle bug ... Message-ID: <9303180007.AA07561@alychne-nc.canon.com> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Thu, 18 Mar 1993 00:07:54 GMT I had a misunderstanding, as it turns out. The bug here is that an unmalloc'd pointer is freed, not that the sample application crashes. This is noted by linking the following program against MallocDebug (I used "cc -g -Wall -o TestStyle TestStyle.m -lMallocDebug -lNeXT_s"): ***** #import <appkit/appkit.h> @interface Delegate:Object { Window *window; } - openMainWindow; @end @implementation Delegate - openMainWindow { NXRect contentRect = {{0.0f, 0.0f}, {200.0f, 200.0f}}; window = [[Window alloc] initContent:&contentRect style:NX_TITLEDSTYLE backing:NX_BUFFERED buttonMask:(NX_CLOSEBUTTONMASK | NX_MINIATURIZEBUTTONMASK) defer :YES]; [window setFreeWhenClosed:YES]; [window center]; [window makeKeyAndOrderFront:self]; return self; } - appDidInit:(const id)sender /* app delegate */ { Text *text; // static NXTabStop ts = {NX_LEFTTAB,5.0}; static NXTextStyle style = {0.0,0.0,0.0,3.0,0,0,NULL}; int timeout = 30000; NXRect contentRect = {{0.0f, 0.0f}, {200.0f, 200.0f}}; [self openMainWindow]; text = [window getFieldEditor:YES for:window]; [[window contentView] addSubview:text]; [text setFrame:&contentRect]; [text setText:"Hello, world."]; [text setParaStyle:(void *)&style]; // [text setParaStyle:[text defaultParaStyle]]; [NXApp perform:@selector(terminate:) with :self afterDelay:timeout cancelPrevious:NO]; return self; } @end /* * This is correct, but gcc will give a warning because methods can't be * qualified. */ volatile int main(void) { Delegate *const delegate = [[Delegate alloc] init]; NXApp = [Application new]; [NXApp setDelegate:delegate]; [NXApp run]; /* NOTREACHED */ } ***** The pointer being freed when the Text object is freed is (deep breath) text->_info->cache.curRun->paraStyle->tabs where _info is a pointer to an NXLayInfo structure, cache is a NXTextCache structure, curRun is a pointer to an NXRun structure, paraStyle is a pointer to an NXTextStyle structure, and tabs is a pointer to an NXTabStop structure, with (presumably) several others following. (I dug all this up by making educated guesses based on the appkit/Text.h header.) If I create my own style structure and use it as the argument to setStyle:, it gets copied into memory somewhere, and that new memory's tabs pointer is freed. However, if that style structure doesn't have any tab stops, then no pointer is freed. So, if you're not using the tab stops feature, you could create a text style structure, set 0 and NULL as the last two fields, and use that. If you are using tab stops, I'd write up a bug report, submit it to NeXT (and to this group, please, so we know the number!) and wait for NS/Intel. I'm not sure if there's another way around it, although I'm not a Text object guru and will gladly defer to those who are. --- Subrata Sircar|ssircar@canon.com (NextMail ok)|Prophet & SPAMIT Charter Member Canon Information Systems and I do not share the same views on everything. "I'm just mad that I missed the sexual revolution." - me "Yes, but you dress much better as a result." - Mike
Newsgroups: comp.sys.next.bugs From: asm@eecg.toronto.edu (Anees S. Munshi) Subject: Re: Mail.app problems ... Message-ID: <1993Mar18.141951.8599@jarvis.csri.toronto.edu> References: <C3pG1q.Mnr@unixhub.SLAC.Stanford.EDU> <1993Mar16.141200.1244@proximus.north.de> Date: 18 Mar 93 19:19:51 GMT gemoe@proximus.north.de (Gerhard Moeller) writes: >In article <C3pG1q.Mnr@unixhub.SLAC.Stanford.EDU> bobcook@slac.stanford.edu >(Bob Cook) writes: >> I did exactly your four steps above, and Mail crashed. When I brought it >back up, the new item was there as you describe. >So there are different versions. Mine didn't crash. The message was there. >Am I lucky? :-) Could it be that 2.1 works, while 3.0 crashes? ------------------------------------------------------------------------- Anees Munshi, University of Toronto, Department of Electrical Engineering Email:asm@eecg.toronto.edu (NeXTMail accepted) Telephone: 416.929.5754 (Toronto), Fax: 416.929.5744 -------------------------------------------------------------------------
Newsgroups: comp.sys.next.bugs From: mmoss@ic.sunysb.edu (Matthew D Moss) Subject: Looking for NS2.0 Fax printing patch Message-ID: <C4Aqz1.6@max.physics.sunysb.edu> Sender: news@max.physics.sunysb.edu (News Administration) Organization: SUNY at Stony Brook, Computer Store Date: Mon, 22 Mar 1993 15:28:13 GMT I had a problem reported to me about a user having trouble printing faxes under NeXT 2.0/2.1. When he asked to print one out, it would not do anything. Does anyone know about this problem, and if such a patch exists to fix it? (Sorry I can't be more specific about the problem) Thanx in advance.... -- +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Matthew David Moss | Blessed are the pure in heart, for they | | INTERNET: mmoss@ic.sunysb.edu | will see God. | | BITNET : mmoss@sbccmail | Matthew 5:8 |
Newsgroups: comp.sys.next.bugs From: gerben@rna.indiv.nluug.nl Subject: On request, the NS 3.0 cc bug (gcc 1.93) Message-ID: <1993Mar22.212600.15179@rna.indiv.nluug.nl> Sender: gerben@rna.indiv.nluug.nl (Gerben Wierda) Organization: G.R.O.S.S. Date: Mon, 22 Mar 1993 21:26:00 GMT /* As requested, this program shows a bug in gcc 1.93 */ /* This demonstrates a bug in gcc 1.93. (see below) */ /* The bug is gone in gcc 2.3.3 */ typedef unsigned long OID; typedef long int32; typedef unsigned long uint32; typedef short int16; typedef unsigned short uint16; typedef unsigned char Boolean; struct attribute { OID attrelid; char attname[16]; OID atttypid; OID attdefrel; uint32 attnvals; OID atttyparg; /* type arg for arrays/spquel/procs */ int16 attlen; int16 attnum; uint16 attbound; Boolean attbyval; Boolean attcanindex; OID attproc; /* spquel? */ uint32 attnelems; int32 attcacheoff; /* char *attlock; */ }; /* * SHORTALIGN(LEN) - length (or address) aligned for shorts */ #define SHORTALIGN(LEN)\ (((long)(LEN) + 1) & ~01) /* * LONGALIGN(LEN) - length (or address) aligned for longs */ #define LONGALIGN(LEN)\ (((long)(LEN) + 3) & ~03) #define fetchatt(A, T) \ ((*(A))->attlen < 0 ? (char *) (LONGALIGN((T) + sizeof(long))) : \ ((*(A))->attbyval ? \ ((*(A))->attlen > sizeof(short) ? (char *) *(long *) (T) : \ ((*(A))->attlen < sizeof(short) ? (char *) *(T) : \ (char *) * (short *) (T))) : (char *) (T))) /* the "attlen < sizeof(short)" branch is `optimized to oblivion'. */ char * testje(att, tp) struct attribute *att[]; char *tp; { return fetchatt(att, tp); } /* *** COMPILER DIAGNOSTICS: *** Reading specs from /lib/m68k/specs gcc version 1.93 (68k, MIT syntax) /lib/m68k/cpp -lang-c -v -undef -D__GNUC__=2 -D__GNU__ -Dmc68000 -Dm68k -DNeXT -Dunix -D__MACH__ -D__BIG_ENDIAN__ -D__ARCHITECTURE__="m68k" -D__mc68000__ -D__m68k__ -D__NeXT__ -D__unix__ -D__MACH__ -D__BIG_ENDIAN__ -D__ARCHITECTURE__="m68k" tupbug.c /usr/tmp/cc001529.cpp GNU CPP version 1.93 (68k, MIT syntax) /lib/m68k/cc1obj /usr/tmp/cc001529.cpp -quiet -dumpbase tupbug.c -version -o tupbug.s GNU C version 1.93 (68k, MIT syntax) compiled by GNU C version NeXT devkit-based CPP 3.0. tupbug.c: In function `testje': tupbug.c:57: warning: cast to pointer from integer of different size tupbug.c:57: warning: cast to pointer from integer of different size *** COMPILER OUTPUT: *** #NO_APP .text .align 1 .globl _testje _testje: link a6,#0 movel a6@(8),a0 movel a0@,a0 tstw a0@(36) jge L2 movel a6@(12),d0 addql #7,d0 moveq #-4,d1 andl d1,d0 jra L3 L2: movel a6@(8),a0 movel a0@,a0 tstb a0@(42) jeq L4 movel a6@(8),a0 movel a0@,a0 cmpw #2,a0@(36) jls L6 movel a6@(12),a0 movel a0@,d0 jra L7 L6: *** ?? Where is my other branch ?? *** movel a6@(12),a0 moveb a0@,d0 extbl d0 L7: jra L5 L4: movel a6@(12),d0 L5: L3: jra L1 L1: unlk a6 rts */ -- Gerben Wierda [NeRD:7539] Tel. (+31) 35 833539 "If you don't know where you are going, any road will take you there." From the Talmud(?), rephrased in Lewis Carroll, "Alice in Wonderland".
Newsgroups: comp.sys.palmtops,comp.sys.nsc.32k,comp.sys.novell,comp.sys.northstar,comp.sys.next.sysadmin,comp.sys.next.misc,comp.sys.next.bugs From: buzy@quads.uchicago.edu (Len Buzyna) Subject: DidYouKnow... Message-ID: <1993Mar23.050634.6576@midway.uchicago.edu> Sender: news@uchinews.uchicago.edu (News System) Organization: University of Chicago Computing Organizations Distribution: usa Date: Tue, 23 Mar 1993 05:06:34 GMT Today Japanese companies own the 7/11 store chain, Dunlop, Universal Pictures, Columbia Pictures, Loews Theaters, MCA Home Entertainment, Tri-Star Pictures, CBS Records, Columbia Records, Spencers stores, Ciniplex Odeon (a big part), Firestone Tires and many many more very large US companies while foreigners are prevented from owning any important Japanese concerns. Ordinary Japanese are kind wonderful people, but what their companies and government are doing is racist and wrong. To find out more about this (and get a more complete list of the above), read (JAPANYES) "Does America Say Yes To Japan?";Leclerc 1992,93 which is available on INTERNET. (most recent edition is v031993). This thoughtfully written and important article has been circulating widely in many of America's biggest corporations & universities like IBM & Harvard. When you read it (it takes about 30 minutes), you'll see why. The essay provides a frightening yet fascinating detailed, referenced overview of the Japanese industrial machine at work and how Japan practices 'business is war' strategies to target and take over strategic critical U.S. industries like high technology, popular media and heavy industry as well as influence the decisions of the US government in favor of Japan. It is a very moving piece and is filled with many verifiable and disturbing examples. You can get JAPANYES 1 of 3 ways: 1)FTP to monu6.cc.monash.edu.au it's in directory: pub/nihongo as: JAPANYES 2)The article has been posted in its entirety (in three sections however) in the misc.test & soc.culture.usa & sci.econ newsgroups. Search on the author 'buzy' or the title 'article' to find the posts. 3)Email a request for JAPANYES to ar12@midway.uchicago.edu He will email you a copy.
Newsgroups: comp.text,comp.sys.next.bugs From: eps@futon.SFSU.EDU (Eric P. Scott) Subject: [Adobe TranScript] ptroff doesn't render bold \(sq correctly Message-ID: <1993Mar23.062610.6505@csus.edu> Followup-To: comp.text Sender: news@csus.edu Organization: San Francisco State University Date: Tue, 23 Mar 1993 06:26:10 GMT In "old" (C/A/T) troff, a bold \(sq is supposed to display as a filled box [_A TROFF Tutorial_ (USD:25)], but pscat translates it to a hollow square. [Since I'm not using ditroff, \(bx isn't available.] troff.font/chartab.inc contains 0202 R 0154 -1 PFONT 0 0 DEFSYM 0140 "square box" 0202 I 0154 -1 PFONT 0 0 DEFSYM 0140 "ital square box" 0202 B 0154 -1 PFONT 0 0 DEFSYM 0140 "bold square box" ^^^^ should be 0141 Tested on: NeXTSTEP 2.x (bundled), NeXTSTEP 3.0 (bundled), SPARC Transcript 2.1. -=EPS=-
Control: cancel <1993Mar23.050634.6576@midway.uchicago.edu> Newsgroups: comp.sys.palmtops,comp.sys.nsc.32k,comp.sys.novell,comp.sys.northstar,comp.sys.next.sysadmin,comp.sys.next.misc,comp.sys.next.bugs From: news@wakinyan.uchicago.edu (Newsmistress) Subject: cmsg cancel <1993Mar23.050634.6576@midway.uchicago.edu> Message-ID: <1993Mar24.030225.6832@midway.uchicago.edu> Sender: news@uchinews.uchicago.edu (News System) Organization: University of Chicago Computing Organizations References: <1993Mar23.050634.6576@midway.uchicago.edu> Distribution: usa Date: Wed, 24 Mar 1993 03:02:25 GMT <1993Mar23.050634.6576@midway.uchicago.edu> was cancelled from within trn.
Newsgroups: comp.sys.next.bugs From: kevins@slow.inslab.uky.edu Subject: Re: DidYouKnow... Message-ID: <C4Cz72.389@ms.uky.edu> Sender: news@ms.uky.edu (USENET News System) Organization: University Of Kentucky, Dept. of Math Sciences References: <1993Mar23.050634.6576@midway.uchicago.edu> Distribution: usa Date: Tue, 23 Mar 1993 20:21:01 GMT Len Buzyna writes -> Today Japanese companies own the 7/11 store chain, Dunlop, Universal Pictures, -> Columbia Pictures, Loews Theaters, MCA Home Entertainment, Tri-Star Pictures, -> CBS Records, Columbia Records, Spencers stores, Ciniplex Odeon (a big part), -> Firestone Tires and many many more very large US companies while foreigners -> are prevented from owning any important Japanese concerns. Ordinary Japanese -> are kind wonderful people, but what their companies and government are doing -> is racist and wrong. -> -> To find out more about this (and get a more complete list of the above), -> read (JAPANYES) "Does America Say Yes To Japan?";Leclerc 1992,93 which is -> available on INTERNET. (most recent edition is v031993). This thoughtfully -> written and important article has been circulating widely in many of America's -> biggest corporations & universities like IBM & Harvard. When you read it (it -> takes about 30 minutes), you'll see why. -> -> The essay provides a frightening yet fascinating detailed, referenced -> overview of the Japanese industrial machine at work and how Japan practices -> 'business is war' strategies to target and take over strategic critical U.S. -> industries like high technology, popular media and heavy industry as well as -> influence the decisions of the US government in favor of Japan. It is a very -> moving piece and is filled with many verifiable and disturbing examples. -> You can get JAPANYES 1 of 3 ways: -> -> 1)FTP to monu6.cc.monash.edu.au it's in directory: pub/nihongo as: JAPANYES -> 2)The article has been posted in its entirety (in three sections however) -> in the misc.test & soc.culture.usa & sci.econ newsgroups. Search on the -> author 'buzy' or the title 'article' to find the posts. -> 3)Email a request for JAPANYES to ar12@midway.uchicago.edu -> He will email you a copy. The above post does not belong in this group. -- _______________________________________________________________________________ I'll cross my heart and hope to die but the needle's already in my eye. _______________________________________________________________________________ Kevin Solie
From: basiji@stein.u.washington.edu (David Basiji) Newsgroups: comp.sys.next.bugs Subject: Re: DidYouKnow... Date: 24 Mar 1993 16:03:42 GMT Organization: University of Washington Distribution: usa Message-ID: <1oq0ouINNcso@shelley.u.washington.edu> References: <1993Mar23.050634.6576@midway.uchicago.edu> <C4Cz72.389@ms.uky.edu> The original poster has had his account privileges revoked. David Basiji UW Bioengineering
From: "Jeremy G. Mereness" <zonker+@CMU.EDU> Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: NFS Server Bug, 3.0? Date: Thu, 25 Mar 1993 16:24:20 -0500 Organization: Graduate School of Industrial Administr., Carnegie Mellon, Pittsburgh, PA Message-ID: <8fgWA4K00WB30VGktH@andrew.cmu.edu> The Incident: creating a new network. all the clients are fine. the server is mounting a set of directories on an external SCSI drive to be mounted by the clients as /LocalApps, /LocalLibrary, and /usr/local. In NetInfo, "/server" exports its directories and "/" imports them as their respective mountpoints. There are also a couple of directories on the server being mounted as "/Net". 6 directories are being broadcast on the network total. I noticed that the server was mounting the directories just like the clients (because I had imported as "/"). "Fine," I thought, "redundant, but okay." Too bad you can't make one or two exceptions from "/". I started installing Mathematica through the mountpoint /LocalApps on the server. Everything went fine until the end of the second disk. Then, the server went offline throughout the network, including to itself (/server not responding... still trying). Any operation would just poll for a disk, but never find one. I have to forcefully shutdown through the monitor. If I umount the mountpoints on the server and recreate them as symbolic links, everything goes fine. Question 1: has anyone seen this kind of behavior on a 3.0 server? Is this c'est la vie or am I putting too big a load on it? Could I help things with the NFS Expert options, like lengthening the time-out period? or is it something else? Question 2: I wanna hack it. Let me modify the rc or rc.local file on the server so that it umounts the NFS directories and replaces them with symbolic links to the external drive. Where in the rc files should I put this so that I can be sure that all the NFS mounts have been mounted already? Thanks in Advance! ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ |Jeremy Mereness | Support | Ye Olde Disclaimer: | |zonker+@cmu.edu (internet) | NeXTStep | This message represents| | or jm7e+@andrew.cmu.edu | Software | my opinions, alone. | |FAST Laboratory, CMU-GSIA |------------| Thank you, Mask Man. | | B.S. Mech Eng. CMU,1992| Certificate Mixology,1992 |No NeXTMail, yet | ----------------------------------------------------------------------
Newsgroups: comp.sys.next.bugs,comp.protocols.ppp From: mbparker@netcom4.com (Michael Benjamin Parker) Subject: NeXT PPP Users: got remote domain name service working? How?! Message-ID: <MBPARKER.93Mar27072036@netcom4.com> Sender: mbparker@netcom.com (Michael Benjamin Parker) Organization: Netcom Online Communications Service Date: Sat, 27 Mar 1993 15:20:36 GMT If you're a NeXT PPP user, were you able to get domain name service working? That is, get your next to translate a normal hostname, such as netcom.com, into an IP address, such as 192.100.81.100, when your nameserver is not local and must be accessed via the PPP? I can't get it to work. If you have, what did you do? Actually, nslookup works fine, but all the other TCP/IP utilities such as ping and telnet can't translate hostnames to IP addresses. Below is a detailed email describing my situation and a response from my PPP vendor saying, `Yep, it's a bug with NeXT' (see below) but not giving a way to fix it. Thanks for any help you can give, -Mike Parker ****************************************************************************** {Return-Path: <mbparker> Date: Wed, 24 Mar 93 04:02:07 -0800 From: mbparker (Michael Benjamin Parker) To: support@morningstar.com Subject: Help! General hostname lookup always fails while nslookup works fine Reply-To: mbparker@mit.edu I am a registered customer (you can look up my customer id if you want; I'm Michael B. Parker from Oracle Corp.) and I have an odd bug. Normal host name lookup (done by ping or telnet or other TCP/IP utility) is failing, yet nslookup works fine. For instance:{ share/etc# ping nntp.netcom.com [A 1 minute delay where I can see nothing happening (probably checking some source which isn't responding, then times out), then the ppp link is started up and I get:] ping: unknown host nntp.netcom.com share/etc# telnet nntp.netcom.com nntp.netcom.com: Host name lookup failure share/etc# nslookup nntp.netcom.com [No delay at all; the ppp link, if down, is started up immediately.] Server: netcomsv.netcom.com Address: 192.100.81.101 res_mkquery(0, nntp.netcom.com.netcom.com, 1, 1) ------------ Got answer: HEADER: opcode = QUERY, id = 2, rcode = NXDOMAIN header flags: response, auth. answer, want recursion, recursion avail. questions = 1, answers = 0, authority records = 1, additional = 0 QUESTIONS: nntp.netcom.com.netcom.com, type = A, class = IN AUTHORITY RECORDS: -> netcom.com ttl = 3600 (1 hour) origin = netcomsv.netcom.com mail addr = root.netcomsv.netcom.com serial = 70000 refresh = 3600 (1 hour) retry = 3600 (1 hour) expire = 3600 (1 hour) minimum ttl = 3600 (1 hour) ------------ res_mkquery(0, nntp.netcom.com, 1, 1) ------------ Got answer: HEADER: opcode = QUERY, id = 3, rcode = NOERROR header flags: response, auth. answer, want recursion, recursion avail. questions = 1, answers = 1, authority records = 0, additional = 0 QUESTIONS: nntp.netcom.com, type = A, class = IN ANSWERS: -> nntp.netcom.com internet address = 192.100.81.105 ttl = 3600 (1 hour) ------------ Name: nntp.netcom.com Address: 192.100.81.105 share/etc# } (``telent 192.100.81.105'' and similar, i.e., using the IP address instead of the hostname, works fine, BTW.) I can't figure out why normal name service isn't working, but nslookup works. This is my situation. I'm running NeXTstep 3.0 and Morning Star PPP Version 1.3 Beta [26-Mar-1992 22:10:50]. /etc/ppp/Startup contains: #setup for share-netcom ppp connection: pppd 192.187.158.6:192.187.158.254 netmask 255.255.255.248 auto name mbparker.slip.netcom.com escape 0x91 escape 0x93 && (echo -n ' ppp connection to netcom') >/dev/console /usr/etc/route add default 192.187.158.254 1 /etc/ppp/Systems contains: #Netcom entry via ZyXEL U-1496e by MBParker on 19930324: 192.187.158.254 Any ACU 38400 328-0766 "" \p TIMEOUT 30 ogin: \pSmbparke word: \<password> PPP--PPP /etc/resolv.conf contains: domain netcom.com nameserver 192.100.81.101 nameserver 192.100.81.105 And /etc/named.boot exists but I don't believe is having any affect because no process named ``named'' appears to be running and any attempt to run ``named'' fails immediately with the error: ``<time> share named[<pid>]: bind(dfd 8, 0.0.0.0[53]): Address already in use''. Do you have any idea what's wrong? Thanks for your help, -Mike Parker } ****************************************************************************** {Return-Path: <bob@MorningStar.Com> Date: Thu, 25 Mar 93 16:13:02 -0500 From: Bob Sutterfield <bob@MorningStar.Com> To: mbparker@mit.edu, mbparker@netcom.com (Michael Benjamin Parker) Cc: support@MorningStar.Com Reply-To: support@MorningStar.Com In-Reply-To: Michael Benjamin Parker's message of Wed, 24 Mar 93 04:02:07 -0800 <9303241202.AA22139@netcom4.netcom.com> Subject: Help! General hostname lookup always fails while nslookup works fine : This sounds like a familiar old problem with NeXT's resolver. They glommed named and NetInfo lookup functions both into a single daemon (I forget its name) that doesn't quite work right. You'll note that pppd dials immediately upon receipt of the bringup packet, but that named/NetInfo daemon holds onto the query for too long. : I wonder whether nslookup was compiled with different resolver libraries, or if it doesn't even ask the local named/NetInfo daemon? : This is a bug, or maybe a designed-in behavioral feature, in NeXT's combined named/NetInfo daemon. } -- Michael B. Parker P.O. Box 60951, Palo Alto, CA 94306-0951; voice 415-325-4451 (24 hrs), email MBParker@MIT.Edu (no NeXTmail yet), fax & modem 415-325-8604. -- Michael B. Parker P.O. Box 60951, Palo Alto, CA 94306-0951; voice 415-325-4451 (24 hrs), email MBParker@MIT.Edu (no NeXTmail yet), fax & modem 415-325-8604.
Newsgroups: comp.sys.next.bugs,comp.protocols.ppp From: bob@MorningStar.Com (Bob Sutterfield) Subject: Re: NeXT PPP Users: got remote domain name service working? How?! In-Reply-To: mbparker@netcom4.com's message of Sat, 27 Mar 1993 15: 20:36 GMT Message-ID: <BOB.93Mar27170757@roughy.MorningStar.Com> Sender: news@MorningStar.Com Organization: Morning Star Technologies References: <MBPARKER.93Mar27072036@netcom4.com> Date: Sat, 27 Mar 1993 22:08:04 GMT In article <MBPARKER.93Mar27072036@netcom4.com> mbparker@netcom4.com (Michael Benjamin Parker) writes: ...when your nameserver is not local and must be accessed via the PPP? So far as I know, the workaround seems to be to run a local nameserver. Arrange to be an authoritative secondary for at least your own domain's forward mappings, and your own network's reverse mappings. And take secondary dumps of those domains with which you often need to connect. I have described the problem to NeXT, as have several of our other customers who are in situations similar to yours, but I haven't heard any response yet. -- Bob Sutterfield, Morning Star Technologies +1 614 451 1883 1760 Zollinger Rd, Columbus Ohio USA, 43221 +1 800 558 7827 bob@MorningStar.Com +1 614 459 5054 (FAX)
Newsgroups: comp.sys.next.bugs Subject: Excessive <mach-tasks> solution? Message-ID: <1p5i0pINNda8@news.gac.edu> From: Nathan Lee <lee@shadow.gac.edu> Date: 29 Mar 1993 01:05:29 GMT Distribution: world Organization: Gustavus Adolphus College Hello, A (long) while back, when NS3.0 first started coming out, I noticed a large discussion on the appearance of many <mach-task> processes on new 3.0 machines, including fixes (I believe). Our campus mail-server is currently generating many,many of these processes, and we would greatly appreciate some of you informed readers out there reminding us of how the problem is best solved. Thanks very much! Nathan -- Nathan Lee (lee@gac.edu) ex-NeXT CC :D Gustavus Adolphus College
Newsgroups: comp.sys.next.bugs Subject: Excessive <mach-tasks> solution? Message-ID: <1993Mar29.074855.17388@urz.unibas.ch> From: frank@ifi.unibas.ch (Robert Frank) Date: Mon, 29 Mar 1993 07:48:55 GMT Sender: news@urz.unibas.ch (USENET News System) References: <1p5i0pINNda8@news.gac.edu> Organization: Institut fuer Informatik In article <1p5i0pINNda8@news.gac.edu> writes: > Hello, > > A (long) while back, when NS3.0 first started coming out, I > noticed a large discussion on the appearance of many <mach-task> > processes on new 3.0 machines, including fixes (I believe). > > Our campus mail-server is currently generating many,many of > these processes, and we would greatly appreciate some of you > informed readers out there reminding us of how the problem is > best solved. > > Thanks very much! > > Nathan > -- > Nathan Lee (lee@gac.edu) > ex-NeXT CC :D > Gustavus Adolphus College Same over here, I'd like an answer too! -- Robert Frank tel. + (061) 321 99 67 Institut fuer Informatik fax + (061) 321 99 15 University of Basel, Switzerland Mittlere Strasse 142 rfc822: frank@ifi.unibas.ch (NeXT mail accepted) CH-4056 Basel X400: S=frank;OU=ifi;O=unibas;P=switch;A=arcom;C=ch ( if all fails try frank@urz.unibas.ch )
From: kwang@data.acs.calpoly.edu (Kevin John Wang) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: /etc/mtab? and the auto-diskmounter? and booting? Keywords: mtab mount fsck blah Message-ID: <1993Mar29.200213.160239@zeus.calpoly.edu> Date: 29 Mar 93 20:02:13 GMT Sender: news@zeus.calpoly.edu Followup-To: comp.sys.next.sysadmin Organization: Cal Poly State University, San Luis Obispo Does anyone else's /etc/mtab keep getting chmod'ed to 600? This is wreaking havoc over here because df and SoundWorks refuse to work if they can't read /etc/mtab. 1: I have verified it in as far as on SOME of our machines the file is 600 after boot. SOME machines. 2: I'm fairly sure that on ALL the machines ejecting a floppy causes /etc/mtab to become 600. inserting does not. 3: I *THINK* that the OD has the same effect. 4: I *THINK* that fsck also twiddles with the permissions. help? (besides putting something in cron...) I think that it is either the AutoDiskmounter or maybe even mount/umount anyone? anyone? - Kevin Wang
From: mfausett@bbn.com (Mark Fausett) Newsgroups: comp.sys.next.bugs,comp.protocols.ppp Subject: Re: NeXT PPP Users: got remote domain name service working? How?! Message-ID: <mfausett.733444036@kirin> Date: 29 Mar 93 22:27:16 GMT References: <MBPARKER.93Mar27072036@netcom4.com> <BOB.93Mar27170757@roughy.MorningStar.Com> >In article <MBPARKER.93Mar27072036@netcom4.com> mbparker@netcom4.com (Michael Benjamin Parker) writes: > ...when your nameserver is not local and must be accessed via the > PPP? Curious; what happens when you wait a few minutes and try again, AFTER the link comes up. I use the merit PPP code and have no problems (of course I don't even try to resolve names BEFORE I bring up the link). mf
Newsgroups: comp.sys.next.bugs From: khw2x@sonja.math.Virginia.EDU (Kevin H. Weiss) Subject: Excessive <mach-tasks> solution? Message-ID: <C4o6E9.34u@murdoch.acc.Virginia.EDU> Sender: usenet@murdoch.acc.Virginia.EDU Organization: University of Virginia References: <1993Mar29.074855.17388@urz.unibas.ch> Date: Mon, 29 Mar 1993 21:30:08 GMT In article <1993Mar29.074855.17388@urz.unibas.ch> writes: > In article <1p5i0pINNda8@news.gac.edu> writes: > > Hello, > > > > A (long) while back, when NS3.0 first started coming out, I > > noticed a large discussion on the appearance of many <mach-task> > > processes on new 3.0 machines, including fixes (I believe). > > > > Our campus mail-server is currently generating many,many of > > these processes, and we would greatly appreciate some of you > > informed readers out there reminding us of how the problem is > > best solved. > > > > Thanks very much! > > > > Nathan > > -- > > Nathan Lee (lee@gac.edu) > > ex-NeXT CC :D > > Gustavus Adolphus College > > Same over here, I'd like an answer too! > > -- > Robert Frank tel. + (061) 321 99 67 > Institut fuer Informatik fax + (061) 321 99 15 > University of Basel, Switzerland > Mittlere Strasse 142 rfc822: frank@ifi.unibas.ch (NeXT mail accepted) > CH-4056 Basel X400: S=frank;OU=ifi;O=unibas;P=switch;A=arcom;C=ch > ( if all fails try frank@urz.unibas.ch ) Once, again I have the same problem, and the swapfile gets out of control, too!!! Please send help... anybody... -- -------------------------------------------------------------- Kevin Weiss Department of Mathematics khw2x@sonja.math.Virginia.edu (NeXT mail) University of Virginia khw2x@virginia.edu (Non-NeXT mail)
Newsgroups: comp.sys.next.bugs From: hwr@snert.ka.sub.org (Heiko W.Rupp) Subject: Re: Excessive <mach-tasks> solution? References: <1p5i0pINNda8@news.gac.edu> <1993Mar29.074855.17388@urz.unibas.ch> Date: Tue, 30 Mar 1993 18:20:37 GMT Organization: The Home of the Pilhuhn Sender: news@pilhuhn.ka.sub.org (Das Newssystem auf pilhuhn) Message-ID: <hwr.733515637@snert.ka.sub.org> frank@ifi.unibas.ch (Robert Frank) writes: >> noticed a large discussion on the appearance of many <mach-task> >> processes on new 3.0 machines, including fixes (I believe). >> Our campus mail-server is currently generating many,many of >> these processes, and we would greatly appreciate some of you >> informed readers out there reminding us of how the problem is >> best solved. >Same over here, I'd like an answer too! One program, which produces <mach-task>s is telnet. -- Heiko W.Rupp Gerwigstr.5 7500 Kh'e 1 hwr@pilhuhn.ka.sub.org +49 721 693642 Actors will happen even in the best-regulated families.
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: CCC_HART@rzmain.rz.uni-ulm.de (Hardy) Subject: Help: error in write(2) ??? Message-ID: <1993Apr2.015050.2183@wega.rz.uni-ulm.de> Sender: news@wega.rz.uni-ulm.de (News Net) Organization: University of Ulm, Germany Date: Fri, 2 Apr 1993 01:50:50 GMT Hello NeXTler, I have a big problem: in a project which uses threads i got this error: (gdb) run [...] Program generated(5): Software exception. 0x50081e0 in write () (gdb) After this the program hangs (without debugger the programm terminates). The write(2) is used to put data to a network port. The error occurs only if the application uses two or more objects, which have each one data port, simultaneous an read/write alternately from these ports. Simultaneous read/write doesnt occur. The error could be reproduced. Has anyone any hints ? Thanx in advance, hardy --- please reply to: hardy@first_next.informatik.uni-ulm.de (Next mail) or: Christoph.Hartmann@ccc.uni-ulm.de (Standard mail)
Newsgroups: comp.sys.next.bugs From: otto@epr.jyu.fi (Otto J. Makela) Subject: NS3.0: Installed packages bug? Message-ID: <OTTO.93Apr2194639@epr.jyu.fi> Sender: otto@jyu.fi (Otto J. Makela) Organization: The Crimson Permanent Assurance Company Date: Fri, 2 Apr 1993 17:46:39 GMT When I installed NS3.0 on our mono NeXTstation, I noticed that it did not keep correct track of which packages had been installed. I seem to remember this being talked about in the news, but now I can't remember the solution. I'd like to uninstall some of the sillier language choices and some such from our system, as we're running a bit low on disk space (even with a 600M drive). -- /* * * Otto J. Makela <otto@jyu.fi> * * * * * * * * * * * * * * * * * * */ /* Phone: +358 41 613 847, BBS: +358 41 211 562 (V.32bis/USR-HST,24h/d) */ /* Mail: Kauppakatu 1B18/SF-40100 Jyvaskyla/Finland, ICBM: 62.14N25.44E */ /* * * Computers Rule 01001111 01001011 * * * * * * * * * * * * * * * * */
Newsgroups: comp.sys.next.misc,comp.sys.next.bugs From: otto@epr.jyu.fi (Otto J. Makela) Subject: NS3.0: Bug with printing peculiar-sized envelopes? Message-ID: <OTTO.93Apr2194743@epr.jyu.fi> Followup-To: comp.sys.next.bugs Sender: otto@jyu.fi (Otto J. Makela) Organization: The Crimson Permanent Assurance Company Date: Fri, 2 Apr 1993 17:47:43 GMT While we were still running our mono NeXTstation+laser under NS2.1, my girlfriend setup WriteNow for some special envelopes she prints. I believe they were B5 size (right for a folded A4). Ever since I upgraded the machine to NS3.0, the envelopes she's tried to produce with this setup have been misprinted so that the address has been placed somewhere too far right so that half of them are lost. No amount of twiddling with the size settings has helped. Known problem with NS3.0? Fix? -- /* * * Otto J. Makela <otto@jyu.fi> * * * * * * * * * * * * * * * * * * */ /* Phone: +358 41 613 847, BBS: +358 41 211 562 (V.32bis/USR-HST,24h/d) */ /* Mail: Kauppakatu 1B18/SF-40100 Jyvaskyla/Finland, ICBM: 62.14N25.44E */ /* * * Computers Rule 01001111 01001011 * * * * * * * * * * * * * * * * */
Newsgroups: comp.sys.next.bugs Subject: Getting other renderers to work under NetInfo Message-ID: <1993Apr4.132838.4753@latcs1.lat.oz.au> From: nicolek@latcs1.lat.oz.au (Nicole KAIYAN) Date: Sun, 4 Apr 1993 13:28:38 GMT Organization: Comp Sci, La Trobe Uni, Australia Keywords: NetInfo Summary: Getting other renderers to work under NetInfo Basically, I have altered my hostname with HostManager and wish to select it as my renderer under NetInfo. I have followed the 3DKit release notes and have inserted localhost as an alias for my new name host, but alas this still causes the problem of trying to connect to the server and errors. Can an elightened soul assist please ? --------------------------------- ------------------------------------- | Nicole Kaiyan | Email: nicolek@latcs1.lat.oz.au | | Computer Science Department | nk@saturn.cs.swin.oz.au | | La Trobe University | | | Melbourne 3083 | NeXTmail welcome | | AUSTRALIA | | --------------------------------- -------------------------------------
Newsgroups: comp.sys.next.bugs From: wching@alice.music.yale.edu (William Ching) Subject: Bug? /dev/ttyp* files are being set to group wheel Message-ID: <1993Apr4.223822.14301@cs.yale.edu> Sender: news@cs.yale.edu (Usenet News) Organization: Yale School of Music, Center for Studies in Music Technology Date: Sun, 4 Apr 1993 22:38:22 GMT I've been having problems with the files in /dev. The pseudo-tty files (ttyp*) are getting set to group wheel and there are also some strange access modes being set (i.e. owner ronly and no access to group or other). Among other things these files are also not being reset to the default state (owner root, grp tty, ugo=rw). The only solution I've come up with is manually resetting these files and a cron entry to reset unused terminals. Evidently it seems that pseudo-ttys created by the telnet daemon and/or rlogin daemon are responsible for this screwiness. I'm running NeXTStep 3.0 on a cube. Thanks in advance. WC ----------------------------------------------------------------------------- William Ching | wching@csmt.music.yale.edu Yale University Department of Music | Center for Studies in Music Technology "To the insane the sane must seem insane"
Newsgroups: comp.sys.next.bugs From: sherwood@space.ualberta.ca (System Administrator) Subject: 2.x diskeater bugs Message-ID: <1993Apr4.235547.14381@kakwa.ucs.ualberta.ca> Sender: news@kakwa.ucs.ualberta.ca Organization: University Of Alberta, Edmonton Canada Date: Sun, 4 Apr 1993 23:55:47 GMT These two bugs exist under 2.1-2. I suspect that they have not been checked for 3.0 1. /usr/spool/appkit. Under unknown circumstances these file are not deleted. The problem is insidious, as they are spread out over time. On one machine I had 18 MB of these with time stamps spread over 2 years. 2. /usr/tmp By default if sort can't work entirely in memory, it works in /usr/tmp. This is particularly obnoxious if you run /usr/lib/find/updatedb and it runs out of room, as it just crashes, and doesn't remove the files. Solution: Modify /etc/rc, and add a line near where it clears /tmp to clear /usr/spool/appkit Modify /usr/lib/find/updatedb to explicitly sort on a partition that has more room (sort -T ..) -- => Sherwood Botsford sherwood@space.ualberta.ca <= => University of Alberta Lab Manager, Space Physics Group <= => tel:403 492-3713 fax: 403 492-4256 <=
From: mcm@vax.oxford.ac.uk Newsgroups: comp.sys.next.bugs Subject: Re: /etc/mtab? and the auto-diskmounter? and booting? Message-ID: <1993Mar30.001253.12872@vax.oxford.ac.uk> Date: 29 Mar 93 23:12:53 GMT References: <1993Mar29.200213.160239@zeus.calpoly.edu> Followup-To: comp.sys.next.sysadmin Organization: Oxford University VAX 6620 In article <1993Mar29.200213.160239@zeus.calpoly.edu>, kwang@data.acs.calpoly.edu (Kevin John Wang) writes: > Does anyone else's /etc/mtab keep getting chmod'ed to 600? This is > wreaking havoc over here because df and SoundWorks refuse to work if they > can't read /etc/mtab. > 1: I have verified it in as far as on SOME of our > machines the file is 600 after boot. SOME machines. > 2: I'm fairly sure that on ALL the machines ejecting a floppy causes > /etc/mtab to become 600. inserting does not. > 3: I *THINK* that the OD has the same effect. > 4: I *THINK* that fsck also twiddles with the permissions. Yes, exactly the same problem, no solution yet unfortunately. /\/\att Millar NeXTmail empire@teaching3.physics.ox.ac.uk
Newsgroups: comp.sys.next.bugs From: gad@eclipse.its.rpi.edu (Garance A. Drosehn) Subject: Re: NS3.0: Installed packages bug? Message-ID: <8ag5h8-@rpi.edu> References: <OTTO.93Apr2194639@epr.jyu.fi> Date: Mon, 5 Apr 1993 07:44:36 GMT otto@epr.jyu.fi (Otto J. Makela) writes: > When I installed NS3.0 on our mono NeXTstation, I noticed that it > did not keep correct track of which packages had been installed. > I seem to remember this being talked about in the news, but now > I can't remember the solution. I'd like to uninstall some of the > sillier language choices and some such from our system, as we're > running a bit low on disk space (even with a 600M drive). I mentioned it in one of the NeXT newsgroups at some point, but I don't know if it's on any of the official "bugs" lists. The mixup is that the NS3.0 install sets up /NextLibrary/Receipts as if all the languages are installed, even for those languages you have not in fact installed. What's worse is that each of those receipt files (well, it's really a ".pkg" directory) is over a megabyte. The problem is that these receipt directories for uninstalled languages include the tar file which would be used to install that language. The second problem is that the tar file is completely useless, because the .pkg directory does not include the information which would tell package how to install that language. So, the thing to do is to double-click on each of the language packages in /NextLibrary/Receipts (except the english one, or whatever your primary one is, I would guess), which will start up the installer. Then delete the package. For any language you *do* want installed, go back to the CD-ROM and install it from wherever it is on the CD-ROM (I'm on vacation or I'd look it up for you...). It would probably be good to cross check each of those packages before deleting it. Any package which was installed correctly should have a ".pkg" file in /NextLibrary/Receipts which is under 200K. In fact, the ones for languages will probably be under 100k. Ones that are mixed up will be over 1meg. I vaguly remember that there was one other (non-language) package which was goofed up this way, but I don't remember which one it was. -- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
Newsgroups: comp.sys.next.bugs From: prail@doc.tip.ameslab.gov (joel prail) Subject: NeXT Turbo gags on floppies Message-ID: <C50LB8.Mq4@news.iastate.edu> Sender: news@news.iastate.edu (USENET News System) Organization: Iowa State University, Ames IA Date: Mon, 5 Apr 1993 14:23:31 GMT Our office NeXTs seem to be having problems with their floppies. Each of the NeXT Turbo machines incorrectly or fail to recognize 3.5" disks. A significant portion of the time someone will stick in a Next-formatted floppy and it will be recognized as a DOS disk, but not the same icon as REAL dos disks. Putting in a dos disk sometimes is completely ignored, other times it mounts but only a folder appears, and other time (not very frequently...) it works. This happens on all 7 turbos, but NOT on any of the un-turbo nexts. We are currently using NS 3.0. -Joel Prail Computer Engineer Technology Integration Program Ames Laboratory, Iowa State University
Newsgroups: comp.sys.next.bugs From: mrb@earth.wustl.edu (Mike Bray) Subject: 3.0 - printing - feature selections dropped for remote printers Message-ID: <1993Apr6.065921.14571@wuecl.wustl.edu> Sender: usenet@wuecl.wustl.edu (News Administrator) Organization: Washington University, St. Louis MO Date: Tue, 6 Apr 1993 06:59:21 GMT We have an HP IIISi printer with the appletalk interface. We have a gatorbox running gatorprint, which appears to the next as a remote unix system that accepts lpd-style print requests. We can print just fine, from the PrintPanel. However, the problem is that selected options on the print panel don't make it to the printer. The ones we really need are upper/lower paper tray selections and duplex (both sides of the page) mode. NeXT explains that the .ppd files are not being read for remote printers, but that it would work ok for local printers. This is consistent with our experience that saving the print job to a device dependent postscript file from the print panel works fine, with the feature selections properly embedded. As yet, NeXT hasn't offered a work-around. Has anyone overcome this problem? Some ideas I'm considering: 1. Work some magic with NetInfo to make it think the printer is local somehow, so that "it just works" would be true for once? 2. Figure out how to get the "admin/InitFiles" property in NetInfo to work. According to the "printing" release note that came with 3.0, by specifying a path to a postscript file as the value of the property "InitFiles", that file will be downloaded before printjobs go to the printer. This didn't work for me. What is the format of the file to be downloaded? I was thinking we could pre-load this file turning on the options we want, before printing. We'd have to be careful because we have 6 NeXTs on the Net printing...
Newsgroups: comp.sys.next.bugs From: nico@imani.cam.org(Nicolas Dore) Subject: Re: Getting other renderers to work under NetInfo Message-ID: <1993Apr5.140320.1424@imani.cam.org> Sender: nico@imani.cam.org References: <1993Apr4.132838.4753@latcs1.lat.oz.au> Date: Mon, 5 Apr 1993 14:03:20 GMT In article <1993Apr4.132838.4753@latcs1.lat.oz.au> writes: > Basically, I have altered my hostname with HostManager and wish to select it > as my renderer under NetInfo. I have followed the 3DKit release notes and > have inserted localhost as an alias for my new name host, but alas this > still causes the problem of trying to connect to the server and errors. > > Can an elightened soul assist please ? > > --------------------------------- ------------------------------------- > | Nicole Kaiyan | Email: nicolek@latcs1.lat.oz.au | > | Computer Science Department | nk@saturn.cs.swin.oz.au | > | La Trobe University | | > | Melbourne 3083 | NeXTmail welcome | > | AUSTRALIA | | > --------------------------------- ------------------------------------- > > This is probably too late because of my uucp link, but anyway... Go into NetInfoManager, open the "locations" directory, and Voil'a! Change "localhost" to your hostname, and you'rew there. Probably too late, but hey... Nicolas BTW Change your hostname in the "/etc/rc*" scripts. At one point, there's a "hostname localhost" line, wich sets the command line prompt to "localhost". I don't know if changing it means anything to rendering, but it does to uucp, and it looks nice...
From: sears@tree.egr.uh.edu (Paul S. Sears) Newsgroups: comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.bugs Subject: Bug in Newsgrazer version 72.3? Date: 6 Apr 1993 16:52:22 GMT Organization: University of Houston Message-ID: <1pscg6$hoi@menudo.uh.edu> Newsgrazer Release 2.0, version 72.3 We are having problems with NewsGrazer. It suddenly quits at random times. If run from the shell window, it returns: "buss error" so I decided to run it via gdb. It seems to crash consistantly with a memory exception error (same address). Anyone else have this problem?? My gdb sessom is below: thanatos> gdb NewsGrazer Reading symbol data from NewsGrazer... (no debugging symbols found)...done. Reading symbol data from /usr/shlib/libNeXT_s.C.shlib...done. Reading symbol data from /usr/shlib/libsys_s.B.shlib...done. (gdb) run Starting program: /NeXTMount2/PublicApps/NewsGrazer.app/NewsGrazer Program generated(1): Memory access exception on address 0xb4a21f8 (invalid address). 0x6060000 in DPSAddFD () (gdb) info program Using the running image of child process 1708. Program stopped at 0x6060000. (gdb) info regi Register Contents (relative to selected stack frame) d0 0x200000 2097152 d1 0x15 21 d2 0x1d1c1a1 30523809 d3 0x3a383435 976761909 d4 0xffffffff -1 d5 0xe69 3689 d6 0xe69 3689 d7 0xffffffff -1 a0 0x4031b74 67312500 a1 0x7615c 483676 a2 0x1d0190 1900944 a3 0x6e35c 451420 a4 0x5002d8c 83897740 a5 0x1a86c8 1738440 fp 0x3ffde50 0x3ffde50 sp 0x3ffde3c 0x3ffde3c ps 0x0 0 pc 0x6060000 0x6060000 fp0 251536.34304000001 (raw 0x40100000fffffff5ffffffa415fffffff45e0b4e12) fp1 2428239930 (raw 0x401e0000ffffff90ffffffbc003a00000000) fp2 251536.34304000001 (raw 0x40100000fffffff5ffffffa415fffffff45e0b5000) fp3 +Infinity (raw 0x7fffffffff0000ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff ffff) fp4 +Infinity (raw 0x7fffffffff0000ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff ffff) fp5 +Infinity (raw 0x7fffffffff0000ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff ffff) fp6 +Infinity (raw 0x7fffffffff0000ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff ffff) fp7 +Infinity (raw 0x7fffffffff0000ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff ffff) fpcontrol 0x0 0 fpstatus 0x88 136 fpiaddr 0x6020cc4 100797636 (gdb) info stack #0 0x6060000 in DPSAddFD () #1 0x144ce in ?? () #2 0x5df6 in ?? () #3 0x59a8 in ?? () #4 0x5424 in ?? () #5 0x54ea in ?? () #6 0x6020cd0 in checkTEs () #7 0x601c344 in _DPSGetOrPeekEvent () #8 0x601d726 in NXGetOrPeekEvent () #9 0x602e47e in -[Application run] () #10 0x11706 in ?? () -- Paul S. Sears * sears@uh.edu (NeXT Mail OK) The University of Houston * suggestions@tree.egr.uh.edu (NeXT Engineering Computing Center * comments, complaints, questions) NeXT System Administration * DoD#1967 '83 NightHawk 650SC >>> SSI Diving Certification #755020059 <<< "Programming is like sex: One mistake and you support it a lifetime."
From: rhess@consilium.COM (Richard L. Hess) Newsgroups: comp.sys.next.bugs Subject: Re: /etc/mtab? and the auto-diskmounter? and booting? Message-ID: <RHESS.93Apr7073601@consilium.COM> Date: 7 Apr 93 14:36:01 GMT References: <1993Mar29.200213.160239@zeus.calpoly.edu> <1993Mar30.001253.12872@vax.oxford.ac.uk> Sender: root@cimshop.UUCP Followup-To: comp.sys.next.sysadmin Organization: Consilium Inc., Mountain View, California. In-reply-to: mcm@vax.oxford.ac.uk's message of 29 Mar 93 23:12:53 GMT Here's something that I posted about this problem in January. +---------------------------------------------------[ forwarded message ] From: rhess@consilium.COM (Richard L. Hess) Date: 25 Jan 93 09:35:44 In-reply-to: davisre@sage.cc.purdue.edu's message of 23 Jan 93 00:01:36 GMT Newsgroups: comp.sys.next.bugs Subject: Re: /etc/mtab permissions >>>>> "Robert" == Robert Davis <davisre@sage.cc.purdue.edu> writes: Robert> This may have been mentioned before (I recently upgraded to Robert> 3.0), but is there a workaround or fix for this: ejecting a mounted Robert> floppy changes the permissions of /etc/mtab so that you can't do a df. I had the same problem. It took me a while to figure it out, but I'd bet that you have very tight permissions set on your "File Creation Mask". If you just use "Expert Preferences" (Unix) in the Preferences tool to open up your permissions to allow "Group" & "Others" Read access, then you won't have any problems with the permissions changing on /etc/mtab. Evidently when the eject script/program executes it runs as root and uses your default permissions and so if your permissions are too "tight" you can't read /etc/mtab after the floppy get's ejected. Hope this helps, Dick... ps. I don't remember whether it was ejecting Macintosh or DOS format floppies that caused this problem, but one worked and the other didn't and I never had problems with the NeXT (Unix) format floppies... +---------------------------------------------------[ End forwarded message ] Hope this helps, Dick... +---------------------------------------------------------- WHO: Richard Hess [ Staff Engineer ] CORP: Consilium Inc, 640 Clyde Court, Mtn View, CA 94043 VOICE: [415] 691-6342 EMAIL: rhess@consilium.COM <-----------------[ NeXTmail ] +
Newsgroups: comp.sys.next.bugs From: Ahti.Kotisaari@helsinki.fi Subject: NeXT400dpiPrinter prints incomplete pages Message-ID: <1993Apr7.185530.23880@klaava.Helsinki.FI> Sender: news@klaava.Helsinki.FI (Uutis Ankka) Organization: University of Helsinki Date: Wed, 7 Apr 1993 18:55:30 GMT Maybe this is not a bug, but... My NeXT 400dpi Laser Printer prints all documents as follows: nearly every page is first printed incompletely (from one to two thirds of page length), and then properly a complete one! Why? The former owner told it didn't behave like that. I just reset the A4 green socket from 'letter' position, cabled and plugged in, and set the driver by PrintManager. I have NS 3.0 installed, the PrintManager should have been updated automatically with 3.0 upgrade (?). Ahti.Kotisaari@helsinki.fi -- 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($UA>6)E('1H:7,@ M:7,@;F]T(&$@8G5G+"!B=70N+BY<"EP*37D@3F585"`T,#!D<&D@3&%S97(@ M4')I;G1E<B!P<FEN=',@86QL(&1O8W5M96YT<R!A<R!F;VQL;W=S.EP*;F5A M<FQY(&5V97)Y('!A9V4@:7,@9FER<W0@<')I;G1E9"!I;F-O;7!L971E;'D@ M*&9R;VT@;VYE('1O('1W;R!T:&ER9',@;V8@<&%G92!L96YG=&@I+"!A;F0@ M=&AE;B!P<F]P97)L>2!A(&-O;7!L971E(&]N92$@5VAY/R`@5&AE(&9O<FUE M<B!O=VYE<B!T;VQD(&ET(&1I9&XG="!B96AA=F4@;&EK92!T:&%T+EP*7`H* M7'!A<F1<='@Q,30P7'1X,C,P,%QT>#,T-#!<='@T-C`P7'1X-3<V,%QT>#8Y M,#!<='@X,#8P7'1X.3(P,%QT>#$P,S8P7'1X,3$U,C!<9F,P7&-F,"!)(&IU M<W0@<F5S970@=&AE($$T(&=R965N('-O8VME="!F<F]M("=L971T97(G('!O M<VET:6]N+"!C86)L960@86YD('!L=6=G960@:6XL(&%N9"!S970@=&AE(&1R M:79E<B!B>2!0<FEN=$UA;F%G97(N("!)(&AA=F4@3E,@,RXP(&EN<W1A;&QE M9"P@=&AE(%!R:6YT36%N86=E<B!S:&]U;&0@:&%V92!B965N('5P9&%T960@ M875T;VUA=&EC86QL>2!W:71H(#,N,"!U<&=R861E("@_*2Y<"EP*7`I!:'1I 9+DMO=&ES86%R:4!H96QS:6YK:2YF:0I]"F=R `
Newsgroups: comp.sys.next.bugs From: rfschtkt%banruc60.bitnet@ohstvma.acs.ohio-state.edu Subject: Re: Using defaultsChanged and public window server Message-ID: <9304051156.AA10810@flexus> Sender: daemon@magnus.acs.ohio-state.edu Organization: The Ohio State University Date: Mon, 5 Apr 1993 11:56:48 GMT LS, Please note that, with my previous message the last I read or wrote about this subject, I realised that Eric Scott, with his use of the word ``it'', was probably talking about what I meant by ``isn't able to let this take effect during a session'' as the reason for the lack of response to defaultsChanged, so I have to revoke what I said as being comments to what he said (I don't think I made any other error). Dumb. Oh well, Eric's message still leaves Loginwindow's "PublicWindowServer" unexplained, right? Raf Schietekat, RfSchtkt@banruc60.bitnet, Flanders, Belgium RfSchtkt@maze.ruca.ac.be apparently isn't functional for public use yet (real, i.e., with triangle in the Deliver button) NeXT Mail preferred I can't reach sites with ! or % in their address, or ending in ``at'' Nederlands, English, Francais, Deutsch a.o.
Newsgroups: comp.sys.next.bugs From: johair@afit.af.mil (John R. O'hair) Subject: Printing Multiple Copies Message-ID: <johair.734408893@afit.af.mil> Sender: news@afit.af.mil Organization: Air Force Institute of Technology Date: Sat, 10 Apr 1993 02:28:13 GMT I encountered a problem while trying to print multiple copies of a document. During the print job, the window server crashes, logging me out. Every time I try to log back in, one page is printed and the server crashes. This is quite a problem when I'm printing 100+ pages. The application I printed from was FrameMaker, but I don't think the problem is there. Does anyone know what is happening and how to fix it? John
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: masud@invnext.worldbank.org (masud cader) Subject: Floppy disk causes panic! Message-ID: <1993Apr10.213543.24952@worldbank.org> Keywords: panic, workspace error, floppy Sender: news@worldbank.org Organization: The World Bank, Washington, DC, USA Date: Sat, 10 Apr 1993 21:35:43 GMT I have experienced a serious error when I inserted a 800K Mac formatted floppy into a NeXTStation running 3.0. I was able to do this four or five times with the same diskette. Apparently the diskette was not formatted properly...but I do not think it should cause a panic. The messages from /usr/adm/messages follow. Apr 10 17:13:11 invnext syslog: Workspace logged in Apr 10 17:13:21 invnext mach: DOS File System: Initialized Apr 10 17:13:23 invnext mach: trap: type 0x14 fcode 0 rw 0 faultaddr 0x0 Apr 10 17:13:23 invnext mach: trap: pc 0x10fec0de sp 0x3ffff5c sr 0x2010 Apr 10 17:13:23 invnext mach: trap: cpu 0 th 0x10166ea0 proc 0x101668e0 pid 171 pcb 0x10ff200c Apr 10 17:13:23 invnext mach: traceback: fp 0x10ff2efc Apr 10 17:13:23 invnext mach: called from pc 10fec27e fp 10ff2f0c 4-args 00000000 10ff0000 10ff2f30 10fee13a Apr 10 17:13:23 invnext mach: called from pc 10fee13a fp 10ff2f30 4-args 1011fbba 00000000 040a3058 10147578 Apr 10 17:13:23 invnext mach: called from pc 0402dbae fp 10ff2f70 4-args 10147578 101159ac 03fffa38 000067a8 Apr 10 17:13:23 invnext mach: called from pc 0409351e fp 10ff2fac 4-args 000067a8 00000004 03fffa38 00000001 Apr 10 17:13:23 invnext mach: called from pc 04001da4 fp 03fffa6c 4-args 000000a7 00006828 000067a8 00000004 Apr 10 17:13:23 invnext mach: last fp 0x3fffa6c Apr 10 17:13:23 invnext mach: panic: (Cpu 0) zero divide Apr 10 17:13:23 invnext mach: NeXT ROM Monitor 2.5 v66 Apr 10 17:13:23 invnext mach: panic: NeXT Mach 3.0: Wed Jul 29 19:43:28 PDT 1992; root(rcbuilder):mk-127.15/BUILD/RELEASE_M68K Apr 10 17:13:23 invnext mach: Apr 10 17:13:26 invnext syslogd: going down on signal 15 Apr 10 17:16:56 invnext mach: Killing all processes NeXT ROM Monitor 2.5 v66 Suggestions? -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Masud Cader Trading Analytics The World Bank masud@invnext.worldbank.org
Newsgroups: comp.sys.next.bugs From: levi@Happy-Man.com (Levi_Bitansky) Subject: Re: Printing Multiple Copies Message-ID: <1993Apr11.025243.24286@Happy-Man.com> Organization: Happy Man Corp., Vashon Island, WA 98070-7399 References: <johair.734408893@afit.af.mil> Date: Sun, 11 Apr 1993 02:52:43 GMT Disclaimer: This is personal opinion and not fact. It is not my employer's opinion, as far as I know; we have not discussed it. In <johair.734408893@afit.af.mil> johair@afit.af.mil (John R. O'hair) writes: >The application I printed from was FrameMaker, but I >don't think the problem is there You are probably mistaken. Frame Technologies seems (based on my interpretation of some co-workers' experiences over some time) to refer bug reports to the equivalent of a public relations department, rather than to real developers. The version Frame put out for Next was so non-NextStep that it had to be revised with minor changes in the operating system. The initial version was so buggy it had to be replaced almost immediately, and in testing their most recent version, they (I'm told; I wasn't involved directly) essentially ignored the testers' comments. Listening to and responding to beta-test reports might have delayed the aggressively pre-scheduled release. Heaven forbid that a small delay should be incurred, in order to produce a reliable, easy-to-use product! Anyway, you are surely not alone in having seen the problem you describe with FrameMaker (and only with FrameMaker). I _have_ seen that with my own eyes. Now that alternative products are available even on NextStep for any but the longest documents, anyone but a large corporation that has standardized on FrameMaker throughout its organization should think twice before purchasing such an ill-supported (and perhaps ill-designed, or it would be easy to support) product, especially when its producer thinks of bug-fixing as a PR function. Perhaps their product is better-supported on platforms with greater unit-sales. I don't know, but my hunch is that that's not the case, and that this company has succeeded only because its only competitor in the large-document market, Interleaf, charges much more, so is out of reach of all but the biggest publishers. Depending on the nature of the specific documents you are publishing, there may be as many as three alternative products, already available on NextStep. So bite the bullet and switch. -- Levi_Bitansky@Happy-Man.com Happy Man Corp. 206/463-9399 x111 4410 SW Pt. Robinson Rd., Vashon Island, WA 98070-7399 fax x118 - Welcome a lost dog into your home; you'll earn much happiness -
Newsgroups: comp.sys.next.bugs From: jjfeiler@relief.com (John Jay Feiler) Subject: Re: Floppy disk causes panic! Message-ID: <1993Apr11.022532.17145@relief.com> Sender: jjfeiler@relief.com Organization: relief consulting References: <1993Apr10.213543.24952@worldbank.org> Date: Sun, 11 Apr 1993 02:25:32 GMT In article <1993Apr10.213543.24952@worldbank.org> masud@invnext.worldbank.org (masud cader) writes: > I have experienced a serious error when I inserted a 800K Mac formatted floppy > into a NeXTStation running 3.0. > > I was able to do this four or five times with the same diskette. Apparently > the diskette was not formatted properly...but I do not think it should cause a > panic. The messages from /usr/adm/messages follow. > > Apr 10 17:13:11 invnext syslog: Workspace logged in > Apr 10 17:13:21 invnext mach: DOS File System: Initialized > Apr 10 17:13:23 invnext mach: trap: type 0x14 fcode 0 rw 0 faultaddr 0x0 > Apr 10 17:13:23 invnext mach: trap: pc 0x10fec0de sp 0x3ffff5c sr 0x2010 > Apr 10 17:13:23 invnext mach: trap: cpu 0 th 0x10166ea0 proc 0x101668e0 pid 171 > pcb 0x10ff200c > Apr 10 17:13:23 invnext mach: traceback: fp 0x10ff2efc > Apr 10 17:13:23 invnext mach: called from pc 10fec27e fp 10ff2f0c 4-args > 00000000 10ff0000 10ff2f30 10fee13a > Apr 10 17:13:23 invnext mach: called from pc 10fee13a fp 10ff2f30 4-args > 1011fbba 00000000 040a3058 10147578 > Apr 10 17:13:23 invnext mach: called from pc 0402dbae fp 10ff2f70 4-args > 10147578 101159ac 03fffa38 000067a8 > Apr 10 17:13:23 invnext mach: called from pc 0409351e fp 10ff2fac 4-args > 000067a8 00000004 03fffa38 00000001 > Apr 10 17:13:23 invnext mach: called from pc 04001da4 fp 03fffa6c 4-args > 000000a7 00006828 000067a8 00000004 > Apr 10 17:13:23 invnext mach: last fp 0x3fffa6c > Apr 10 17:13:23 invnext mach: panic: (Cpu 0) zero divide > Apr 10 17:13:23 invnext mach: NeXT ROM Monitor 2.5 v66 > Apr 10 17:13:23 invnext mach: panic: NeXT Mach 3.0: Wed Jul 29 19:43:28 PDT > 1992; root(rcbuilder):mk-127.15/BUILD/RELEASE_M68K > Apr 10 17:13:23 invnext mach: > Apr 10 17:13:26 invnext syslogd: going down on signal 15 > Apr 10 17:16:56 invnext mach: Killing all processes NeXT ROM Monitor 2.5 v66 > > Suggestions? > -- > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > Masud Cader > Trading Analytics > The World Bank > masud@invnext.worldbank.org You can also panic a NeXT by inserting a Linux boot disk. I've already reported this to bug_next. Perhaps you could do the same. John -- John Feiler jjfeiler@relief.com 4926 152nd St. SW NeXTmail Welcome!!! Edmonds, WA 98026-3344 Independent NeXTSTEP Developer
Newsgroups: comp.sys.next.bugs From: francisr@stupid.ucs.indiana.edu (Rob Francis) Subject: Re: Printing Multiple Copies Message-ID: <C5CAGK.Et@usenet.ucs.indiana.edu> Sender: news@usenet.ucs.indiana.edu (USENET News System) Organization: Indiana University References: <johair.734408893@afit.af.mil> <1993Apr11.025243.24286@Happy-Man.com> Date: Sun, 11 Apr 1993 22:00:19 GMT I agree with the post about FrameMaker. The great thing about Frame is that it's available on all platforms. Moving from a Mac to a PC to a NeXT has always been seamless. But you're gonna want to create the doc on a Mac or PC, because the NeXT version sucks bird. It's very buggy. Did I mention it's VERY buggy? Oft times, it crashes the window server on machines in a lab I maintain. Junk. Considering the interface sucks too, I'd move to something if I can (and will --I'm looking at PasteUp and will check out Pages when it comes out). Of course the interface sucks on all the other platforms too. They get an "A" for keeping things consistent, if not good. -rob --- Rob Francis Unix Systems Specialist, Indiana University francisr@indiana.edu (812)855-0327
Newsgroups: comp.sys.next.bugs From: kelley@kiwi.ATMOS.ColoState.Edu (Kelley Wittmeyer) Subject: Re: Printing Multiple Copies Sender: news@yuma.ACNS.ColoState.EDU (News Account) Message-ID: <Apr12.145434.56335@yuma.ACNS.ColoState.EDU> Date: Mon, 12 Apr 1993 14:54:34 GMT References: <johair.734408893@afit.af.mil> Organization: Colorado State University, Fort Collins, CO 80523 In article <johair.734408893@afit.af.mil> johair@afit.af.mil (John R. O'hair) writes: > I encountered a problem while trying to print multiple > copies of a document. During the print job, the window > server crashes, logging me out. Every time I try to > log back in, one page is printed and the server crashes. > This is quite a problem when I'm printing 100+ pages. > The application I printed from was FrameMaker, but I > don't think the problem is there. Does anyone know > what is happening and how to fix it? > > John i have a problem printing multiple copies but it happens on just one printer, the others in our group work ok. more often than not, i do not get all of my copies printed. no error messages, just ~10% less copies than i requested. this happens from Edit.app. any clues? kelley wittmeyer dept of atmospheric science colorado state university
Newsgroups: comp.sys.next.bugs From: alex@hal.rhein-main.de (Alexander Lehmann) Subject: Re: Floppy disk causes panic! Message-ID: <C5Do8J.8w@hal.rhein-main.de> References: <1993Apr10.213543.24952@worldbank.org> Date: Mon, 12 Apr 1993 15:55:30 GMT masud cader (masud@invnext.worldbank.org) wrote: : I have experienced a serious error when I inserted a 800K Mac formatted : floppy into a NeXTStation running 3.0. : I was able to do this four or five times with the same diskette. Apparently : the diskette was not formatted properly...but I do not think it should cause : a panic. [...] : Suggestions? I just tested this with a disk where I wrote 00 over the format information in the boot sector (at least one use for my old ST, a disk monitor). The problem is caused by /usr/filesystems/DOS.fs/dosfs_reloc, which is dynamically linked into the kernel when a DOS disk is inserted. The code that reads the boot sector doesn't check the data for validity (only whether f9 or fb is the first byte of the sector) and in 2 or 3 functions divides by the number of sector per cluster. If this number is 0 due to a corrupted disk, a division by zero exception causes the kernal panic. I have checked a bit into the code, the bug shouldn't the hard to fix. I'll post a fix as soon as I get it figured out. (Debugging is a bit difficult when any program error causes fsck to run for a couple of minutes). bye ... Alexander Lehmann -- Alexander Lehmann, alex@hal.rhein-main.de | "Wopp" or alexlehm@iti.informatik.th-darmstadt.de | - ( THHGTTG Pt.3 )
Newsgroups: comp.sys.next.bugs From: alex@hal.rhein-main.de (Alexander Lehmann) Subject: Wrong time with DOS filesystem + fix Message-ID: <C5Dr5x.Dv@hal.rhein-main.de> Date: Mon, 12 Apr 1993 16:58:43 GMT First of all, if you live on the westcoast and doen't care about DST, you can ignore this. I ran across a problem with the MSDOS filesystem in NS3.0. All files that are copied from a MSDOS disk onto the NeXT filesystems have a datestamp that is 9 hours newer that when read on a real (oh boy) MSDOS system. I checked into this problem an found out that the time are listed correctly, if PST is the current timezone. This could be considered a problem you have to live with rather than a bug, since the DOS driver is linked into the kernel and struct tm doesn't have a field for the timezone. On the other hand, it rather annoying to have the files suddenly beamed into the future. So I disassembled the dosfs_reloc program and found the functions that convert the MSDOS style times to UNIX times and vice versa. The problem is very simple to fix, since there only two addl # statements that add resp. subtract the second offset from GMT. I wrote a small perl program to do this automatically. (at the end of this post) If you want to apply this patch, first of all create a backup of /usr/filesystems/DOS.fs, since you might break something. (I don't think so, but who knows) You will have to change the $localtime variable at the start of the program to your timezone. e.g. GMT 0 PDT DST -7*3600 MET DST +2*3600 (that's where I live) then just execute 'patchmsd.pl' (this will not change anything, even if you are root). this will print the current time offset and complain, if it wants to change it. To actually change the offset, you have to become superuser and execute 'patchmsd.pl -p' If you have inserted a DOS disk since the last boot, you will have to reboot for the changes to become active (I think that dynamically loaded kernel programs are copied into memory and not paged readonly). After this procedure, DOS filetimes should be correct. To make life easy, the program will change the what string in the file to identify the timezone the driver is patched, this will look something like this PROGRAM:dosfs_reloc PROJECT:dos-28.0.1 DEVELOPER:mgillon PATCHED: to GMT+2 Once again, I'd like to stress that it is important to create a backup copy of the filesystem driver and, of course, you should do this when nobody does any real work on the machine, since a kernel panic might occur, if something goes wrong. (It didn't happen with this fix, but you have to expect problems when dealing with the intestines of a system, see my other post from today) And, of course, you're on your own, I feel not responsible when you screw up your box. O.K. here come the program: ------------------------------------------------------------------------------- #!/bin/sh # This is a shell archive (produced by shar 3.49) # To extract the files from this archive, save it to a file, remove # everything above the "!/bin/sh" line above, and type "sh file_name". # # made 04/12/1993 16:53 UTC by alex@hal # Source directory /alex/DOS.fs # # existing files will NOT be overwritten unless -c is specified # # This shar contains: # length mode name # ------ ---------- ------------------------------------------ # 2243 -rwxr-xr-x patchmsd.pl # # ============= patchmsd.pl ============== if test -f 'patchmsd.pl' -a X"$1" != X"-c"; then echo 'x - skipping patchmsd.pl (File already exists)' else echo 'x - extracting patchmsd.pl (Text)' sed 's/^X//' << 'SHAR_EOF' > 'patchmsd.pl' && #! /usr/local/bin/perl # # patchmsd.pl: patch the time offset for MSDOS disks # # $Id: patchmsd.pl,v 1.6 1993/04/12 16:35:33 alex Exp $ # # written by Alexander Lehmann <alex@hal.rhein-main.de> # # use this at your own risk # X $offset1=0x1e00+988; $offset2=0x1b58+988; $whatstr=0x6f94; X $localtime=+2*3600; # CET DST (change this to suit your needs) X #$fsfile="/usr/filesystems/DOS.fs/dosfs_reloc"; $fsfile="dosfs_reloc"; X open(FILE,$fsfile) || die "Can't read $fsfile\n"; X seek(FILE,$offset1,0) || die "Can't seek to filepos\n"; read(FILE,$inp,6)==6 || die "Can't read data\n"; ($c,$v1)=unpack("Sl",$inp); X $c==0x0680 || die "not an addl # command\n"; X seek(FILE,$offset2,0) || die "Can't seek to filepos\n"; read(FILE,$inp,6)==6 || die "Can't read data\n"; ($c,$v2)=unpack("Sl",$inp); close FILE; X $c==0x0680 || die "not an addl # command\n"; X $v1==-$v2 || die "positive and negative time offsets don't aggree\n"; X print "time offset is set to GMT"; if($v2<0) { X print $v2/3600; } elsif($v2>0) { X print "+",$v2/3600; } X print "\n"; X if($v2==$localtime) { X print "MSDOS timeoffset agrees with local time\n"; X exit 0; } X if( int(@ARGV)!=1 || $ARGV[0] ne "-p" ) { X print "Use patchmsd.pl -p to patch the time offset\n"; X exit 0; } X open(FILE,"+>>$fsfile") || die "Can't write $fsfile\n"; X seek(FILE,$offset1+2,0) || die "Can't seek to filepos\n"; syswrite(FILE,pack("l",-$localtime),4) || die "Can't write data\n"; X seek(FILE,$offset2+2,0) || die "Can't seek to filepos\n"; syswrite(FILE,pack("l",$localtime),4) || die "Can't write data\n"; X seek(FILE,$whatstr,0) || die "Can't seek to filepos\n"; read(FILE,$str,1024)>0 || die "Can't read data\n"; X if(substr($str,0,4) ne "@(#)") { X die "Can't change what string\n"; } X if(($null=index($str,"\0"))!=$[) { X $str=substr($str,0,$null); } X if(($start=index($str,"BUILT"))!=$[-1 || ($start=index($str,"PATCHED"))!=$[-1) { X $len=length($str)-$built; X $pat="PATCHED: to GMT"; X if($localtime<0) { X $pat .= sprintf ("%d", $localtime/3600); X } elsif($localtime>0) { X $pat .= sprintf ("+%d", $localtime/3600); X } X X substr($str,$start,$len)="$pat\0"; X X seek(FILE,$whatstr,0) || die "Can't seek to filepos\n"; X print FILE $str; } X close FILE; X print "file successfully patched\n"; X SHAR_EOF chmod 0755 patchmsd.pl || echo 'restore of patchmsd.pl failed' Wc_c="`wc -c < 'patchmsd.pl'`" test 2243 -eq "$Wc_c" || echo 'patchmsd.pl: original size 2243, current size' "$Wc_c" fi exit 0 ------------------------------------------------------------------------------- You are free to distribute this program as you wish, but please leave my name in. One last thing, if someone can suggest a simple method for finding the current timezone in perl, I can change the program to automatically adjust to daylight saving time and other things, but the environment TZ is not always present and I didn't find any way to do it in perl (One could call localtime and gmtime and compare the results, but this seems rather complicated). That's about it, bye ... Alexander Lehmann -- Alexander Lehmann, alex@hal.rhein-main.de | "Wopp" or alexlehm@iti.informatik.th-darmstadt.de | - ( THHGTTG Pt.3 )
From: andrewd@sematech.tamu.edu (Andrew Duchowski) Newsgroups: comp.sys.next.bugs Subject: Formatting DOS floppies ?? Date: 13 Apr 1993 16:18:26 GMT Organization: Sematech Research, Texas A&M University Distribution: world Message-ID: <1qep4iINNecu@tamsun.tamu.edu> I apologize if this has already been discussed. If so, can someone point me to a FAQ or something similar? Otherwise: I've never been able to successfully format a DOS diskette on my NeXT (currently running 3.0). I just did this recently and copied some text files over to the diskette using mtools. The files show up on a 'dir' on a DOS system, but I can't 'cat' any of them. Furthermore, I can't re-format the disk since DOS says that it has a bad 0 sector. Suggestions? Thanks in advance! -- -- Andrew Duchowski | -- -- Texas A&M University | This page intentionally blank -- -- andrewd@cs.tamu.edu (non-NeXT mail) | -- -- andrewd@visual2.cs.tamu.edu (NeXT mail) | --
From: ts110@cus.cam.ac.uk (T. Slivnik) Newsgroups: comp.sys.next.bugs Subject: Swollen swapfile Message-ID: <1993Apr13.222636.15730@infodev.cam.ac.uk> Date: 13 Apr 93 22:26:36 GMT Sender: news@infodev.cam.ac.uk (USENET news) Organization: U of Cambridge, England My NeXT keeps running out of disk space, even though there is at least 30M available on my disk. I often need to reboot, after which SOME of the disk space magically reappears. E.g. Before reboot ============= localhost> ls -l /private/vm total 104856 -rw------t 1 root 43048960 Apr 12 15:42 swapfile -rw------t 1 root 64290816 Apr 12 15:42 swapfile.front (Workspace Manager shows ~5M free) After reboot ============ localhost> ls -l /private/vm total 18320 -rw------t 1 root 16777216 Apr 12 15:58 swapfile -rw------t 1 root 1974272 Apr 12 15:58 swapfile.front (Workspace Manager shows 30.1M free) Why does my swapfile keep growing so drastically in size? Why do I only get 25M back when it seems swapfile loses 27M and swapfile.front loses 44M after reboot? Is there any way I could keep these files smaller? I would be extremely grateful to anyone who can offer me some advice. Tomaz Slivnik
Newsgroups: comp.sys.next.bugs From: lewine@cheshirecat.webo.dg.com (Donald Lewine) Subject: Logout "ejects" external hard disks Sender: usenet@webo.dg.com (Usenet Administration) Message-ID: <1993Apr13.150137.8745@webo.dg.com> Date: Tue, 13 Apr 93 15:01:37 GMT Organization: Data General Corporation I have an external hard drive attached to my NeXT color slab and exported for NFS use. When I logoff my NeXT the hard drive is unmounted and other users cannot access it. I think that NeXTStep must think it is some kind of floppy and should be ejected on logout. I also think the behavior is a bug in NeXTStep. Has anyone else ever seen it? Did I set something up wrong? -- -------------------------------------------------------------------- Donald A. Lewine (508) 870-9008 Voice Data General Corporation (508) 366-0750 FAX 4400 Computer Drive. MS D112A Westboro, MA 01580 U.S.A. uucp: uunet!dg!lewine Internet: lewine@cheshirecat.webo.dg.com
Newsgroups: comp.sys.next.bugs From: garton@cunixa.cc.columbia.edu (Bradford Garton) Subject: Re: Logout "ejects" external hard disks Message-ID: <1993Apr14.141153.24934@news.columbia.edu> Sender: usenet@news.columbia.edu (The Network News) Organization: Columbia University References: <1993Apr13.150137.8745@webo.dg.com> Date: Wed, 14 Apr 1993 14:11:53 GMT In article <1993Apr13.150137.8745@webo.dg.com> lewine@cheshirecat.webo.dg.com writes: >I have an external hard drive attached to my NeXT color slab and >exported for NFS use. When I logoff my NeXT the hard drive is >unmounted and other users cannot access it. > >I think that NeXTStep must think it is some kind of floppy and >should be ejected on logout. I also think the behavior is a bug >in NeXTStep. Has anyone else ever seen it? Did I set something >up wrong? I have a related problem: I temporarily attached a SCSI hard disk to one of our slabs, and now NeXTStep tells users who logon that it cannot read the SCSI hard disk and generates some minor errors. For the life of me I cannot find where the "phantom" disk is being referenced, I've checked all the *tab files, NetInfo, all of the users' defaults for possible shelf-stashing, I even tried rebuilding the device entries for *sd1a* in /dev (rebooting obviously didn't work...). Thinking that perhaps the disk simply had not been umounted correctly, I attached a SCSI disk again, umounted it by hand, and now I get *two* errors generated whenever someone logs on. Is this a known problem, and I've missed some discussion about it? Is it fixable, or do I have to live with "Cannot read SCSI disk, reinitialize or ignore..." forever? Brad Garton Music brad@woof.music.columbia.edu
Newsgroups: comp.sys.next.bugs From: peter@beast.math.ualberta.ca (Peter Karbaliotis) Subject: Re: Bug in Newsgrazer version 72.3? Message-ID: <1993Apr14.152831.1676@kakwa.ucs.ualberta.ca> Sender: news@kakwa.ucs.ualberta.ca Organization: University Of Alberta, Edmonton Canada References: <1pscg6$hoi@menudo.uh.edu> Date: Wed, 14 Apr 1993 15:28:31 GMT In article <1pscg6$hoi@menudo.uh.edu> sears@tree.egr.uh.edu (Paul S. Sears) writes: > >Newsgrazer Release 2.0, version 72.3 > >We are having problems with NewsGrazer. It suddenly quits at random times. >If run from the shell window, it returns: > >"buss error" > >so I decided to run it via gdb. It seems to crash consistantly with a >memory exception error (same address). Anyone else have this problem?? > >My gdb sessom is below: > [munch] >(gdb) info stack >#0 0x6060000 in DPSAddFD () >#1 0x144ce in ?? () >#2 0x5df6 in ?? () >#3 0x59a8 in ?? () >#4 0x5424 in ?? () >#5 0x54ea in ?? () >#6 0x6020cd0 in checkTEs () >#7 0x601c344 in _DPSGetOrPeekEvent () >#8 0x601d726 in NXGetOrPeekEvent () >#9 0x602e47e in -[Application run] () >#10 0x11706 in ?? () > I had exactly the same problem with NG 72.3 last week (Monday April 5, in fact). That is, it crapped out in DPSAddFD called from ??, ..., called from checkTEs(). After I looked at the new groups, the problem went away and has not recurred. Hmm... -- Peter Karbaliotis - peter@beast.math.ualberta.ca - NeXTmail ok When a particle you don't even know gives you momentum, that's impulse.
Newsgroups: comp.sys.next.bugs From: sxurcb@fnma.com(Rodney Blake) Subject: Unsure how to debug these problems Message-ID: <1993Apr14.215908.952@almserv.uucp> Sender: usenet@almserv.uucp Organization: Fannie Mae Date: Wed, 14 Apr 1993 21:59:08 GMT I have a couple of messages being sent to my console window. They do not seem to blow-up my program, but I am curious on what might be the cause(s) of these messages and where do I begin to debug these problems. In gdb or some other way ? First Message: Apr 14 17:42:14 thalia CodeManager[959]: dpsclient assertion failed: an fd passed to DPSAddFD was closed before calling DPSRemoveFD Second Message: Apr 14 17:43:17 thalia last message repeated 27 times Please be kind, Just a beginner NeXT programmer ! ! -- Rodney Blake [sxurcb@fnma.com (NeXT Mail Okay)] Mail Stop: 4H-5S:02 Federal National Mortgage Association (Fannie Mae) 3900 Wisconsin Ave., NW Washington, DC 20016 phone: (202) 752-3705
Newsgroups: comp.sys.next.bugs From: kramer@fragile.uucp (Mike Andrews) Subject: Re: Logout "ejects" external hard disks Message-ID: <C5HxC8.2LI@fragile.uucp> Sender: kramer@fragile.uucp (Mike Andrews) Organization: Terminal Frost References: <1993Apr14.141153.24934@news.columbia.edu> Date: Wed, 14 Apr 1993 23:02:31 GMT In article <1993Apr14.141153.24934@news.columbia.edu> garton@cunixa.cc.columbia.edu (Bradford Garton) writes: > In article <1993Apr13.150137.8745@webo.dg.com> lewine@cheshirecat.webo.dg.com writes: > >I have an external hard drive attached to my NeXT color slab and > >exported for NFS use. When I logoff my NeXT the hard drive is > >unmounted and other users cannot access it. > > > >I think that NeXTStep must think it is some kind of floppy and > >should be ejected on logout. I also think the behavior is a bug > >in NeXTStep. Has anyone else ever seen it? Did I set something > >up wrong? Put the drive in /etc/fstab -- then the machine will mount the disk at bootup, and you won't have any more problems. This is in the network/sysadmin docs. > I have a related problem: I temporarily attached a SCSI hard disk to one of > our slabs, and now NeXTStep tells users who logon that it cannot read the > SCSI hard disk and generates some minor errors. For the life of me I cannot > find where the "phantom" disk is being referenced, I've checked all the *tab > files, NetInfo, all of the users' defaults for possible shelf-stashing, I > even tried rebuilding the device entries for *sd1a* in /dev (rebooting > obviously didn't work...). Thinking that perhaps the disk simply had not > been umounted correctly, I attached a SCSI disk again, umounted it by hand, > and now I get *two* errors generated whenever someone logs on. Hmmm, don't know about this one, unless you also didn't put it in /etc/fstab... If a drive isn't in there, it only gets mounted when someone logs in on the console, apparently, and it doesn't do file permissions right. My /etc/fstab: /dev/sd0a / 4.3 rw,noquota,noauto 0 1 /dev/sd1a /fragile2 4.3 rw,noquota 0 2 -- Mike Andrews root@fragile.uucp [NeXTmail OK] kramer@wittenberg.edu (school) "This guy's pretty bizarre, Gus." kramer@mik.uky.edu (hometown) -- Primus
Newsgroups: comp.sys.next.bugs From: coconut@crash.cts.com (Brian Dear) Subject: Mail bug? Organization: coconut computing, inc., la jolla, california Date: 15 Apr 93 20:00:29 PDT Message-ID: <1993Apr15.200029.2460@crash> This has undoubtedly been reported already. Perhaps there's a fix? Oftentimes I will be composing a new message in Mail.app, and when I click the "Deliver" button, all my Mail.app windows go away, and the program exits. The message goes through, but jeepers, I kinda wanted to still do some more stuff in Mail.app. This is in NS3.0 on a NeXTcube with 24MB RAM. Any ideas? -- brian dear coconut computing, inc. brian@coconut.com
From: cjs@po.CWRU.Edu (Christopher J. Seline) Newsgroups: comp.sys.next.bugs Subject: Why doesn't NeXT issue patches (like most other computer companies) Date: 17 Apr 1993 23:19:54 GMT Organization: Case Western Reserve University, Cleveland, OH (USA) Message-ID: <1qq3aq$4sq@usenet.INS.CWRU.Edu> I'd like to suggest that NeXT issue patches anytime they fix a bug. DEC, IBM, and the rest do this, why can't NeXT? cjs p.s. please no flames
From: madler@cco.caltech.edu (Mark Adler) Newsgroups: comp.sys.next.bugs Subject: Re: Why doesn't NeXT issue patches (like most other computer companies) Date: 18 Apr 1993 15:51:31 GMT Organization: California Institute of Technology, Pasadena Message-ID: <1qrte3INNr15@gap.caltech.edu> References: <1qq3aq$4sq@usenet.INS.CWRU.Edu> >> I'd like to suggest that NeXT issue patches anytime they fix a bug. They did once. They issued (though only for those with ftp, I think) a patch for the 3.0 disk eject bug. I was ecstatic when they did that, thinking that they finally reversed their policy to not issue bug fixes--only new releases. However, they have since released none. I was hoping that they'd release new CD-ROM drivers to fix the several bugs there, but nothing. (By the way, they do on occasion issue patches to close discovered security holes, but I don't count those, since NeXT obviously treats them specially.) So, anyway, I echo the plea. Yes, I know they're busy, but I also know that they've fixed most or all of the ISO 9660 CDROM driver bugs, so it couldn't take more than 15 minutes for one person to put together a package and place on the next.com ftp site. It would go a long long way toward improving their image as a customer responsive company. Mark Adler madler@cco.caltech.edu
Newsgroups: comp.sys.next.bugs From: gerben@rna.indiv.nluug.nl Subject: Let's do it: NeXT should deliver patches and bugfixes! Message-ID: <1993Apr18.173439.1025@rna.indiv.nluug.nl> Sender: gerben@rna.indiv.nluug.nl (Gerben Wierda) Organization: G.R.O.S.S. References: <1qrte3INNr15@gap.caltech.edu> Date: Sun, 18 Apr 1993 17:34:39 GMT In article <1qrte3INNr15@gap.caltech.edu> madler@cco.caltech.edu (Mark Adler) writes: > So, anyway, I echo the plea. Yes, I know they're busy, but I also > know that they've fixed most or all of the ISO 9660 CDROM driver > bugs, so it couldn't take more than 15 minutes for one person to > put together a package and place on the next.com ftp site. It would > go a long long way toward improving their image as a customer > responsive company. I second that plea. Let's create a message for NeXT management that is signed by e-mail signatures. The short message could read: Dear NeXT, We all think that issuing patches and bugfixes is very good for both your image as a user-friendly company and for the support we can get. Creating and submitting those patches doesn't cost much either. Please let us know if you would follow our advise and make patches and fixes available. Signed, I'll volunteer to collect signatures. Send me a mail message with the subject of this mail message (just reply). I'll wait for 1 or 2 weeks and send the message off. -- Gerben Wierda [NeRD:7539] Tel. (+31) 35 833539 "If you don't know where you are going, any road will take you there." From the Talmud(?), rephrased in Lewis Carroll, "Alice in Wonderland".
Newsgroups: comp.sys.next.bugs,comp.sys.next.software From: Benoit Grange <Benoit_Grange@gateway.qm.apple.com> Subject: Re: compiler problem Sender: news@gallant.apple.com Message-ID: <1993Apr21.090936.25598@gallant.apple.com> Date: Wed, 21 Apr 1993 09:09:36 GMT References: <hheller.735327541@Physik.TU-Muenchen.DE> Organization: Apple Europe In article <hheller.735327541@Physik.TU-Muenchen.DE> Helmut Heller, hheller@Physik.TU-Muenchen.DE writes: In article <hheller.735327541@Physik.TU-Muenchen.DE> Helmut Heller, hheller@Physik.TU-Muenchen.DE writes: > Hello, > I recently tried to compile MH6.8 (POP mail client) on my NeXT, but I got > errors like the following: > > /NextDeveloper/Headers/ansi/ctype.h:31: parse error before `char' (etc.) Many errors like this are caused by syntax errors in the c source before including the standard headers. Check the files at the beginning (first few lines). An other problem may be that the includes of MH define some weird things clashing with the includes (unterminated structures, etc.). Benoit.
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: taluskie@utpapa.ph.utexas.edu (Vince Taluskie) Subject: sendmail: bus error ?! Message-ID: <1993Apr21.222732.21280@chpc.utexas.edu> Sender: root@chpc.utexas.edu (Convex UNIX) Organization: Physics Department, University of Texas at Austin Date: Wed, 21 Apr 93 22:27:32 GMT Hi, I'm not too familiar with Nexts but I was asked to try to help out a group in our department with a problem they were having with their machine. sendmail isn't starting out of the boot scripts. When I try to start sendmail by hand: /usr/lib/sendmail -bd -q1h I get the message "Bus error" BTW, /usr/lib/sendmail.cf is a symbolic link to sendmail.subsidiary.cf . Any idea what I can do about this, please email if you have any advice you can offer. Thanks in Advance, Vince Taluskie -- Vince Taluskie UNIX Systems Administration Physics Computer Group taluskie@utpapa.ph.utexas.edu The University of Texas at Austin
Newsgroups: comp.sys.next.bugs From: orange@alfred.carleton.ca (Paul Vallee) Subject: Workspace crashes when Help requested Message-ID: <orange.735432464@cunews> Sender: news@cunews.carleton.ca (News Administrator) Organization: Carleton University Date: Wed, 21 Apr 1993 22:47:44 GMT Ok, here's the situation: Whenever help for Workspace is requested, workspace crashes, and I get the 'Workspace Manager Error, try to save your changes before logging out.' panel. This happens regardless of whether I [Info][Help] or whether I use the {Help} key. Unfortunately, I have no idea when the problem occurred. I don't use help in the workspace very often. I have tested the help application in other programs (Mesa 1.4 demo) and it works fine. Any ideas what might have gone wrong? I have been root on this machine regularly for setup and program application purposes, so I am open to the idea that I screwed something up. Thanks for your help, orange@ccs.carleton.ca ------- Paul Vallee ---- 542429@acadvm1.uottawa.ca "It is preferable to incur a mild punishment than to perform an onerous task." --- Roald Dahl ------------Home: (613)234-6625 --------- Work: (613)996-3263------------
Newsgroups: comp.sys.next.bugs From: kira!davidjohn (David John Burrowes) Subject: Re: Why doesn't NeXT issue patches (like most other computer companies) Message-ID: <1993Apr23.035228.383@kira.net.netcom.com> Sender: davidjohn@kira.net.netcom.com Organization: No organization at this time. References: <1qrte3INNr15@gap.caltech.edu> Date: Fri, 23 Apr 1993 03:52:28 GMT In article <1qrte3INNr15@gap.caltech.edu> madler@cco.caltech.edu (Mark Adler) writes: > > >> I'd like to suggest that NeXT issue patches anytime they fix a bug. > > So, anyway, I echo the plea. Yes, I know they're busy, but I also > know that they've fixed most or all of the ISO 9660 CDROM driver > bugs, so it couldn't take more than 15 minutes for one person to > put together a package and place on the next.com ftp site. It would > go a long long way toward improving their image as a customer > responsive company. > > Mark Adler > madler@cco.caltech.edu I suppose I'd agree that patches would be a good thing. But I'd disagree about that `15' minutes statistic. Someone will have to create the patch, and the documentaton will have to be written and checked. Then the whole will have to go through whatever Quality Assurance process they have, making sure that it in fact installs properly, the doc makes sense, it functions at a production level adequately. Presumably it'll have to be blessed by some legal OK, and then it can be posted. If they do, in fact, spend a small ammount of time on it, then I'm not sure I'd want to get it anywhere near my machine. So, I guess I just wanted to point out that it is probably by no means something that can be done properly during someone's lunch break, and given their Intel priorities, might explain a lack of attention given to patches. \david john burrowes
From: madler@cco.caltech.edu (Mark Adler) Newsgroups: comp.sys.next.bugs Subject: Re: Why doesn't NeXT issue patches (like most other computer companies) Date: 23 Apr 1993 11:34:16 GMT Organization: California Institute of Technology, Pasadena Message-ID: <1r8k7oINN8lh@gap.caltech.edu> References: <1qrte3INNr15@gap.caltech.edu> <1993Apr23.035228.383@kira.net.netcom.com> >> I suppose I'd agree that patches would be a good thing. But I'd disagree >> about that `15' minutes statistic. ... etc. about documenting, quality assurance, etc. If it were a patch to the kernel, I might agree. However, the patch I'm talking about (fixing the CDROM driver bugs) involves simply replacing one file, so assembling the package and writing the documentation was included in my 15 minute estimate. As for quality assurance, I don't care if the patch doesn't work! At least the first time. I'll be more than happy to test it for them (as will others--we'll be testing different bugs, since there are more than one in the cdrom drivers). If it doesn't work, because they didn't actually fix the bug, then they'd want to know that before the next release. If it bothers someone to test patches like this, then the documentation can clearly state "UNTESTED PATCH" or somesuch. I'd much prefer untested patches than no patches at all. Mark Adler madler@cco.caltech.edu
Newsgroups: comp.sys.next.bugs From: gcolello@biosphere.Stanford.EDU (Greg Colello) Subject: Re: Why doesn't NeXT issue patches (like most other computer companies) Message-ID: <1993Apr23.175428.21866@leland.Stanford.EDU> Sender: news@leland.Stanford.EDU (Mr News) Organization: DSO, Stanford University References: <1r8k7oINN8lh@gap.caltech.edu> Date: Fri, 23 Apr 93 17:54:28 GMT In article <1r8k7oINN8lh@gap.caltech.edu> madler@cco.caltech.edu (Mark Adler) writes: > >> I suppose I'd agree that patches would be a good thing. But I'd >> disagree about that `15' minutes statistic. >> (.... etc. about documenting, quality assurance, etc.) >> -\david john burrowes > > If it were a patch to the kernel, I might agree. However, the patch > I'm talking about (fixing the CDROM driver bugs) involves simply > replacing one file, so assembling the package and writing the > documentation was included in my 15 minute estimate. > > As for quality assurance, I don't care if the patch doesn't work! > At least the first time. I'll be more than happy to test it for them > (as will others--we'll be testing different bugs, since there are > more than one in the cdrom drivers). If it doesn't work, because > they didn't actually fix the bug, then they'd want to know that > before the next release. > > If it bothers someone to test patches like this, then the documentation > can clearly state "UNTESTED PATCH" or somesuch. I'd much prefer > untested patches than no patches at all. > > Mark Adler > madler@cco.caltech.edu Having worked in two different computer companies in the QA division, I can vouge for David's statements. Patches are a bummer for a computer company. Frequently by the time you get the patch out the next release is ready anyway. Another major headache with patches that was not mentioned by David is putting the patch on some media for distribution and distributing it. On the other hand the net removes the media and distribution problems. Removing the limitation of the patch going through QA as suggested above is another major obstacle removed. The question is whether NeXT wants to face the problems associated with releasing untested patches to the public (even with a disclaimer). If it were me at NeXT, I would say, "Well let's give it a try". Anybody listening? I'll say one thing that DiskEjectFix.daemon "patch" was a great success at our site. Thanks NeXT! ----------------------------------------------------------------- Greg Colello Carnegie Institution, Department of Plant Biology Stanford University gcolello@biosphere.stanford.edu (NeXT mail OK)
Newsgroups: comp.sys.next.bugs From: nathan@laplace.csb.yale.edu (Nathan F. Janette) Subject: Re: Why doesn't NeXT issue patches (like most other computer companies) Message-ID: <1993Apr23.203339.24611@cs.yale.edu> Sender: news@cs.yale.edu (Usenet News) Organization: Yale University, Department of Computer Science, New Haven, CT References: <1993Apr23.175428.21866@leland.Stanford.EDU> Date: Fri, 23 Apr 1993 20:33:39 GMT In article <1993Apr23.175428.21866@leland.Stanford.EDU> gcolello@biosphere.Stanford.EDU (Greg Colello) writes: > Having worked in two different computer companies in the QA division, I > can vouge for David's statements. Patches are a bummer for a computer > company. Frequently by the time you get the patch out the next release is > ready anyway. Another major headache with patches that was not mentioned > by David is putting the patch on some media for distribution and > distributing it. Wasn't one the arguments used to make NeXT owners shell out for the CDROM drive that CDROMs were cheap and easy to use for such distributions? Somewhere between NeXT's one-and-only-one major release per year and joining the patch-of-the-week-club there must be a happy medium. Why not release a 3.1.n release every 3 months on CDROM, containing bug fixes *only*, not the newest layered product. > I'll say one thing that DiskEjectFix.daemon "patch" was a great success at > our site. Thanks NeXT! Damn straight, the 3.1.n releases should be ftp'able as well. -- Nathan Janette # "As I walk I hear my longing thoughts subsiding. Dept MB&B # Upon your cross I bleed the thoughts that I've been hiding. Yale Univ/HHMI # I'm all used up; there's not much more for me to give. New Haven, CT # Echoes of the life that we all want to live." nathan@laplace.csb.yale.edu
Newsgroups: comp.sys.next.bugs,comp.sys.next.software From: alex@hal.rhein-main.de (Alexander Lehmann) Subject: Re: compiler problem Message-ID: <C5wn5z.vq@hal.rhein-main.de> Followup-To: comp.sys.next.bugs,comp.sys.next.software References: <hheller.735327541@Physik.TU-Muenchen.DE> Date: Thu, 22 Apr 1993 21:46:46 GMT Helmut Heller (hheller@Physik.TU-Muenchen.DE) wrote: : Hello, : I recently tried to compile MH6.8 (POP mail client) on my NeXT, but I got errors like the following: : /NextDeveloper/Headers/ansi/ctype.h:31: parse error before `char' : /NextDeveloper/Headers/ansi/ctype.h:31: syntax error before `char' : /NextDeveloper/Headers/ansi/machine/setjmp.h:11: #include expects "fname" or <fname> : /NextDeveloper/Headers/bsd/machine/label_t.h:11: #include expects "fname" or <fname> : /NextDeveloper/Headers/bsd/machine/machparam.h:11: #include expects "fname" or <fname> : /NextDeveloper/Headers/bsd/machine/signal.h:11: #include expects "fname" or <fname> : I had used the flags -traditional -bsd for the compiler. : It looks to me as if there is something wrong with the way the machine specific header files are included : --- or did I just not specify the correct compiler options? : Are there any workarounds? The problem comes from the use of -traditional. The NeXT include files rely on the feature of cpp that it can handle defines in include lines. This is turned of by the -traditional option. Something like #include ARCH_INCLUDE(ansi/, setjmp.h) is not allowed in trad C. The second problem is that the NeXT include files use const and voilatile in function prototypes and variable declarations. This can be fixed by adding -Dconst= or -Dvoilatile= to the compiler flags. The include problem can only be fixed by not using -traditional. If the MH sources really rely on traditional features like xxx/**/yyy token catenation or writable strings, I think that should be changed. BTW. I came across another bug in cpp. Try the following: #define XXX "stdio.h" #include XXX This makes cpp stop with signal 10 (bus error). bye ... Alexander Lehmann -- Alexander Lehmann, alex@hal.rhein-main.de | "Wopp" or alexlehm@iti.informatik.th-darmstadt.de | - ( THHGTTG Pt.3 )
Control: cancel <1993Apr13.222636.15730@infodev.cam.ac.uk> Newsgroups: comp.sys.next.bugs From: ts110@cus.cam.ac.uk (T. Slivnik) Subject: cmsg cancel <1993Apr13.222636.15730@infodev.cam.ac.uk> Message-ID: <1993Apr24.160115.26055@infodev.cam.ac.uk> Originator: ts110@bootes.cus.cam.ac.uk Sender: news@infodev.cam.ac.uk (USENET news) Organization: U of Cambridge, England References: <1993Apr13.222636.15730@infodev.cam.ac.uk> Date: Sat, 24 Apr 1993 16:01:15 GMT <1993Apr13.222636.15730@infodev.cam.ac.uk> was cancelled from within rn.
Newsgroups: comp.sys.next.bugs,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.sysadmin From: kgnome@cs.concordia.ca (MATIS stephane) Subject: Smalltalk building under NeXTSTEP 3.0 Message-ID: <C62Avt.AEp@newsflash.concordia.ca> Summary: GNU Smalltalk 1.1.1 Sender: usenet@newsflash.concordia.ca (USENET News System) Organization: Computer Science, Concordia University, Montreal, Quebec Date: Sun, 25 Apr 1993 23:07:05 GMT The following is the answer that I have received on my difficulty in "making" GNU Smallatlk 1.1.1 under NEXTSTEP 3.0. I have yet to try the posted "success", but I'm sure it will work. [Include Message]------------------------------------------------------- Date: Sat, 24 Apr 93 15:24:58 -0500 From: Chris Cleeland <milo!chris@wpa.com> To: MATIS stephane <kgnome@cs.concordia.ca> Subject: Re: GNU Smalltalk 1.1.1 Binaries sought Reply-To: chris%milo@wpa.com In comp.sys.next.programmer article <C5v5uy.HM1@newsflash.concordia.ca> you wrote: > Hello Folks! > > I'm looking for a set of GNU Smalltalk Binaries that run under 3.0, > but certainlly weren't compiled under 3.0, as it seems to be impossible. My > two attemptes failed, and according to gnu.smalltalk.bug, that's the general > gist of it. > > So please, could a good samaritan upload to cs.orst.edu or sonnata > a set of binaries for Smalltalk. I'm learning the language through the > Smalltalk tuorial, and have been very taken by it. I like the ploding > methodology ... after I'm done with this, I may finally get time to finish > my Garfinkel&Mahoney book and code in ObjC. > > In the meantime, I'll keep using my DGUX compiled edition, which > "seems" to work ok ... ( I'm having a strange bug related to a tutorial > example) > > > +---------------------------------+ ___ ___ ___ > | Stephane I. Matis | / \_BATLLETECH /\__\ Viva NeXT! > | E-Mail : kgnome@cs.concordia.ca | \___/ \___/ \/__/ NeXTSTEP 3.0! > | "It Just Works..." - Steve Jobs | \___/ > +---------------------------------+ Wolfnet Operative & NeXThead You really can get it to compile under NS3.0. It takes some work, though. The problem lies in compiling mstinterp.c and using the NeXT-tweaked preprocessor. If you look in /lib, you'll see both "cpp" and "cpp.precomp". The default preprocessor is the latter. It appears to have a bug in it which processing mstinterp provokes. What I did was to simply make a copy of mstinterp.c to mstinterp.c.orig. Then, I ran /lib/cpp (with the appropriate arguments from the Makefile) to produce mstinterp.i; after this, I ran cc on mstinterp.i and viola', it linked! I need to report these to both gnu.smalltalk.bug and NeXT. I'm having some problems posting from my machine at the moment, and I don't get gnu.smalltalk.bug. Would you mind passing it on? I've got to figure out where to mail for NeXT bugs. Good luck! And let me know if you need more help! -- ------------------------------------------------------------- Chris Cleeland | Internet: chris%milo@wpa.com Consultant/NeXT Advocate | UUCP: wupost!nimno!milo!chris Milo Designs | BellNet: (314) 771-3860 [End of Include]------------------------------------------------------- +---------------------------------+ ___ ___ ___ | Stephane I. Matis | / \_BATLLETECH /\__\ Viva NeXT! | E-Mail : kgnome@cs.concordia.ca | \___/ \___/ \/__/ NeXTSTEP 3.0! | "It Just Works..." - Steve Jobs | \___/ +---------------------------------+ Wolfnet Operative & NeXThead
From: jgshir@athena.mit.edu (John G Shirlaw) Newsgroups: comp.sys.next.bugs Subject: Known bugs and their fixes. Date: 26 Apr 1993 03:52:21 GMT Organization: Massachusetts Institute of Technology Distribution: world Message-ID: <1rfm9lINNog3@senator-bedfellow.MIT.EDU> Has a list of known bugs in 3.0 been created yet and if so where do I find it and preferably with the work arrounds? Thanks john.
Newsgroups: comp.sys.next.bugs From: dkk@cbnewsl.cb.att.com (david.k.kallman) Subject: Mail.app buttons dying Organization: AT&T Bell Laboratories Date: Mon, 26 Apr 1993 15:14:26 GMT Message-ID: <C63Jo4.4DI@cbnewsl.cb.att.com> Folks, Has anyone seen this behavior? Under 3.0, Mail.app wouldn't let me press the any of the square buttons (Delete, Compose, Mailboxes, Find). I quit out of Mail and restarted it, again the square buttons wouldn't work (the up/down arrows worked until I tried using the square buttons). I tried quitting/restarting a few times, one time Mail came up without any buttons, once it came up with one big rectangular gray window (except for the Active.mbox header) without any text. I rebooted the system, and Mail seems to be working OK now. Thanks. Dave Kallman, AT&T, 480 Red Hill Rd., Middletown, NJ 07748 d_k_kallman@att.com, (908)615-2989, fax: (908)615-2507 P.S.: My /usr/spool/mail is another system, NFS mounted, if this matters. -- ---- Dave Kallman, AT&T, 480 Red Hill Rd., Middletown, NJ 07748 d_k_kallman@att.com, (908)615-2989, fax: (908)615-2507
From: gary@msu.edu (Gary J LaPointe) Newsgroups: comp.sys.next.bugs Subject: Mail.app is goofy (2.2) Date: 26 Apr 1993 18:32:47 GMT Organization: Michigan State University Message-ID: <1rh9sf$n8c@msuinfo.cl.msu.edu> When I send NeXTmail (mail.app) if the preferences for NeXTmail are are not NeXTmail when I click the Icon for it and do font stufff, it doesn't show up very well (goofs up most of the time) if I send it it works fine. If I set the default for NeXTmail to true It works fine 99% of the time. It used to work fine. I cleaned out a bunch of preference files out of my directory once and it was fine for a while then too, any suggestsions? Gary -- Gary J LaPointe gary@ah3.cal.msu.edu Michigan State University Center For Integrative Studies, Arts & Humanities
Newsgroups: comp.sys.next.bugs From: gvh@metrosoft.com (Gordon Van Huizen) Subject: Re: Mail.app buttons dying Message-ID: <1993Apr26.181822.10801@metrosoft.com> Sender: gvh@metrosoft.com Organization: Metrosoft References: <C63Jo4.4DI@cbnewsl.cb.att.com> Date: Mon, 26 Apr 1993 18:18:22 GMT In article <C63Jo4.4DI@cbnewsl.cb.att.com> dkk@cbnewsl.cb.att.com (david.k.kallman) writes: > Folks, > > Has anyone seen this behavior? Under 3.0, Mail.app > wouldn't let me press the any of the square buttons > (Delete, Compose, Mailboxes, Find). I quit out of Mail and > restarted it, again the square buttons wouldn't work (the > up/down arrows worked until I tried using the square > buttons). I tried quitting/restarting a few times, one time > Mail came up without any buttons, once it came up with one > big rectangular gray window (except for the Active.mbox > header) without any text. I've had all these things happen when my mailbox contains mail without a From: field (believe it or not!). After selecting and reading a letter without a From: field (and nearly always after deleting one), the square buttons would stop working. Also when deleting a From:-less letter, I've had the phenomenon of the whole window going gray (a very big delete, indeed!). Gordon -- Gordon Van Huizen vox: 619.488.9411 fax: 619.488.3045 Metrosoft gvh@metrosoft.com [NeXTmail welcome] "Our ship is coming in, it just isn't black." - MTD 2/93
From: wfischer@ac.dal.ca Newsgroups: comp.sys.next.bugs Subject: Digital Librarian Trouble Message-ID: <1993Apr26.203517.13197@ac.dal.ca> Date: 26 Apr 93 23:35:16 GMT Organization: Dalhousie University, Halifax, Nova Scotia, Canada Got trouble with Digital Librarian: DL seems to lose track of the indices for certain files. My previous post to comp.sys.next.software generated five "me too" messages, and someone else posted the same problem to comp.sys.next.misc a couple of days ago. Sometimes when I start up DL (by clicking my bookshelf), the indices have vanished from the icons on my shelf (though they're still in the directories). The "i" is missing from the button on the icon, and DL won't find things that I know are in the files. The .index.store files still exist (I have read permission on them, but not write). It's only the files I don't personally have write permissions on (i.e. Shakespeare, Unix man pages, etc. -- the stuff in /NextLibrary) that this happens. I can update each index, but ... If I drag in the icons from the places they live, that makes it all better. Again, big pain! One other clue (maybe): the Shakespeare icon is smaller when I drag it in (and the index is there) than when I start up DL again. Any clues? I'm running NS 3.0 on a mono, turbo, non-ADB station. Tell me I'm doing some simple, stupid thing ... -- Will Fischer Nooooo signature!
Newsgroups: comp.sys.next.bugs From: dg@synex.andi.org (David M. Gursky) Subject: Floppy handling Message-ID: <C640AH.2u2@nextsrv1.andi.org> Sender: usenet@nextsrv1.andi.org (usenet) Organization: Association of NeXTSTEP Developers International Date: Mon, 26 Apr 1993 21:13:28 GMT Was there ever a fix or patch issued to correct the problem whereby DOS and Mac floppy disks could not be ejected (save logging out)?
Newsgroups: comp.sys.next.bugs From: climpach@avalon.physik.unizh.ch (Christian Limpach) Subject: why is 'cc' not compiling this right ??? Message-ID: <1993Apr27.155507.7859@ifi.unizh.ch> Sender: news@ifi.unizh.ch (USENET News Admin) Organization: University of Zurich, Department of Computer Science Date: Tue, 27 Apr 1993 15:55:07 GMT [ Article crossposted from comp.sys.next.programmer ] [ Author was Christian Limpach ] [ Posted on Mon, 26 Apr 1993 13:04:37 GMT ] From climpach Sat Apr 24 22:53:24 1993 Subject: next modfile player To: kesseler@restena.lu (Georges Kesseler) Date: Sat, 24 Apr 1993 22:53:24 +0100 (MET) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 2106 Status: OR hi, in the following piece of code, the two calls to Paula( int) are not identical. In the one which is commented out, the parameter is wrong. int FramePlayer(SNDSoundStruct *sound, int tag, int err) int i,offset; int s,sloc,samp; offset=0; for(i=0;i<precalc;i++) { offset+=4*AudioFrameSize; s=(int)sound; sloc=sound->dataLocation; samp=s+sloc; /* Paula((int)sound+sound->dataLocation+offset-4*AudioFrameSize);*/ Paula(samp+offset-4*AudioFrameSize); mt_music(); } sound->dataSize=offset; SNDStartPlaying(sound, tag+2, 5, 0, SND_NULL_FUN, FramePlayer); } The compiler generates completely wrong code, there are multiplications by 112 and 28 and things like this. Is there a bug in 'cc' ??? christian
Newsgroups: comp.sys.next.bugs From: climpach@avalon.physik.unizh.ch (Christian Limpach) Subject: is 'as' able to produce pc relative code ??? Message-ID: <1993Apr27.155803.8190@ifi.unizh.ch> Sender: news@ifi.unizh.ch (USENET News Admin) Organization: University of Zurich, Department of Computer Science Date: Tue, 27 Apr 1993 15:58:03 GMT [ Article crossposted from comp.sys.next.programmer ] [ Author was Christian Limpach ] [ Posted on Mon, 26 Apr 1993 12:55:23 GMT ] Hi, I have to use (because of the speed) some assembler parts in my programm. It seems to me like 'as' is not able to generate pc relative addressing. Something like 'lea pc@(test),a0' just doesn't put the address of test in a0. Is this a problem with 'as' or am I doing something wrong ??? If it's an 'as' problem, will a new version of 'as' help ??? christian
Newsgroups: comp.sys.next.bugs From: frank@fnbc.com (Frank Mitchell) Subject: Re: IB connections to DBTableView go off-screen Message-ID: <1993Apr27.153155.25060@fnbc.com> Sender: news@fnbc.com Organization: First National Bank Of Chicago, Chicago IL, USA References: <1993Apr27.034917.21884@tkymail.sps.mot.com> Date: Tue, 27 Apr 93 15:31:55 GMT In article <1993Apr27.034917.21884@tkymail.sps.mot.com> mshaler@tdocad.sps.mot.com (Michael Shaler) writes: > While inspecting connections between a DBModule and a DBTableView that is > not fully shown (its window is smaller than its full size and > must scroll right to show several hidden columns), the > IBconnection from the DBTableVector in the DBModule window > goes left and, as they say, south, unless the window is large > enough to fully show all columns. This happens also between a DBTableView and one of our (non-View) objects that has a custom Connection Inspector. I haven't checked simple objects yet. -- Frank Mitchell email(work): frank@fnbc.com (NeXTmail) (home): frank@gagme.chi.il.us "That is one last thing to remember:_writers are always selling somebody out_." -- Joan Didion, Preface to _Slouching Toward Bethlehem_
Newsgroups: comp.sys.next.bugs From: drew@fnbc.com (Drew Davidson) Subject: Re: IB connections to DBTableView go off-screen Message-ID: <1993Apr27.155352.26723@fnbc.com> Sender: news@fnbc.com Organization: First National Bank Of Chicago, Chicago IL, USA References: <1993Apr27.153155.25060@fnbc.com> Date: Tue, 27 Apr 93 15:53:52 GMT In article <1993Apr27.153155.25060@fnbc.com> frank@fnbc.com (Frank Mitchell) writes: > In article <1993Apr27.034917.21884@tkymail.sps.mot.com> > mshaler@tdocad.sps.mot.com (Michael Shaler) writes: > > While inspecting connections between a DBModule and a DBTableView that is > > not fully shown (its window is smaller than its full size and > > must scroll right to show several hidden columns), the > > IBconnection from the DBTableVector in the DBModule window > > goes left and, as they say, south, unless the window is large > > enough to fully show all columns. > > This happens also between a DBTableView and one of our (non-View) objects that > has a custom Connection Inspector. I haven't checked simple objects yet. Indeed. This has to do with IB's [NXApp displayConnectionBetween:source and:dest] method. DBTableView is funky in that the DBTableVectors are not Views. I think that IB is hacked to treat them specially (just like DBModels take advantage of undocumented API to do their drag-and-drop properties). Other weirdness ensues if the DBTableView is sized to obscure a vector and you select a connection to that vector (it draws the frame in the wrong place). This really p***es me off! NeXT lets DBKit look really slick by taking advantage of undocumented API, then messes it up! Feh. P.S. Hi, Frank. Get back to work! :-) (maybe *I* should get back to work too!) > -- > Frank Mitchell > email(work): frank@fnbc.com (NeXTmail) (home): frank@gagme.chi.il.us > > "That is one last thing to remember:_writers are always selling somebody out_." > -- Joan Didion, Preface to _Slouching Toward Bethlehem_ -- +--------------------------------+-------------------------------------------+ | Drew Davidson | "Never ask a programmer if he'll have | | Software Guy | another cup of coffee because it's | | First National Bank of Chicago | nobody's damn business how much he's | | drew@fnbc.com (NeXTmail) | already had!" - me | +--------------------------------+-------------------------------------------+
Newsgroups: comp.sys.next.bugs From: Benoit Grange <Benoit_Grange@gateway.qm.apple.com> Subject: Webster crashes Sender: news@gallant.apple.com Message-ID: <1993Apr28.083752.24721@gallant.apple.com> Date: Wed, 28 Apr 1993 08:37:52 GMT Organization: Apple Europe R&D On NS 3.0 try to define the word 'a' in Webster. Boom !!! Who does software quality assurance within NeXT ??? ---- Benoit Grange Paris - France
Newsgroups: comp.sys.next.bugs From: gad@eclipse.its.rpi.edu (Garance A. Drosehn) Subject: Re: Mail.app is goofy (2.2) Message-ID: <-2653rg@rpi.edu> References: <1rh9sf$n8c@msuinfo.cl.msu.edu> Date: Tue, 27 Apr 1993 23:15:34 GMT gary@msu.edu (Gary J LaPointe) writes: > When I send NeXTmail (mail.app) if the preferences for NeXTmail are are > not NeXTmail when I click the Icon for it and do font stufff, it doesn't > show up very well (goofs up most of the time) if I send it it works fine. > If I set the default for NeXTmail to true It works fine 99% of the time. > > It used to work fine. > > I cleaned out a bunch of preference files out of my directory > once and it was fine for a while then too, any suggestions? In release 2.x, the problem you describe comes up based on what "default font" you have selected for your Mail windows. I forget the details, but since I want non-NeXTmail to be my default, I've left my fonts set to Helvetica in mail. It means that incoming, non-NeXTmail looks stupid (since the sender is often assuming a fixed-pitch font), but it's the way to get around this problem in NeXTSTEP 2.x. I believe this all works fine in NeXTSTEP 3.0. To be more specific, here's the "font-ish" defaults I have set for mail. I don't remember if the font-size was important. Mail NXFontSize 14 Mail TocFont Helvetica Mail NXFont Helvetica -- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
From: isbell@cats.ucsc.edu (Art Isbell) Newsgroups: comp.sys.next.bugs Subject: Re: Webster crashes Date: 28 Apr 1993 16:56:57 GMT Organization: Cubic Solutions - NeXT software development and consulting Message-ID: <1rmd0qINNrgj@darkstar.UCSC.EDU> References: <1993Apr28.083752.24721@gallant.apple.com> In article <1993Apr28.083752.24721@gallant.apple.com> Benoit Grange <Benoit_Grange@gateway.qm.apple.com> writes: >On NS 3.0 try to define the word 'a' in Webster. > >Boom !!! No boom in 3.1 PR1, but "Word not found" is returned. Could the word "a" really not be in Digital Webster? "an" is found and seems to refer to "a". -- Art Isbell Cubic Solutions NeXT Registered Developer #745 NeXT software development and consulting NeXTmail: isbell@cats.UCSC.EDU Voice: (408)335-1154 USmail: 95018-9442 Fax: (408)335-2515
Newsgroups: comp.sys.next.bugs From: thompson@netcom.com Subject: Re: Webster crashes Message-ID: <thompsonC67HB3.132@netcom.com> Organization: Netcom - Online Communication Services (408 241-9760 guest) References: <1993Apr28.083752.24721@gallant.apple.com> <1rmd0qINNrgj@darkstar.UCSC.EDU> Date: Wed, 28 Apr 1993 18:13:51 GMT >>On NS 3.0 try to define the word 'a' in Webster. >> >>Boom !!! > >No boom in 3.1 PR1, but "Word not found" is returned. Could the word "a" >really not be in Digital Webster? "an" is found and seems to refer to "a". Well, try "b"! :-) Actually, it used to die on "5"--does it still? Eric
From: isbell@cats.ucsc.edu (Art Isbell) Newsgroups: comp.sys.next.bugs Subject: Re: Webster crashes Date: 28 Apr 1993 19:40:42 GMT Organization: Cubic Solutions - NeXT software development and consulting Message-ID: <1rmmjqINN1q4@darkstar.UCSC.EDU> References: <1993Apr28.083752.24721@gallant.apple.com> <1rmd0qINNrgj@darkstar.UCSC.EDU> <thompsonC67HB3.132@netcom.com> In article <thompsonC67HB3.132@netcom.com> thompson@netcom.com writes: >>>On NS 3.0 try to define the word 'a' in Webster. >>>Boom !!! >>No boom in 3.1 PR1, but "Word not found" is returned. Could the word "a" >>really not be in Digital Webster? "an" is found and seems to refer to "a". >Well, try "b"! :-) Actually, it used to die on "5"--does it still? b \'beE\ 1a: the 2d letter of the English alphabet b: a graphic representation of this letter c: a speech counterpart of orthographic b 2: the 7th tone of a C-major scale 3: a graphic device for reproducing the letter b 4: one designated b esp. as the 2d in order or class 5a: a grade rating a student's work as good but short of excellent b: one graded or rated with a B 6: something shaped like the letter B A search for "5" returns "Word not found". Seems like NeXT has fixed some of D.W.'s problems. -- Art Isbell Cubic Solutions NeXT Registered Developer #745 NeXT software development and consulting NeXTmail: isbell@cats.UCSC.EDU Voice: (408)335-1154 USmail: 95018-9442 Fax: (408)335-2515
From: anstine@uniblab.sas.upenn.edu (David Anstine) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: Bug in Upgrade.app Message-ID: <123861@netnews.upenn.edu> Date: 29 Apr 93 01:11:24 GMT Sender: news@netnews.upenn.edu Followup-To: comp.sys.next.bugs Organization: University of Pennsylvania A few months ago I upgraded a machine to 3.0. I used this machine to netupgrade ~20 more machines. I later discovered that none of these machines could properly run netinfo applications, ie: whenever they called an Open Domain dialog box, they crashed. The problem was tracked to /usr/lib/netinfo on the upgrade server which was corrupt and hence propagated throughout my network. I wrote the corruption off as a fluke. Today on the next-managers list, someone complained about the same exact thing. Replacing /usr/lib/netinfo from CD also solved his problem. Anyone else experience this? -dave
Newsgroups: comp.sys.next.bugs From: kira!davidjohn (David John Burrowes) Subject: Re: Mail.app buttons dying Message-ID: <1993Apr28.020612.299@kira.net.netcom.com> Sender: davidjohn@kira.net.netcom.com Organization: No organization at this time. References: <C63Jo4.4DI@cbnewsl.cb.att.com> Date: Wed, 28 Apr 1993 02:06:12 GMT In article <C63Jo4.4DI@cbnewsl.cb.att.com> dkk@cbnewsl.cb.att.com (david.k.kallman) writes: > Folks, > > Has anyone seen this behavior? Under 3.0, Mail.app > wouldn't let me press the any of the square buttons > (Delete, Compose, Mailboxes, Find). I quit out of Mail and > restarted it, again the square buttons wouldn't work (the > up/down arrows worked until I tried using the square > buttons). I tried quitting/restarting a few times, one time > Mail came up without any buttons, once it came up with one > big rectangular gray window (except for the Active.mbox > header) without any text. > P.S.: My /usr/spool/mail is another system, NFS mounted, > if this matters. I've seen this, though all my stuff is on my local machine. I've also had cases where mail.app seems to have gotten it's coordinate system inverted (bottom button scrolls up, top scrolls down) and one time the entire contents of the window appeared (text and all) upside down (buttons down by the scrollbar, etc). I've never been able to figure out why this happened so sporadically. The bit that was suggested about no From: line is interesting. But, my problems weren't at all consistant... \david john burrowes
From: dockd@storm.cs.orst.edu (Dion Dock) Newsgroups: comp.sys.next.bugs Subject: bug in help panel Message-ID: <1rnddiINNf7u@flop.ENGR.ORST.EDU> Date: 29 Apr 93 02:09:54 GMT Article-I.D.: flop.1rnddiINNf7u Distribution: usa Organization: Computer Science Department, Oregon State University I found something which may or may not have been reported before. If you try to search for words separated by a space, like "info panel", in a NS3.0 help panel, your program will hang. This has been fixed in NS3.1 for Intel. Dion -- Dion Dock __ __ NeXT mail? working on it / ) / ) / dockd@storm.cs.orst.edu / / o ______ / / _____. /_ /__/_<_(_) / <_ /__/_(_) (__/ <_
Newsgroups: comp.sys.next.bugs From: dleon@seas.smu.edu (David Leon) Subject: Re: Webster crashes Message-ID: <1993Apr30.013838.1074@seas.smu.edu> Sender: news@seas.smu.edu (USENET News System) Organization: SMU - School of Engineering & Applied Science - Dallas References: <1993Apr28.083752.24721@gallant.apple.com> <1rmd0qINNrgj@darkstar.UCSC.EDU> <thompsonC67HB3.132@netcom.com> Date: Fri, 30 Apr 1993 01:38:38 GMT My 3.0 Digi Web crashes immediately after it has been launched. I have tried it with different programs in the background and by itself. I have seen it work in monthes.
Newsgroups: comp.sys.next.bugs From: matt@zeb.ame.nd.edu (Matthew J. Grismer) Subject: Terminal and Mail app bugs Message-ID: <1993Apr29.142719.3361@news.nd.edu> Keywords: mail, terminal,bugs Sender: news@news.nd.edu (USENET News System) Organization: University of Notre Dame Date: Thu, 29 Apr 1993 14:27:19 GMT Anyone noticed the following bugs? Terminal - I find that when I telnet to another workstation, long lines no longer wrap, they just get cut off at the edge of the window. I've found that if I go to window option in the preferences panel and click on the set window button, everything is fixed. Mail - After I log in I must use the get new mail option with the Mail.app, otherwise it will never show me that I have new mail for the rest of the day. -- Matthew J. Grismer M M JJJJJJJJJJ GGGGGGGG 300 Cushing Hall MM MM J G University of Notre Dame M M M M J G GGGG e-mail: matt@zeb.ame.nd.edu M M M M J G G NeXTmail preferred M M M M J J G G M MM M JJJJJ GGGGGGGG
Newsgroups: comp.sys.next.bugs From: gordie@cyclesoft.com (Gordie Freedman) Subject: Re: Terminal and Mail app bugs Message-ID: <1993May3.075517.2470@netcom.com> Sender: gordie@netcom.com Organization: Cyclesoft Media Works References: <1993Apr29.142719.3361@news.nd.edu> Date: Mon, 3 May 1993 07:55:17 GMT In article <1993Apr29.142719.3361@news.nd.edu> matt@zeb.ame.nd.edu (Matthew J. Grismer) writes: > Anyone noticed the following bugs? > > Mail - After I log in I must use the get new mail option with the > Mail.app, otherwise it will never show me that I have new mail for the > rest of the day. > -- This is a known bug - if you autolaunch mail and hide it on autolaunch (a Mail pref) it doesn't check for new mail until you do it once explicitly. I don't hide it on autolaunch, and things seem to work. -- >>> Gordie Freedman -> gordie@cyclesoft.com NeXTMail Yes! >>> Thou shalt not inline functions more complicated than 20
Newsgroups: comp.sys.next.bugs From: nwc (Nick Christopher) Subject: sendmail.cf bug! Message-ID: <C6GBFq.ALG@ny.shl.com> Sender: usenet@ny.shl.com (Net News) Organization: SHL Systemhouse Inc. Date: Mon, 3 May 1993 12:45:26 GMT Well its actually just a possible problem rather than a bug.... The sendmail.sharedsubsidiary.cf shipped by NeXT includes the line: FV|/usr/bin/uuname Why would a subsidiary have uucp connections? The above caused me MUCH GRIEF!!! We had to move our modem from one cpu to another. I correctly switched senmail.cf files on both machines, but I forgot to kill the L.sys on the old mailhost. The whole net was fine - so the swap worked right? Nope as I eventually realized mail on the old mailhost, sent to our uucp partner, WAS GOING INTO THE BIT BUCKET!!!!! This was happening because uuname was still returning a value and the sharedsubsidiary.cf was using it. \n Nicholas Christopher Systemhouse nwc@ny.shl.com
Newsgroups: comp.sys.next.bugs From: dan@ennex1.eng.utsa.edu (Daniel Shelton) Subject: Re: Terminal and Mail app bugs Message-ID: <1993May3.133300.3180@ringer.cs.utsa.edu> Sender: news@ringer.cs.utsa.edu Organization: Univ of Texas at San Antonio References: <1993May3.075517.2470@netcom.com> Date: Mon, 3 May 1993 13:33:00 GMT In article <1993May3.075517.2470@netcom.com> gordie@cyclesoft.com (Gordie Freedman) writes: > In article <1993Apr29.142719.3361@news.nd.edu> matt@zeb.ame.nd.edu (Matthew J. > Grismer) writes: > > Anyone noticed the following bugs? > > > > Mail - After I log in I must use the get new mail option with the > > Mail.app, otherwise it will never show me that I have new mail for the > > rest of the day. > > -- > > This is a known bug - if you autolaunch mail and hide it on autolaunch (a Mail > pref) it doesn't check for new mail until you do it once explicitly. I don't > hide it on autolaunch, and things seem to work. > -- > >>> Gordie Freedman -> gordie@cyclesoft.com NeXTMail Yes! > >>> Thou shalt not inline functions more complicated than 20 I am using 3.0 on a network and I must be lucky because my mail does not exhibit this bug. I have my mail.app hide on autolaunch and it works fine. Also all the user accounts on the system do not show this bug. Are you absolutely sure there is not a system setup error? Dan (dan@ennex1.eng.utsa.edu)
From: kjell@oops (Kjell Nilsson) Newsgroups: comp.sys.next.bugs Subject: Can't start IB Message-ID: <1993May3.142344.20430@oops.se> Date: 3 May 93 14:23:44 GMT Sender: kjell@oops.se Organization: OOPSart - ObjectWareHouse Help Suddenly I could not start IB. All I get is a runtimeerror, and it comes before anything else happens. Runtime error:NXReadOnlyString : does not recognice selector -replaceWith:. The IB I use is the 3.0 one (cant se the versionnr now). here is the console output May 3 15:16:12 oops InterfaceBuilder[20385]: An uncaught exception was raised May 3 15:16:12 oops InterfaceBuilder[20385]: Unknown error code 314159 in NXReportError Has someone had this before??? Kjell Nilsson OOPS - ObjectWareHouse
Newsgroups: comp.sys.next.bugs From: wave@media.mit.edu (Michael B. Johnson) Subject: Re: Can't start IB In-Reply-To: kjell@oops's message of 3 May 93 14:23:44 GMT Message-ID: <WAVE.93May3164318@media-lab.media.mit.edu> Sender: news@news.media.mit.edu (USENET News System) Organization: M.I.T. Media Laboratory References: <1993May3.142344.20430@oops.se> Date: Mon, 3 May 1993 21:43:18 GMT In article <1993May3.142344.20430@oops.se> kjell@oops (Kjell Nilsson) writes: Help Suddenly I could not start IB. All I get is a runtimeerror, and it comes before anything else happens. Runtime error:NXReadOnlyString : does not recognice selector -replaceWith:. The IB I use is the 3.0 one (cant se the versionnr now). here is the console output May 3 15:16:12 oops InterfaceBuilder[20385]: An uncaught exception was raised May 3 15:16:12 oops InterfaceBuilder[20385]: Unknown error code 314159 in NXReportError Has someone had this before??? You've most likely loaded a palette into IB that has a bug in it. If you just want to start IB, edit ~/.NeXT/defaults.nibd to remove reference to the offending palette. If you actually want to find the bug (i.e. you wrote the buggy palette and want to fix it), run IB inside of gdb, i.e. in a Terminal Window: % gdb /NextDeveloper/Apps/InterfaceBuilder.app/InterfaceBuilder Type "r" and you're off. When you get the error, sniff around and find the bug. -- --> Michael B. Johnson --> MIT Media Lab -- Computer Graphics & Animation Group --> (617) 253-0663 -- wave@media-lab.media.mit.edu
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: eps@futon.SFSU.EDU (Eric P. Scott) Subject: SNDConvertSound() BROKEN under 3.0?!? Message-ID: <1993May4.073343.13071@csus.edu> Followup-To: comp.sys.next.bugs Summary: 3.0 is a pile of sh*t that keeps getting deeper Sender: news@csus.edu Organization: San Francisco State University Date: Tue, 4 May 1993 07:33:43 GMT The attached adaptation of NeXT's converttest.c example illustrates SERIOUS breakage in 3.0's Sound Library. Try it with any 8-bit mu law sound. This bug is not mentioned in the 3.0 Release Notes, NeXTanswers, nor the "3.0 known bugs" list. As is, SNDConvertSound() fails with SND_ERR_CANNOT_ALLOC. Compiled with -DORIGINAL, it works just fine. Compiled on a NeXTSTEP 2.2 system, it works correctly either way. If I run the 2.2 binaries on a 3.0 system, it dies with a SIGSEGV (the gdb backtrace is amazing!). The -DORIGINAL version works (as expected). Would someone with 3.1PRx on Black Hardware please test this? -=EPS=- ------- /* * converttest.c - an example conversion that takes a mulaw codec file * as input and writes a new file (11 times bigger) that is the equivalent * 22kHz mono 16 bit linear soundfile */ #import <sound/sound.h> #import <stdio.h> check_error(int err) { if (err) { printf("Error : %s\n",SNDSoundError(err)); exit(1); } return err; } main (int argc, char *argv[]) { void showsnd(SNDSoundStruct *); int err; SNDSoundStruct *s1, *s2; SNDSoundStruct header = { SND_MAGIC, 0, 0, SND_FORMAT_LINEAR_16, (int)SND_RATE_LOW, 1, "" }; SNDSoundStruct codecsnd; check_error(argc != 3); err = SNDReadSoundfile(argv[1],&s1); check_error(err); (void)fputs("s1\n", stderr); showsnd(s1); codecsnd.magic=SND_MAGIC; codecsnd.dataLocation=((char *)s1)+s1->dataLocation-(char *)&codecsnd; codecsnd.dataSize=s1->dataSize; codecsnd.dataFormat=SND_FORMAT_MULAW_8; codecsnd.samplingRate=(int)SND_RATE_CODEC; codecsnd.channelCount=1; codecsnd.info[0]='\0'; (void)fputs("codecsnd\n", stderr); showsnd(&codecsnd); s2 = &header; #ifdef ORIGINAL err = SNDConvertSound(s1,&s2); #else err = SNDConvertSound(&codecsnd,&s2); #endif check_error(err); err = SNDWriteSoundfile(argv[2],s2); check_error(err); exit(0); } void showsnd(SNDSoundStruct *s1) { (void)printf("\t.magic=0x%08x\n\ \t.dataLocation=0x%08x (%p)\n\ \t.dataSize=%d\n\ \t.dataFormat=%d\n\ \t.samplingRate=%d\n\ \t.channelCount=%d\n\ \t.info[0]=%d\n", s1->magic, s1->dataLocation, ((char *)s1)+s1->dataLocation, s1->dataSize, s1->dataFormat, s1->samplingRate, s1->channelCount, s1->info[0]); }
From: doko@cs.tu-berlin.de (Matthias Klose) Newsgroups: comp.sys.next.bugs Subject: mounting NFS directories from a Sun running Solaris ? Date: 04 May 93 12:38:29 GMT Organization: Technical University of Berlin, Germany Message-ID: <1s5kjn$q2j@news.cs.tu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit I am not sure if this is a bug in the NeXTstep Window Manager or a unknown feature/bug in the NFS implementation of Solaris. If a directory sun:/export/next/LocalApps is mounted on the NeXT in /LocalApps, the FileViewer shows in BrowserMode a directory loop after the second directory: /LocalApps/DemoApp.app/Dir/Dir/Dir ... However there is no loop if I open a shell window and change to this directory. I have got a message (probably) describing the problem; can you tell me on which side the problem lays and/or it is resolved in NS3.1? Thanks, Matthias Newsgroups: comp.sys.sun.admin Path: cs.tu-berlin.de!math.fu-berlin.de!news.netmbx.de!Germany.EU.net!mcsun!sun4nl!fwi.uva.nl!casper From: casper@fwi.uva.nl (Casper H.S. Dik) Subject: Re: C compile on NFS on Solaris 2.1 Message-ID: <1993Apr29.080803.14109@fwi.uva.nl> Sender: news@fwi.uva.nl Nntp-Posting-Host: adam.fwi.uva.nl Organization: FWI, University of Amsterdam References: <C5L9Fq.HwH@cup.hp.com> <1620@racerx.bridge.COM> <17646@auspex-gw.auspex.com> Date: Thu, 29 Apr 1993 08:08:03 GMT Lines: 39 guy@Auspex.COM (Guy Harris) writes: >>Is Solaris doing >>something in the NFS protocol that is in the spec and supported by Sun >>but not well by others because it's never been an issue before Solaris >>came along and started doing it? >It's certainly possible, although much of SunOS 4.x's NFS server code >dates back before SunOS 5.x, so if it doesn't screw up SunOS 4.x >servers, it's probably because they discovered, early on, at least some >of the problems it had with 4.x servers, and fixed them before they >shipped 5.x (for some value of "x"). >Then again, it's also possible that it's doing something *not* in the >spec, but that SunOS 4.x's server code happens to be able to handle. That's probably the case: Patch-ID# 100939-01 Keywords: wrong file size full file system nfs module fs server mount readdir NF S Synopsis: SunOS 5.1: Fixes to NFS module Date: Mar/19/93 [ .... ] This bug causes Concurrent's "pwd" command to report incomplete pathnames. This may also affect other vendors who rely on the dirent data to include accurate name length data. KRBrown The description field as copied from bug report 1119254 follows: In fastpath, the readdir result is counting the null byte in the file name. For example the name "." has a count of 2 and a value of ".\0" Probable cause is the use of copystr() which returns the string length including the NULL byte. [ ..... ] Casper
Newsgroups: comp.sys.next.bugs From: stabl@antigone.ppp.informatik.uni-muenchen.de (Robert Stabl) Subject: Re: mounting NFS directories from a Sun running Solaris ? Message-ID: <1993May4.232044.14947@antigone.ppp.informatik.uni-muenchen.de> Sender: stabl@antigone.ppp.informatik.uni-muenchen.de Organization: Institut fuer Informatik der Universitaet Muenchen References: <1s5kjn$q2j@news.cs.tu-berlin.de> Date: Tue, 4 May 1993 23:20:44 GMT In article <1s5kjn$q2j@news.cs.tu-berlin.de> doko@cs.tu-berlin.de (Matthias Klose) writes: > I am not sure if this is a bug in the NeXTstep Window Manager or a > unknown feature/bug in the NFS implementation of Solaris. > > If a directory sun:/export/next/LocalApps is mounted on the NeXT in > /LocalApps, the FileViewer shows in BrowserMode a directory loop after > the second directory: /LocalApps/DemoApp.app/Dir/Dir/Dir ... > However there is no loop if I open a shell window and change to this > directory. > > I have got a message (probably) describing the problem; can you tell > me on which side the problem lays and/or it is resolved in NS3.1? > That's probably the case: > > Patch-ID# 100939-01 > Keywords: wrong file size full file system nfs module fs server mount readdir NF > S > Synopsis: SunOS 5.1: Fixes to NFS module > Date: Mar/19/93 We had the same problem. It is fixed by the above mentioned patch on the Solaris (SUN) side. But remember that there is a severe bug in the NeXT NFS implemenation when performing a chown system call on a mounted Solaris file system on the NeXT side. To fix this bug we had to patch the NeXT kernel. Robert Stabl. -- Robert Stabl email: stabl@informatik.uni-muenchen.de Computer Science Institute Tel: +(49) 89 2180 6316 University of Munich FAX: +(49) 89 2180 6310 Leopoldstr. 11B "Wer die Wahrheit nicht weiss, der ist bloss ein D-8000 Muenchen 40 Dummkopf. Aber wer sie weiss und sie eine Luege Germany nennt, der ist ein Verbrecher!" B.Brecht,"Galileo"
From: fxg@damabus.informatik.rwth-aachen.de (Felix Gatzemeier) Newsgroups: comp.sys.next.bugs Subject: Localization.rtfd/PB.app inconsistent => Localization impossible Date: 5 May 93 15:06:28 GMT Organization: Rechnerbetrieb Informatik - RWTH Aachen Message-ID: <fxg.736614388@damabus> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Summary: PB doesn't work as described in Localization.rtfd, features missing Keywords: Localization, languages, nextstep3.0 Problems: 1) The Concepts/Localizatin.rtfd file describes genstrings twice, each time differently, genstrings itself doesn't work quite the way any try describes it (no big problem, though) 2) The Project Builder doesn't offer the functionality related to in the file mentioned above. 3) genstrings has no man page (quite nerdy, I know). Details: 1) Speaks for itself. The differences are bearable and can be resolved by consulting genstrings itself. 2) This is grave. At some point, you have to tell PB about the different languages you are using. If this is possible at all, it's well hidden. I read ProjectBuilder.rtf, no hint. I even searched ProjectBuilder.app for strings like .lproj, no success. When adding a folder to a project, PB adds its contents and bickers about undetermineable file types etc. I've tried for hours: NO WAY. 3) Small problem, small solution. Comments: If NeXT is announcing multilinguality as a feature of their OS, they should really make implementing it as easy as it's described in the documentation. The description sounds nice, not real smooth, but worth the effort. Like most of us, I hadn't had a chance to look at 3.1. -- Felix
Newsgroups: comp.sys.next.bugs From: lorinr@altsys.com (Lorin Rivers III) Subject: Re: mounting NFS directories from a Sun running Solaris ? Message-ID: <1993May5.145506.21581@altsys.com> Organization: Altsys Corporation, Richardson, TX References: <1s5kjn$q2j@news.cs.tu-berlin.de> Date: Wed, 5 May 1993 14:55:06 GMT In article <1s5kjn$q2j@news.cs.tu-berlin.de> doko@cs.tu-berlin.de writes: >I am not sure if this is a bug in the NeXTstep Window Manager or a >unknown feature/bug in the NFS implementation of Solaris. > >If a directory sun:/export/next/LocalApps is mounted on the NeXT in >/LocalApps, the FileViewer shows in BrowserMode a directory loop after >the second directory: /LocalApps/DemoApp.app/Dir/Dir/Dir ... >However there is no loop if I open a shell window and change to this >directory. > >I have got a message (probably) describing the problem; can you tell >me on which side the problem lays and/or it is resolved in NS3.1? > > Thanks, Matthias > -- Lorin Rivers Lorin_Rivers@altsys.com Altsys Technical Support 214.680.2518 269 W. Renner Parkway NeXT Mail Expected Richardson, Texas 75080 I said it, not my boss
Newsgroups: comp.sys.next.bugs From: speters@maverick.oracle.com (Stephen Peters) Subject: Re: Localization.rtfd/PB.app inconsistent => Localization impossible Message-ID: <1993May5.170632.9767@oracle.us.oracle.com> Keywords: Localization, languages, nextstep3.0 Sender: usenet@oracle.us.oracle.com (Oracle News Poster) Organization: Oracle Corp. References: <fxg.736614388@damabus> Date: Wed, 5 May 1993 17:06:32 GMT at Oracle Corporation. The opinions expressed are those of the user and not necessarily those of Oracle. In article <fxg.736614388@damabus>, fxg@damabus.informatik.rwth-aachen.de (Felix Gatzemeier) writes: > Problems: > 1) The Concepts/Localizatin.rtfd file describes genstrings twice, > each time differently, genstrings itself doesn't work quite the way any > try describes it (no big problem, though) I forget what the problem was the second time it describes genstrings -- if I recall, it (correctly) stated to run "genstrings [source files] > Localizable.strings". Of course, it would be nice if genstrings worked the way it said in the first part of the doc -- you type genstrings and it makes all the appropriate tables... > 2) The Project Builder doesn't offer the functionality related to in the > file mentioned above. see below. > 3) genstrings has no man page (quite nerdy, I know). This is annoying, especially because the RTF file contradicts itself -- the man page is an appropriate place to go. Hopefully this is merely an oversight and is fixed in 3.1. >Details: > 2) This is grave. At some point, you have to tell PB about the different > languages you are using. If this is possible at all, it's well hidden. I > read ProjectBuilder.rtf, no hint. I even searched ProjectBuilder.app > for strings like .lproj, no success. When adding a folder to a project, > PB adds its contents and bickers about undetermineable file types etc. > I've tried for hours: NO WAY. That's strange. I didn't have any problems -- I just made a French.lproj folder, put it in my source directory, and then added it to the project normally (under "Other Resources", I believe), and it worked without a hitch. If you want more information, Felix, write me (speters@us.oracle.com) and I'll go home and try to figure out exactly what I did. > Comments: > If NeXT is announcing multilinguality as a feature of their OS, they should > really make implementing it as easy as it's described in the documentation. > The description sounds nice, not real smooth, but worth the effort. Although it sounds like I didn't have quite as much trouble as you, I agree with you completely. Localization.rtfd sounds like it got a bit confused when they were writing it, and when the genstrings change occurred, they forgot to update the doc or write a man page. Attention to details was kind of low here. Overall, though, I like the way that they've made Localization work. It's very nicely done, and I was able to localize my test program (a home library catalog that I've done partially for usefulness and partially as a learning tool) in less than an hour. > Like most of us, I hadn't had a chance to look at 3.1. Is anyone *with* 3.1 able to say whether a man page now exists for genstrings? :-) Stephen Peters
From: isbell@cats.ucsc.edu (Art Isbell) Newsgroups: comp.sys.next.bugs,comp.sys.next.software Subject: Sybase and DBModeler bugs Date: 5 May 1993 18:43:08 GMT Organization: Cubic Solutions - NeXT software development and consulting Distribution: world Message-ID: <1s91rsINNnj5@darkstar.UCSC.EDU> Summary: You'll be bitten if you're using primary keys with > 4 columns. Keywords: Sybase,DBModeler We've been having strange problems with our DBKit app running with a Sybase database under both 3.0 and 3.1 PR1. The bottom line was that our dbmodel created by DBKit was missing some of the attributes in the compound keys of some of our entities. What's happening is caused by bugs in both Sybase's sp_primarykey system procedure and in DBModeler. Below is a patch which fixes sp_primarykey in the /usr/sybase/scripts/installermaster script that is used when a server is being created. For existing servers, the system administrator should dump sp_primarykey using sp_helptext and copy the dump to a file, drop sp_primarykey, edit the sp_primarykey code as is indicated in the patch and also correct any spacing errors which seem to creep in during the dump, then create a new sp_primarykey system procedure. -------------------------------- cut here ------------------------------------- *** installmaster.orig Thu Sep 6 08:36:18 1990 --- installmaster Wed May 5 10:42:13 1993 *************** *** 35,40 **** --- 35,45 ---- ** CA 94608. ** ** Stored procedures for Data Dictionary Operations + ** + ** Fixed bug in sp_primarykey that caused key6 to be identical to key4 + ** resulting in an invalid compound key where the number of key columns + ** exceeds 5. Note that this doesn't affect Sybase operations, only those + ** client applications that read the Sybase data dictionary (e.g. DBModeler). */ go /* a_permitboth 45.1 12/15/89 */ *************** *** 9489,9495 **** begin select @cnt = @cnt + 1, @key6 = colid from syscolumns ! where name = @col4 and id = @objid if @key6 is NULL begin --- 9494,9500 ---- begin select @cnt = @cnt + 1, @key6 = colid from syscolumns ! where name = @col6 and id = @objid if @key6 is NULL begin -------------------------------- cut here ------------------------------------- Note that this bug has been in installmaster at least since Sept. 1990, but it doesn't affect the operation of Sybase itself because the indices seem to be correct. However any client applications that read Sybase's data dictionary for key information will be missing key 6. DBModeler reads Sybase's data dictionary. However, the 3.1 PR-1 (and apparently the 3.0) DBModeler adds to Sybase's problems by missing some of the key columns that Sybase knows about :-( Under 3.0, the missing keys can be added manually in DBModeler or programmatically using private API. In addition, under 3.1, an ASCII model file can be created manually or programatically. Both of these workarounds are OK for small databases, but for large schemas, the workarounds are very time-consuming. Bug reports have been filed with NeXT. -- Art Isbell Cubic Solutions NeXT Registered Developer #745 NeXT software development and consulting NeXTmail: isbell@cats.UCSC.EDU Voice: (408)335-1154 USmail: 95018-9442 Fax: (408)335-2515
Newsgroups: comp.sys.next.bugs From: lorinr@altsys.com (Lorin Rivers III) Subject: Re: mounting NFS directories from a Sun running Solaris ? Message-ID: <1993May5.161252.22410@altsys.com> Organization: Altsys Corporation, Richardson, TX References: <1s5kjn$q2j@news.cs.tu-berlin.de> <1993May5.145506.21581@altsys.com> Date: Wed, 5 May 1993 16:12:52 GMT In article <1993May5.145506.21581@altsys.com> lorinr@altsys.com (Lorin Rivers III) writes: >In article <1s5kjn$q2j@news.cs.tu-berlin.de> doko@cs.tu-berlin.de writes: >>I am not sure if this is a bug in the NeXTstep Window Manager or a >>unknown feature/bug in the NFS implementation of Solaris. >> >>If a directory sun:/export/next/LocalApps is mounted on the NeXT in >>/LocalApps, the FileViewer shows in BrowserMode a directory loop after >>the second directory: /LocalApps/DemoApp.app/Dir/Dir/Dir ... >>However there is no loop if I open a shell window and change to this >>directory. >> >>I have got a message (probably) describing the problem; can you tell >>me on which side the problem lays and/or it is resolved in NS3.1? >> >> Thanks, Matthias >> What I meant to say is, I too have had this experience. -- Lorin Rivers Lorin_Rivers@altsys.com Altsys Technical Support 214.680.2518 269 W. Renner Parkway NeXT Mail Expected Richardson, Texas 75080 I said it, not my boss
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: eps@futon.SFSU.EDU (Eric P. Scott) Subject: Re: SNDConvertSound() BROKEN under 3.0?!? Message-ID: <1993May7.021956.3759@csus.edu> Sender: news@csus.edu Organization: San Francisco State University References: <1993May4.073343.13071@csus.edu> Date: Fri, 7 May 1993 02:19:56 GMT The problem has been confirmed by several sites as unfixed in 3.1PR1/Lightning3A. I've received a response from NeXT, which while not admitting the existence of a problem, suggests that a workaround _may_ exist involving SND_FORMAT_INDIRECT. This looks like a job for NeXTanswers!! -=EPS=-
Newsgroups: comp.sys.next.bugs From: rob@shw.optics.rochester.edu (Rob Smith) Subject: patches for apps when going to 3.0? Message-ID: <1993May7.170557.14345@galileo.cc.rochester.edu> Sender: news@galileo.cc.rochester.edu Organization: University of Rochester Computing Center Date: Fri, 7 May 93 17:05:57 GMT Having upgrades to 3.0 I am wording if there are any patches out there that I need to make my apps that ran under 2.1 run better. Or is all this crashing out of apps just general softwear rot in general. My biggest concerns are, WriteNow 2.0 Adobe Illustrator 3.01 sorry if this is an old question. email preferred thanks rob
Newsgroups: comp.sys.next.bugs From: rob@shw.optics.rochester.edu (Rob Smith) Subject: patches for apps when going to 3.0? (less typos) Message-ID: <1993May7.171111.14563@galileo.cc.rochester.edu> Sender: news@galileo.cc.rochester.edu Organization: University of Rochester Computing Center Date: Fri, 7 May 93 17:11:11 GMT (sorry about that) Having upgraded to 3.0 I am wondering if there are any patches out there that I need to make my apps that ran under 2.1 run better. Or is all this crashing out of apps just general softwear rot. My biggest concerns are, WriteNow 2.0 Adobe Illustrator 3.01 sorry if this is an old question. email preferred thanks rob
Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs,comp.sys.next.misc From: bebeada@elof.iit.edu (Adam Beberg) Subject: HELP!...missing header files... Message-ID: <1993May8.002857.22055@iitmax.iit.edu> Summary: lost headers.. Sender: Adam Beberg Organization: Illinois Institute of Technology, Chicago Date: Sat, 8 May 93 00:28:57 GMT hello, my problem: i'm using cc and trying to compile POVray.. and i need the header file that defines the math co-processor error conditions DOMAIN,UNDERFLOW,SING,OVERFLOW i have grep'ed etc. every file on the machine with no luck. Machine: NeXT 68040, NeXT Mach && bsd 4.3 if anyone out there in NeXTland can help please mail me the header or a site where i can FTP it from. thanks in advance Adam Beberg bebeada@elof.acc.iit.edu ------------------------------------------------------------------------------ Emergency... Emergency... There's an emergency going on... It's still going on.. It's still an emergency... This is an emergency announcement... ))) In Stereo Where Available (((
From: rfschtkt%banruc60.BITNET@ohstvma.acs.ohio-state.EDU Newsgroups: comp.sys.next.bugs Subject: NXBrowser with NXImages in some Cells Date: 10 May 1993 06:19:16 -0400 Organization: The Ohio State University Sender: daemon@magnus.acs.ohio-state.edu Message-ID: <9305081636.AA03767@flexus> LS, I have found a second bug related to NXBrowserCells with an NXImage. This is its report in knownbugsin3.0v10.1.rtf.Z (it's clearer in RTF). >>>>> reloadColumn: with some NXBrowserCells having an NXImage, NS_3.0.10.1.021 I have found that whenever I reloadColumn:0 with some Cells carrying an image, my application crashes because of a freed object being freed again. Running under gdb (with and without -O) shows that this happens during a setImage: message. Here's some further evidence (a piece of my code, and its result): printf("smallUnknownIcon=%p, [smallUnknownIcon class]=%p before\n", smallUnknownIcon , [smallUnknownIcon class]); if(0!=((struct segment_command *)p_lc)->filesize) [[matrix cellAt:row :0] setImage:smallUnknownIcon]; printf("smallUnknownIcon=%p, [smallUnknownIcon class]=%p after\n", smallUnknownIcon , [smallUnknownIcon class]); flexus> Otool.debug/Otool smallUnknownIcon=0xed410, [smallUnknownIcon class]=0x6160c30 before smallUnknownIcon is +initialized once smallUnknownIcon=0xed410, [smallUnknownIcon class]=0x6160c30 after smallUnknownIcon=0xed410, [smallUnknownIcon class]=0x6160c30 before [...] smallUnknownIcon=0xed410, [smallUnknownIcon class]=0x6160c30 after (until here the first load of the column) /bin/strip: /Users/rfschtkt/WorkBench/NeXTApps/robot/robot already stripped (pressed a button which reloads the document) smallUnknownIcon=0xed410, [smallUnknownIcon class]=0x6160c30 before smallUnknownIcon=0xed410, [smallUnknownIcon class]=0x6160c30 after (this first NXBrowserCell never has an image associated with it) smallUnknownIcon=0xed410, [smallUnknownIcon class]=0x6160c30 before (this one has, before and after) objc: FREED(id): message class sent to freed object=0xed410 (I didn't do it!) IOT trap (core dumped) flexus> Workaround, anyone? <<<<< knownbugsin3.0v10.1.rtf.Z should appear on sonata by the end of next week; the current version is 10.0, in the form of 8.0 and 9.1 put together, in sonata.cc.purdue.edu:next/submissions. This is the place to check for the state of public knowledge about bugs in NeXTSTEP_3.0; it's over 100 kB in compressed state. Do you know a bug that's not mentioned yet? Tell me! Raf Schietekat, RfSchtkt@banruc60.bitnet, Flanders, Belgium (real, i.e., with triangle in the Deliver button) NeXT Mail preferred I can't reach sites with ! or % in their address, or ending in ``at'' Nederlands, English, Francais, Deutsch a.o. PS: Should I submit the next version of knownbugs... to orst as well?
Newsgroups: comp.sys.next.bugs From: test@ie1next.me.umn.edu () Subject: NeXT.default problem Message-ID: <C6uDrt.4MM@news2.cis.umn.edu> Sender: news@news2.cis.umn.edu (Usenet News Administration) Organization: University of Minnesota Date: Tue, 11 May 1993 02:58:48 GMT Hello: I was trying to load a IB palette. For some reason, my IB cannot be launched anymore. Therefore, I checked into the NeXT.default and modified the file. The IB can be launched but all applications cannot save any default information anymore. Can anybody give me some hints on how to solve this problem? Ben
From: ratlifc@ctron.com (Christian A. Ratliff) Newsgroups: comp.sys.next.bugs Subject: compressing $HOME/.NeXT Followup-To: comp.sys.next.bugs Date: 11 May 1993 17:36:55 GMT Organization: Cabletron Systems, Inc. Distribution: world Message-ID: <ratlifc-110593133339@134.141.4.94> A few weeks ago I noticed something rather odd. If you compress your .NeXT directory, which I did by accident, when you uncompress it the files it places in the .NeXT directory are not those it tarred out. Very peculiar. I wouldn't call this a bug, excepting that you cannot _restore_ the old files. I was able to do it by zcat/tar from a terminal window. later, christian -------- Christian Ratliff Cabletron Systems, Inc. EDGE System Developer Rochester, NH 03867 ratlifc@ctron.com <NeXTmail OK> Work: (603) 337-1209 "I'm a NeXTSTEP man; I'm an SGI guy." Home: (207) 780-NeXT Nobody at Cabletron knows, approves of, or recalls my opinions.
Newsgroups: comp.sys.next.bugs From: dgursky@nextsrv1.andi.org (David M. Gursky) Subject: Netmasks Message-ID: <C6x6uG.81o@nextsrv1.andi.org> Keywords: Netmasks, Netinfo Sender: usenet@nextsrv1.andi.org (usenet) Organization: Association of NeXTSTEP Developers International Date: Wed, 12 May 1993 15:25:27 GMT I noticed the following oddity when we recently installed a NeXTStation on our local NeXT network. I had manually configured the netmask in /etc/hostconfig to 255.255.255.0, yet when this new station was powered-up, the netmask appeared to be 255.255.0.0 (derived through the use of the ifconfig command), even though the Netinfo tools and other manual inspections suggested it was 255.255.255.0. I solved this problem by changing /etc/hostconfig to pick up the Netmask automatically from the server, but I was wondering if other folks on the net had seen this behavior and had any thoughts as to why it happens?
Newsgroups: comp.sys.next.bugs From: francisr@stupid.ucs.indiana.edu (Rob Francis) Subject: Re: Netmasks Message-ID: <C6yz1K.BDM@usenet.ucs.indiana.edu> Keywords: Netmasks, Netinfo Sender: news@usenet.ucs.indiana.edu (USENET News System) Organization: Indiana University References: <C6x6uG.81o@nextsrv1.andi.org> Date: Thu, 13 May 1993 14:32:07 GMT In article <C6x6uG.81o@nextsrv1.andi.org> dgursky@nextsrv1.andi.org (David M. Gursky) writes: >I noticed the following oddity when we recently installed a NeXTStation on >our local NeXT network. I had manually configured the netmask in >/etc/hostconfig to 255.255.255.0, yet when this new station was >powered-up, the netmask appeared to be 255.255.0.0 (derived through the >use of the ifconfig command), even though the Netinfo tools and other >manual inspections suggested it was 255.255.255.0. I solved this problem >by changing /etc/hostconfig to pick up the Netmask automatically from the >server, but I was wondering if other folks on the net had seen this >behavior and had any thoughts as to why it happens? I had that happen too. Did the same work around. You wanna send it to bug_next, or me? (-: -rob
Newsgroups: comp.sys.next.bugs From: fm91hn@hik.se (HENRIK NORSELL) Subject: OPINION POLL! Message-ID: <fm91hn.45.737319566@hik.se> Sender: news@tintin.hik.se Organization: University of Kalmar Date: Thu, 13 May 1993 18:59:26 GMT Net citizens! This is a desperate try to save our last course in university. We are writing a study about the Net, how it all started, about the people living in it, however trying to explain the basics of how it all works. That includes you, reader of this message. We would be more than grateful if we could get your answers to the following questions; 1. For how many years have you known that Internet existed? 2. How often do you use the Net? (occasions per month) 3. Whatfor? (hobby, in your profession, socialy...) 4. How do you access the Net? (university, profession, friends, private...) 5. Has the Net taken over roles that other media played before? (telephone, newspapers, TV, girlfriend...) 6. What newsgroups/type of information do you take part of? 7. Male or Female? 8. Age? If you have the time; 9. What's your future visions about the Net? Limits and/or possibilities. 10.How do you think/hope law and censorship will change over time ahead? We also want to apologize for taking up so much bandwidth with this. This request has been spread to 60 newsgroups, chosen at random, but, you know how it is, term end is closing up, panic spreads. Email address: fm91hn@hik.se or fm91pb@hik.se Sincere Respect And May The Force Be With You All! Peter & Henrik
Newsgroups: comp.sys.next.bugs From: yann@dagobah.fdn.org (Yann Lechelle) Subject: Regarding Fax/language.lproj/Cover.draw Message-ID: <1993May14.115619.887@dagobah.fdn.org> Sender: yann@dagobah.fdn.org Organization: Individual Date: Fri, 14 May 1993 11:56:19 GMT How come the system creates a local ~/Fax/English.lproj folder in each directory ? It seems that the covers redesigned in this directory are not considered by the fax cover panel. It works if you put the covers in the /NextLibrary/Fax/English.lproj/... The pop-up shows all possible covers, with a neat thumb nail redraw. But everyone, individually, should be able to put his own cover. Is there a protection flag that would allow this ? (as it makes sense that this would be a normal behavior for companies that want to authorize users to only use their official cover sheet). Is this meant to be, or simply a bug ? Yann. -- / ================================================= \ \ Yann Lechelle (Paris, FRANCE) / / yann@dagobah.fdn.org Email/NeXTmail \ \ "grep me no patterns and I'll tell you no lines." /
Newsgroups: comp.sys.next.bugs From: dennis.glatting@mccaw.com Subject: Re: Netmasks Message-ID: <1993May16.010349.25444@mccaw.com> Keywords: sdsd Sender: news@mccaw.com Organization: McCaw Cellular Communications, Inc. References: <C6yz1K.BDM@usenet.ucs.indiana.edu> Distribution: usa Date: Sun, 16 May 1993 01:03:49 GMT In article <C6yz1K.BDM@usenet.ucs.indiana.edu> francisr@stupid.ucs.indiana.edu (Rob Francis) writes: > In article <C6x6uG.81o@nextsrv1.andi.org> dgursky@nextsrv1.andi.org (David M. Gursky) writes: > >I noticed the following oddity when we recently installed a NeXTStation on > >our local NeXT network. I had manually configured the netmask in > >/etc/hostconfig to 255.255.255.0, yet when this new station was > >powered-up, the netmask appeared to be 255.255.0.0 (derived through the > >use of the ifconfig command), even though the Netinfo tools and other > >manual inspections suggested it was 255.255.255.0. I solved this problem > >by changing /etc/hostconfig to pick up the Netmask automatically from the > >server, but I was wondering if other folks on the net had seen this > >behavior and had any thoughts as to why it happens? > > > I had that happen too. Did the same work around. You wanna send it > to bug_next, or me? (-: > > -rob No bug. RFC 1395. Something is returning a bad mask. It may be the NetInfo server or a ULTRIX box -- anything running bootpd on the subnet. NeXT's marked as NETMASTER=-YES- should have their subnet mask hard coded (along with IP address, etc). For this to work, specify NETMASK=-AUTOMATIC- -- and fix the bad bootpd host. If NETMASK isn't set then the it is derived from the host's network class. 255.255.0.0 would indicate your machine is a class B. -- Dennis P. Glatting / Technical Director / Paradigm Systems Corp. of WA
From: izumi@pinoko.berkeley.edu (Izumi Ohzawa) Newsgroups: comp.sys.next.bugs Subject: Re: Regarding Fax/language.lproj/Cover.draw Date: 16 May 1993 03:26:42 GMT Organization: University of California, Berkeley Distribution: world Message-ID: <1t4c9i$qd6@agate.berkeley.edu> References: <1993May14.115619.887@dagobah.fdn.org> In article <1993May14.115619.887@dagobah.fdn.org> yann@dagobah.fdn.org (Yann Lechelle) writes: >How come the system creates a local ~/Fax/English.lproj folder in each >directory ? It seems that the covers redesigned in this directory are not >considered by the fax cover panel. > >It works if you put the covers in the /NextLibrary/Fax/English.lproj/... >The pop-up shows all possible covers, with a neat thumb nail redraw. >But everyone, individually, should be able to put his own cover. /LocalLibrary/Fax/*.draw are shown in the pop-up list. That's where shared cover sheets should go. I believe that ~/Library/Fax/*.draw will also work for each user's customized cover sheet. Putting in English.lproj in these directories won't work. /NextLibrary/Fax/English.lproj/.... is special. This is documented somewhere, as I recall, but I can't find it now. -- Izumi Ohzawa [ 大澤五住 ] USMail: University of California, 360 Minor Hall, Berkeley, CA 94720 Telephone: (510) 642-6440 Fax: (510) 642-3323 Internet: izumi@pinoko.berkeley.edu (NeXTMail OK)
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: alex@hal.rhein-main.de (Alexander Lehmann) Subject: optimizer bug with gzip Content-Type: text/plain; charset=ISO-8859-1 Message-ID: <C72HF3.qJ@hal.rhein-main.de> Content-Transfer-Encoding: 8bit Mime-Version: 1.0 Date: Sat, 15 May 1993 12:01:49 GMT Hello all, I found a bug in gzip that occurs due to a bug with the optimizer of the NS3.0 cc. The bug makes gzip cause a segment violation when zcat'ing a file that was created on another OS (e.g. Messy). Apparently, gzip stores the original filename when it cannot be restored due to filesystem restrictions. When uncompressing to stdout, the filename is skipped with a simple while loop that in compiled incorrectly. As code piece that displays the bug is the following (more or less from gzip): unsigned char inbuf[1000]; unsigned inptr; unsigned insize; int fill_inbuf(void); #define get_byte() (inptr < insize ? inbuf[inptr++] : fill_inbuf()) main(void) { while(get_byte()!=0) ; } The resulting assembler code looks like this: #NO_APP .text .align 1 .globl _main _main: link a6,#0 movel _inptr,d1 cmpl _insize,d1 jcc L4 lea _inbuf,a1 movel _inptr,a0 jra L7 --- (1) L4: jbsr _fill_inbuf tstl d0 jeq L3 L2: movel _inptr,d1 cmpl _insize,d1 jcc L4 lea _inbuf,a0 movel _inptr,d0 L7: moveb a0@(d0:l),d0 --- (2) andl #0xFF,d0 addql #1,_inptr tstl d0 jne L2 L3: unlk a6 rts .comm _inbuf,1000 .comm _inptr,4 .comm _insize,4 When jumping to label L7 (1), the address of the char array is in a1 and the offset is in a0, but the move instruction (2) uses a0 and d0 as address. Since this register is not initialized, anything can happen. A slight change in the code makes it work: main(void) { char c; while((c=get_byte())!=0) ; } If the optimizer is not used, the code is also correct. I have fixed gzip.c as follows: *** ../gzip-1.0.7./gzip.c Thu Mar 18 19:14:56 1993 --- gzip.c Sat May 15 13:32:18 1993 *************** *** 1056,1062 **** if ((flags & ORIG_NAME) != 0) { if (to_stdout || part_nb > 1) { /* Discard the old name */ ! while (get_byte() != 0) /* null */ ; } else { /* Copy the base name. Keep a directory prefix intact. */ char *p = basename(ofname); --- 1056,1063 ---- if ((flags & ORIG_NAME) != 0) { if (to_stdout || part_nb > 1) { /* Discard the old name */ ! char c; /* fix for NeXTstep 3.0 cc optimizer bug */ ! while ((c=get_byte()) != 0) /* null */ ; } else { /* Copy the base name. Keep a directory prefix intact. */ char *p = basename(ofname); bye, Alexander Lehmann -- Alexander Lehmann, alex@hal.rhein-main.de | "Wopp" or alexlehm@iti.informatik.th-darmstadt.de | - ( THHGTTG Pt.3 )
Newsgroups: comp.sys.next.bugs From: harit@kripalu.com Subject: DBTableView resizing Message-ID: <1993May16.142509.11333@uunet!cbmvax!xmws!kripalu> Sender: harit@uunet!cbmvax!xmws!kripalu Organization: Kripalu Center Date: Sun, 16 May 93 14:25:09 GMT Has anyone noticed problems with resizing a DBTableView? Both in IB and at run time I have had it hang when I make it smaller. -- Michael Allen Latta Kripalu Center harit@kripalu.com (413)448-3288
Newsgroups: comp.sys.next.bugs,comp.sys.next.software From: mshaler@tdocad.sps.mot.com (Michael Shaler) Subject: How to hang Websters... Date: Mon, 17 May 1993 11:33:30 GMT Organization: Nippon Motorola Ltd., Tokyo, Japan Sender: news@tkymail.sps.mot.com Message-ID: <1993May17.113330.3203@tkymail.sps.mot.com> My limited vocabulary required a visit to the Thesaurus. Double click on "prospects," command-=, and whoah, nobody's home... "Prospect" is fine, as are others, but watch out elsewhere. (Also hangs when directly entered from Webster.app.) --- Michael Shaler Tokyo Design Center Nippon Motorola Ltd. +813 3280 8245 voice +813 3440 0033 fax mshaler@tdocad.sps.mot.com [NeXTmail]
Newsgroups: comp.sys.next.bugs,comp.sys.next.software From: timm@zaphod.ncsa.uiuc.edu (Tim McClarren) Subject: Re: How to hang Websters... Date: Mon, 17 May 1993 15:37:04 GMT Message-ID: <timm.737653024@zaphod> References: <1993May17.113330.3203@tkymail.sps.mot.com> Sender: usenet@news.cso.uiuc.edu (Net Noise owner) Organization: University of Illinois at Urbana mshaler@tdocad.sps.mot.com (Michael Shaler) writes: >My limited vocabulary required a visit to the Thesaurus. >Double click on "prospects," command-=, and whoah, nobody's home... >"Prospect" is fine, as are others, but watch out elsewhere. >(Also hangs when directly entered from Webster.app.) I wonder if this isn't some sort of harbinger that should have been discovered long ago. -- Tim McClarren | "...a bajillion brilliant Jobsian lithium licks." grug@shell.portal.com
From: Michael_Pizolato@afs.com Newsgroups: comp.sys.next.bugs Subject: Re: Regarding Fax/language.lproj/Cover.draw Message-ID: <1993May17.140833.268@afs.com> Date: 17 May 93 14:08:33 GMT Article-I.D.: afs.1993May17.140833.268 References: <1t4c9i$qd6@agate.berkeley.edu> Sender: Michael_Pizolato@afs.com In article <1t4c9i$qd6@agate.berkeley.edu> izumi@pinoko.berkeley.edu (Izumi Ohzawa) writes: > In article <1993May14.115619.887@dagobah.fdn.org> yann@dagobah.fdn.org (Yann Lechelle) writes: > > >How come the system creates a local ‾/Fax/English.lproj folder in each > >directory ? It seems that the covers redesigned in this directory are > >not considered by the fax cover panel. > > > >It works if you put the covers in the /NextLibrary/Fax/English.lproj/... > >The pop-up shows all possible covers, with a neat thumb nail redraw. > >But everyone, individually, should be able to put his own cover. > > /LocalLibrary/Fax/*.draw are shown in the pop-up list. That's > where shared cover sheets should go. > > I believe that ‾/Library/Fax/*.draw will also work for each user's > customized cover sheet. Putting in English.lproj in these directories > won't work. /NextLibrary/Fax/English.lproj/.... is special. ‾/Library/Fax/*.draw is correct. I have had a personal fax cover sheet there for many months now, and all is fine. Michael -- Michael_Pizolato@afs.com If you can't control your peanut ‾18 kyu butter, how can you expect to Q16 control the rest of your life? NeXTMail appreciated - Calvin
Newsgroups: comp.sys.next.bugs From: nico@imani.cam.org(Nicolas Dore) Subject: Freezing on logout Message-ID: <1993May18.022310.291@imani.cam.org> Sender: nico@imani.cam.org Date: Tue, 18 May 1993 02:23:10 GMT Subject has it all. On various occasions, upon logout, the machine freezes, and only the rom-monitor or nmi (whichever works, but only one of them does, I forget) can bring it down to a restart. I can't seem to notice any particular pattern, and since I have to reboot, I can't look at error messages in /tmp/logfile. Any idea? Is this an old one? If so, what's the fix? Thanks Nicolas Dore
From: aberno@godel.questor.wimsey.bc.ca (Anthony Berno) Newsgroups: comp.sys.next.bugs Subject: fmt manpage Message-ID: <1993May18.031046.2416@godel.questor.wimsey.bc.ca> Date: Tue, 18 May 1993 03:10:46 GMT Sender: aberno@godel.questor.wimsey.bc.ca The fmt manpage seems to be out of date. Terminal Services uses "fmt 58 60" to "wrap to 60 columns", but the man page makes no mention of numerical arguments. Does anyone care? -Anthony
From: sfurth@dc.shl.com (Stephen Furth) Newsgroups: comp.sys.next.bugs Subject: Re: Freezing on logout Date: 18 May 1993 12:29:24 GMT Organization: SHL Systemhouse Inc. Message-ID: <1takr4INNkd4@technet1.shl.com> References: <1993May18.022310.291@imani.cam.org> In article <1993May18.022310.291@imani.cam.org> nico@imani.cam.org(Nicolas Dore) writes: > Subject has it all. > > On various occasions, upon logout, the machine freezes, and only the rom-monitor > or nmi (whichever works, but only one of them does, I forget) can bring it down to > a restart. I can't seem to notice any particular pattern, and since I have to > reboot, I can't look at error messages in /tmp/logfile. > > Any idea? Is this an old one? If so, what's the fix? > > Thanks > > Nicolas Dore This is not a fix, but might help in the search... Why not try changing the /etc/rc* scripts so that /tmp is not emptied on boot-up and/or copy the /tmp/logfile file somewhere else? It will not solve the problem, but you would at least get the info to begin solving it. -- Stephen M. Furth | Systemhouse, Inc. Technical Architect | 1010 North Glebe Road Mid-Atlantic Region - DC Office | Arlington, VA 22201 Internet: sfurth@dc.shl.com | Voice: (703) 284-4800 x388 or: sfurth@shl.com | Fax: (703) 524-4782
Newsgroups: comp.sys.next.bugs From: yann@dagobah.fdn.org (Yann Lechelle) Subject: Re: Regarding Fax/language.lproj/Cover.draw Message-ID: <1993May17.231430.791@dagobah.fdn.org> Sender: yann@dagobah.fdn.org Organization: Individual References: <1t4c9i$qd6@agate.berkeley.edu> Date: Mon, 17 May 1993 23:14:30 GMT Thanks to Izumi Ohzawa, Robert Lutwak, and Art Isbell for their enlightening answers... THERE IS INDEED NO BUG WITH THE FAX COVERS. I don't know how I got messed up there... I must have copied simply what was found in /NextLibrary/Fax, including the erroneous English.lproj... Alles klar. Yann. -- / ================================================= ¥ ¥ Yann Lechelle (Paris, FRANCE) / / yann@dagobah.fdn.org Email/NeXTmail ¥ ¥ "grep me no patterns and I'll tell you no lines." /
Control: cancel <C72HF3.qJ@hal.rhein-main.de> Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: alex@hal.rhein-main.de (Alexander Lehmann) Subject: cancel <C72HF3.qJ@hal.rhein-main.de> Message-ID: <C78IzB.9B@hal.rhein-main.de> Date: Tue, 18 May 1993 18:21:11 GMT cancel <C72HF3.qJ@hal.rhein-main.de> -- Alexander Lehmann, alex@hal.rhein-main.de | "Wopp" or alexlehm@iti.informatik.th-darmstadt.de | - ( THHGTTG Pt.3 )
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: alex@hal.rhein-main.de (Alexander Lehmann) Subject: optimizer bug with gzip Content-Type: text/plain; charset=ISO-8859-1 Message-ID: <C78J2p.B8@hal.rhein-main.de> Content-Transfer-Encoding: 8bit Mime-Version: 1.0 Date: Tue, 18 May 1993 18:23:11 GMT [ Article crossposted from comp.sys.next.bugs,comp.sys.next.programmer ] [ Author was Alexander Lehmann (alex@hal.rhein-main.de) ] [ Posted on Sat, 15 May 1993 12:01:49 GMT ] [ It seems that the article didn't leave my machine, so I'm posting it again ] [ If you see this article twice, please forgive me, but my news system ] [ (semi)crashed when I posted it ] Hello all, I found a bug in gzip that occurs due to a bug with the optimizer of the NS3.0 cc. The bug makes gzip cause a segment violation when zcat'ing a file that was created on another OS (e.g. Messy). Apparently, gzip stores the original filename when it cannot be restored due to filesystem restrictions. When uncompressing to stdout, the filename is skipped with a simple while loop that in compiled incorrectly. As code piece that displays the bug is the following (more or less from gzip): unsigned char inbuf[1000]; unsigned inptr; unsigned insize; int fill_inbuf(void); #define get_byte() (inptr < insize ? inbuf[inptr++] : fill_inbuf()) main(void) { while(get_byte()!=0) ; } The resulting assembler code looks like this: #NO_APP .text .align 1 .globl _main _main: link a6,#0 movel _inptr,d1 cmpl _insize,d1 jcc L4 lea _inbuf,a1 movel _inptr,a0 jra L7 --- (1) L4: jbsr _fill_inbuf tstl d0 jeq L3 L2: movel _inptr,d1 cmpl _insize,d1 jcc L4 lea _inbuf,a0 movel _inptr,d0 L7: moveb a0@(d0:l),d0 --- (2) andl #0xFF,d0 addql #1,_inptr tstl d0 jne L2 L3: unlk a6 rts .comm _inbuf,1000 .comm _inptr,4 .comm _insize,4 When jumping to label L7 (1), the address of the char array is in a1 and the offset is in a0, but the move instruction (2) uses a0 and d0 as address. Since this register is not initialized, anything can happen. A slight change in the code makes it work: main(void) { char c; while((c=get_byte())!=0) ; } If the optimizer is not used, the code is also correct. I have fixed gzip.c as follows: *** ../gzip-1.0.7./gzip.c Thu Mar 18 19:14:56 1993 --- gzip.c Sat May 15 13:32:18 1993 *************** *** 1056,1062 **** if ((flags & ORIG_NAME) != 0) { if (to_stdout || part_nb > 1) { /* Discard the old name */ ! while (get_byte() != 0) /* null */ ; } else { /* Copy the base name. Keep a directory prefix intact. */ char *p = basename(ofname); --- 1056,1063 ---- if ((flags & ORIG_NAME) != 0) { if (to_stdout || part_nb > 1) { /* Discard the old name */ ! char c; /* fix for NeXTstep 3.0 cc optimizer bug */ ! while ((c=get_byte()) != 0) /* null */ ; } else { /* Copy the base name. Keep a directory prefix intact. */ char *p = basename(ofname); bye, Alexander Lehmann -- Alexander Lehmann, alex@hal.rhein-main.de | "Wopp" or alexlehm@iti.informatik.th-darmstadt.de | - ( THHGTTG Pt.3 )
Control: cancel <C78ILu.6J@hal.rhein-main.de> Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: alex@hal.rhein-main.de (Alexander Lehmann) Subject: cancel <C78ILu.6J@hal.rhein-main.de> Message-ID: <C78J6M.Cp@hal.rhein-main.de> Date: Tue, 18 May 1993 18:25:34 GMT cancel <C78ILu.6J@hal.rhein-main.de> -- Alexander Lehmann, alex@hal.rhein-main.de | "Wopp" or alexlehm@iti.informatik.th-darmstadt.de | - ( THHGTTG Pt.3 )
Newsgroups: comp.sys.next.bugs From: frank@fnbc.com (Frank Mitchell) Subject: Re: How to hang Websters... Message-ID: <1993May19.151718.754@fnbc.com> Sender: news@fnbc.com Organization: First National Bank Of Chicago, Chicago IL, USA References: <1993May17.113330.3203@tkymail.sps.mot.com> Date: Wed, 19 May 93 15:17:18 GMT In article <1993May17.113330.3203@tkymail.sps.mot.com> mshaler@tdocad.sps.mot.com (Michael Shaler) writes: > My limited vocabulary required a visit to the Thesaurus. > > Double click on "prospects," command-=, and whoah, nobody's home... > > "Prospect" is fine, as are others, but watch out elsewhere. > > (Also hangs when directly entered from Webster.app.) Huh? I tried both, and "prospects" doesn't produce any weird results. It's got to be something deeper. Did you check your dictionary files in /NextLibrary/References/*? What else were you doing to it before it crashed? -- Frank Mitchell email(work): frank@fnbc.com (NeXTmail) (home): frank@gagme.chi.il.us "That is one last thing to remember:_writers are always selling somebody out_." -- Joan Didion, Preface to _Slouching Toward Bethlehem_
Newsgroups: comp.sys.next.bugs Subject: Re: How to hang Websters... Message-ID: <1993May20.011139.24266@husc3.harvard.edu> From: jmcaulif@husc.harvard.edu (Jon McAuliffe) Date: 20 May 93 01:11:38 EDT References: <1993May19.151718.754@fnbc.com> Distribution: world In article <1993May19.151718.754@fnbc.com> writes: > In article <1993May17.113330.3203@tkymail.sps.mot.com> > mshaler@tdocad.sps.mot.com (Michael Shaler) writes: > > My limited vocabulary required a visit to the Thesaurus. > > > > Double click on "prospects," command-=, and whoah, nobody's home... > > > > "Prospect" is fine, as are others, but watch out elsewhere. > > > > (Also hangs when directly entered from Webster.app.) > > Huh? > > I tried both, and "prospects" doesn't produce any weird results. It's got to > be something deeper. Did you check your dictionary files in > /NextLibrary/References/*? What else were you doing to it before it crashed? "Prospects" caused no misbehavior under 3.0, Workspace v.341, PostScript v.81.16, on a 16/400 NSC. (Webster's v.3.0). Strange things afoot... _________________________________________________________________________ Jon McAuliffe jmcaulif@husc.harvard.edu NeXT ex-Campus Consultant 617.493.3004 (voice/fax) Harvard University
europa.eng.gtefsd.com!emory!ogicse!flop.ENGR.ORST.EDU!storm.cs.orst.edu!dockd From: dockd@storm.cs.orst.edu (Dion Dock) Newsgroups: comp.sys.next.bugs Subject: typo in a Webster's definition Message-ID: <1teqo6INNh30@flop.ENGR.ORST.EDU> Date: 20 May 93 02:34:46 GMT Article-I.D.: flop.1teqo6INNh30 Distribution: usa Organization: Computer Science Department, Oregon State University Take a look at the definition for 'empirical' and ask yourself what 'theor' means. Dion -- Dion Dock __ __ dockd@storm.cs.orst.edu / ) / ) / NeXT archive admin / / o ______ / / _____. /_ next-ftp@cs.orst.edu (no NeXTmail) /__/_<_(_) / <_ /__/_(_) (__/ <_
From: rfschtkt%banruc60.BITNET@ohstvma.acs.ohio-state.EDU Newsgroups: comp.sys.next.bugs Subject: Re: NXBrowser with NXImages in some Cells Date: 23 May 1993 03:28:51 -0400 Organization: The Ohio State University Sender: daemon@magnus.acs.ohio-state.edu Message-ID: <9305221845.AA01538@flexus> LS, Please change my report about NXBrowserCell's setImage: method and application crashing (Bug_NeXT #44209) to the following. Sorry for the confusion. NextDev/GeneralRef/02_ApplicationKit/Classes/NXBrowserCell.rtf NS_3.0-0.0.10.2.010 * NXBrowserCell The following methods have been added to NXBrowserCell to set an image that is displayed on the left side of the cell. The alternate image is used when the cell is hightlighted. - setImage:backgroundImage; - image; - setAltImage:newAltImage; - altImage; These methods aren't documented yet, although they are announced in ReleaseNotes/AppKit.rtf. Don't forget to mention that the NXBrowserCells won't make a copy of the NXImage, they will use the very instance provided in setImage:. In fact, the application will crash if the same NXImage is assigned to multiple NXBrowserCells, so a copy must be provided each time. Is this the same behaviour as for other setImage: methods? Raf Schietekat, RfSchtkt@banruc60.bitnet, Flanders, Belgium (real, i.e., with triangle in the Deliver button) NeXT Mail preferred I can't reach sites with ! or % in their address, or ending in ``at'' Nederlands, English, Francais, Deutsch a.o.
Newsgroups: comp.sys.next.bugs From: steve@eps.com Subject: test Message-ID: <netnewsC7Lwxx.8wv@netcom.com> Keywords: Netmasks, Netinfo Sender: netnews@netcom.com (USENET Administration) Organization: NETCOM On-line Communication Services (408 241-9760 guest) References: <C6x6uG.81o@nextsrv1.andi.org> Distribution: world Date: Tue, 25 May 1993 23:51:32 GMT test
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin From: brad@instep.wimsey.bc.ca Subject: /etc/hosts intro comment is bullshit... Lies! Lies! Lies! Message-ID: <1993May28.180158.389@instep.wimsey.bc.ca> Sender: brad@instep.wimsey.bc.ca (Bradley Head) Organization: InStep Mobile Communications Inc. Date: Fri, 28 May 1993 18:01:58 GMT I'm getting rather annoyed at NeXT and their ignorance of BSD UNIX. How is it that with each release of NEXTSTEP, NeXT manages to break one more UNIX tool? Do they have no shame? Case in point: The first few lines of the /etc/hosts file it says the following: # # NOTE: This file is never consulted if NetInfo or Yellow Pages is # running. # Life Rule Number 4,002,983: Never believe _never_. If you do, you'll be wasting your time and your companies time. As of Release 3.0, NeXT has yet to understand the fragile nature of UNIX. I've been trying all day to get the following syslog daemon configuration file (on host whiterock) to work: # # /etc/syslog.conf - on host whiterock. # # forward local1, and local2 to syslogd on host vancouver # local1.debug @vancouver local2.debug @vancouver When syslogd starts up it says it can't find host vancouver. Okay, why not? Well, it starts up as an early daemon, so perhaps netinfo isn't available at the time syslogd needs the host information. That would seem logical. So, I send syslogd a HUP signal which causes it to re-read its configuration file. There, that ought to take care of it... Nope! Still no logging is taking place on host vancouver. That's when I call bullshit to the _NOTE_ in /etc/hosts file. I enter the ip address and name of vancouver into /etc/hosts file, in this case : # # /etc/hosts - on host whiterock # 192.42.172.1 vancouver NOW, I send a HUP signal to syslogd on whiterock, and, well what do you know? - logging is now taking place. Syslogd on whiterock can now forward messages to the syslogd on host vancouver. Gee, that seems to conflict with the statement in /etc/hosts. Quit feeding us bullshit, NeXT! I wish NeXT would appreciate the POWER of UNIX and its collection of very useful tools, and learn to respect that POWER by leaving well enough alone what they do not fully understand. brad. -- Bradley Head Software Developer, InStep Mobile Communications Inc. brad@instep.wimsey.bc.ca (NeXTmail accepted) 604 872-7116 fax: 604 872-7125
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin From: eps@futon.SFSU.EDU (Eric P. Scott) Subject: Re: /etc/hosts intro comment is bullshit... Lies! Lies! Lies! Message-ID: <1993May29.123822.6621@csus.edu> Followup-To: comp.sys.next.sysadmin Sender: news@csus.edu Organization: San Francisco State University References: <1993May28.180158.389@instep.wimsey.bc.ca> Date: Sat, 29 May 1993 12:38:22 GMT In article <1993May28.180158.389@instep.wimsey.bc.ca> brad@instep.wimsey.bc.ca writes: >The first few lines of the /etc/hosts file it says the following: ># NOTE: This file is never consulted if NetInfo or Yellow Pages is ># running. >I've been trying all day to get the following syslog daemon configuration >file (on host whiterock) to work: >local1.debug @vancouver >local2.debug @vancouver >When syslogd starts up it says it can't find host vancouver. Okay, why >not? Well, it starts up as an early daemon, so perhaps netinfo isn't >available at the time syslogd needs the host information. That would seem >logical. If you look at /etc/rc, you'll see that syslogd starts up long before either NIS or NetInfo are available. So it's pretty obvious that any hosts referenced in /etc/syslog.conf need to be identified in /etc/hosts. > So, I send syslogd a HUP signal which causes it to re-read its >configuration file. There, that ought to take care of it... Nope! Hmm... >Still no logging is taking place on host vancouver. >That's when I call bullshit to the _NOTE_ in /etc/hosts file. Is lookupd running at that point? Does starting syslogd with the -d option produce any meaningful diagnostic output? >Quit feeding us bullshit, NeXT! I wish NeXT would appreciate the POWER of >UNIX and its collection of very useful tools, and learn to respect that >POWER by leaving well enough alone what they do not fully understand. Uh, yeah. What does Sun do? They start syslogd late, preceded by a comment that begins # syslogd doesn't belong here, but needs to be started before the others. No kidding! Not only are startup problems not logged, but this arguably introduces a potential security vulnerability. I just don't see any major problem--and in any case, there's nothing wrong with /etc/hosts, or its embedded comment. I could suggest NeXT eliminate any ambiguity by calling _old_gethostbyname() [or worse?] ... does it really make sense to compromise syslogd's self-sufficiency? syslogd is a can of worms. I doubt much good comes from opening it. And be careful what you wish for--it could go away and be replaced by "NEXTLOG with Portable Distributed Objects Support" which they'd later sell to a third party (and make you pay extra for). Anyway, this complaint sounds more like "pilot error" than anything else. -=EPS=-
Newsgroups: comp.sys.next.bugs From: hketola@agsm.ucla.edu (Heikki Ketola) Subject: In Workspace, no more Tools -> Processes !! How come? Message-ID: <hketola.738705315@uclagsm> Date: 29 May 93 13:01:18 PDT I run NS3.0 on a NeXTDimension. In Workspace, Tools -> Processes produces no Processes panel. Also, the alert panel doesn't show if you move files and the receiving folder contains files that have the same name(s) (the "Skip, Replace, Stop" alert panel). Rebooting does not help. I surmise that the panels' default coordinates have been accidentally set to somewhere outside the visible screen (in fact, to somewhere outside of even VirtSSpace's virtual screen!). Does anybody know where these default coordinates are stored (I couldn't find them with DefaultManager)?? I'd appreciate any suggestions on how to exorcise my machine. Heikki Ketola
From: thor@sushi.uib.no Newsgroups: comp.sys.next.bugs Subject: Re: Regarding Fax/language.lproj/Cover.draw Message-ID: <1993May29.202014.6891@alf.uib.no> Date: 29 May 93 20:20:14 GMT References: <1993May17.140833.268@afs.com> Sender: usenet@alf.uib.no (Bergen University Newsaccount) Organization: University of Bergen, Norway In article <1993May17.140833.268@afs.com> Michael_Pizolato@afs.com writes: > In article <1t4c9i$qd6@agate.berkeley.edu> izumi@pinoko.berkeley.edu (Izumi > Ohzawa) writes: > > In article <1993May14.115619.887@dagobah.fdn.org> yann@dagobah.fdn.org (Yann Lechelle) writes: > > > > >How come the system creates a local ‾/Fax/English.lproj folder in each > > >directory ? It seems that the covers redesigned in this directory are > > >not considered by the fax cover panel. > > > > > >It works if you put the covers in the /NextLibrary/Fax/English.lproj/... > > >The pop-up shows all possible covers, with a neat thumb nail redraw. > > >But everyone, individually, should be able to put his own cover. > > > > /LocalLibrary/Fax/*.draw are shown in the pop-up list. That's > > where shared cover sheets should go. > > > > I believe that ‾/Library/Fax/*.draw will also work for each user's > > customized cover sheet. Putting in English.lproj in these directories > > won't work. /NextLibrary/Fax/English.lproj/.... is special. > > ‾/Library/Fax/*.draw is correct. I have had a personal > fax cover sheet there for many months now, and all is fine. I have also tried this, but it has (up to now anyway) not yet worked. All I get is the standard (in English or Swedish - I created one in Norwegian but it refuses to show up) > > Michael Regards, Thor h
Newsgroups: comp.sys.next.bugs From: mrothste@keiko.acs.calpoly.edu (Rothstein) Subject: Need help selling NeXTSTEP to manager Message-ID: <1993May31.035720.135053@zeus.calpoly.edu> Sender: news@zeus.calpoly.edu Organization: Cal Poly State University, San Luis Obispo Date: Mon, 31 May 1993 03:57:20 GMT I am trying to convince my place of employment to go to NeXTSTEP and need help. I need any figures that show time savings of NeXTSTEP over DOS (I have the afs casestudy). Also if anyone has gone from a DOS system to NeXTSTEP and would be willing to talk to my manager he would like to talk to someone who has already done it. What is the best source for NeXTSTEP FIP (i.e. price)? NeXT, NeXTconnection, or somewhere else? We are currently using a relational database called Dataflex under DOS with a novell network. Our front end was written by another company and we are looking at $8000 to purchase the source code. Our network is 10 machines + 1 server, only five machines use the DB (we are considering moving only those machines that use the DB to NeXTSTEP). Six of the machines would need motherboard upgrades, all would need more memory and a svga video card. I appreciate any help or advice that anyone can give. -Mont NeXTmail OK :-) President CP-NUG (Cal Poly NeXT User Group, SLO) mrothste@data.acs.calpoly.edu
Newsgroups: comp.sys.next.bugs From: stabl@antigone.ppp.informatik.uni-muenchen.de (Robert Stabl) Subject: How to fix a severe NFS bug in sdmach for NeXTstep 3.0 Message-ID: <1993May31.161553.1676@antigone.ppp.informatik.uni-muenchen.de> Sender: stabl@antigone.ppp.informatik.uni-muenchen.de Organization: Institut fuer Informatik der Universitaet Muenchen Date: Mon, 31 May 1993 16:15:53 GMT Since we got many requests for more detailed information on the NFS bug in the sdmach kernel of NeXTstep 3.0 here is a repost of an article by Christian Baur <cbaur@informatik.uni-muenchen.de> about the NFS bug. We discovered this severe bug at Munich University when we got a SPARCstation 10/41 (running Solaris 2.1) as NFS file server for our NeXTs. Christian wrote: >jblue@ocotillo.mitre.org (Jason D. Blue) writes: >I remember reading at some point of a bug that appears when NeXT >writes to NFS mounted SUN devices. Can anyone provide more details >(i.e. what version of NEXTSTEP effected, patches, etc.) I posted this a while ago. We had this problem with NS3.0 (not tested with <3.0 or with SunOS). The patch below also did it for Mike Matthews : ========================================================================= matthews@oberon.umd.edu (Mike Matthews) writes: >We're trying to use a SPARC LX running Solaris 2.1 as our mail host to NeXT >and SunOS 4.x machines. I've got /var/mail exported to everybody, who's >mounting it as /usr/spool/mail, and that part seems to work OK. > >The problem is, even though the sendmail.cf file (which is a main.cf file >with the asked-for values) specifies mode 0600 for the files, they come out >as mode 660, and when NeXTmail gets it, it becomes mode 4777. Not very secure. >2) what's up with mail? I have no idea what's going on there. Ok, we had the same problems here with Solaris 2.1 as a file server for our NeXTStations. Doing a chown (and sendmail does a chown) yields in a wrong filemode for that file. All bits are set. After tracing the network traffic it was clear. Whats going on?? In NFS the attributes of a file are transferred in a structure called fattr. All entries in this structure which should not be modified in a set attributes NFS call (NFSPROC_SETATTR) have to be -1. This is also valid for the the vnode (virtual node) file mode structure (vattr) of the physical filesystem. The attribute structure is converted from the vnode attributes (vattr) to the NFS structure (sattr). In vattr the mode field is u_short in sattr it is u_long (32bit int's as god intended int's to be). Seems to be no problem. But the vattr_to_sattr routine in Mach 'converts' the -1 (16bit) to a -1 (also 16bit)... Here is a binpatch for the 3.0 mach kernel (NeXT Mach 3.0: Wed Jul 29 19:43:28 PDT 1992). I don't know if there is also a problem with 2.1, but I think a similiar patch will do it also for 2.1 [This patch ONLY applies to 3.0!!!]. [binpatch it with 'binpatch -f nfspatch.binpatch /sdmach' after making a backup from your /sdmach] nfspatch.binpatch: 256163 61 41 256166 2 0 [this patches a 'movew a2,a0@(0x2:w)' to a 'movel a2,a0@(0x0:w)'] binpatch is available from ftp.informatik.uni-muenchen.de in /pub/next/Tools/hack/. This patch works for us, but remember: We are not responsible for any damage you cause to your filesystem with this patch. You use this patch on your own risk!!! -- Robert Stabl email: stabl@informatik.uni-muenchen.de Computer Science Institute Tel: +(49) 89 2180 6316 University of Munich FAX: +(49) 89 2180 6310 Leopoldstr. 11B "Wer die Wahrheit nicht weiss, der ist bloss ein D-8000 Muenchen 40 Dummkopf. Aber wer sie weiss und sie eine Luege Germany nennt, der ist ein Verbrecher!" B.Brecht,"Galileo"
Newsgroups: comp.sys.next.bugs From: bkr@drdhh.hanse.de (Bjoern Kriews) Subject: IP addresses case sensitive on NeXTstep ? Message-ID: <1993May31.033430.1133@drdhh.hanse.de> Organization: Digital Island Date: Mon, 31 May 1993 03:34:30 GMT Am I the first one ever experiencing that $ ping myhost works but $ ping Myhost does not ? I experienced this bug under 2.1 but a friend I asked told me it's in 3.0 too. -- bkr@drdhh.hanse.de - Bjoern Kriews - Stormsweg 6 - D-2000 Hamburg 76 - FRG ------------------------------------------------------------------------------ Yesterday Moelln, Solingen today. Do we need tomorrow ? Yes, another one.
Newsgroups: comp.sys.next.bugs From: geelong@curacao.dartmouth.edu (Peter Kostelec) Subject: machine won't startup Message-ID: <C7x2K4.ox@dartvax.dartmouth.edu> Originator: geelong@curacao.dartmouth.edu Sender: news@dartvax.dartmouth.edu (The News Manager) Organization: Dartmouth College, Hanover, NH Distribution: usa Date: Tue, 1 Jun 1993 00:26:27 GMT Yesterday morning there was a brief power outage here. When I tried to startup my machine (a NeXTstation color turbo) that afternoon, it didn't start. Despite numerous attempts, the machine never got further than "checking system files ..." (as written in the little dialogue window). I turned the machine off and tried again this morning, without success. I left the machine "checking system files" for over 7 hours and it still didn't startup. Has anyone else ever had this problem before ? Can anything be done ? Thank you for any information. Peter Kostelec geelong@curacao.dartmouth.edu
Newsgroups: comp.sys.next.bugs From: hketola@agsm.ucla.edu (Heikki Ketola) Subject: No more Tools->Processes. Problem solved! Message-ID: <hketola.738947996@uclagsm> Date: 1 Jun 93 08:23:22 PDT A few days ago I posted that I have lost my Tools->Processes panel. Well, I found the panel. I dragged down the dock so that only the neXT icon was visible in the lower right hand corner. The Tools->Processes panel was barely visible at the right hand edge of the screen. I rescued my long-lost panel from there. Beats me how it got there in the first place. Thank to everyone for suggestions! Heikki Ketola
From: gary@ah3.cal.msu.edu (Gary J LaPointe) Newsgroups: comp.sys.next.bugs Subject: LAST command puked again! Date: 1 Jun 1993 17:21:24 GMT Organization: Michigan State University Message-ID: <1ug36k$k20@msuinfo.cl.msu.edu> Well, it's the beginning of the month and now when I do a LAST I just get garbage. This happenened last month too. Any suggestions? Last time, I just got the file (/usr/adm/wtmp, i think?) off a different machine and the first few entries were wrong but then it worked okay. Thanks! Gary -- Gary J LaPointe gary@ah3.cal.msu.edu Michigan State University Center For Integrative Studies, Arts & Humanities "There's nothing normal in life Except for the setting on your washing machine."
Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: kurt@frsvnsvn.cts.com (Kurt Werle) Subject: Novell Probs. (NCP errors) Message-ID: <1993Jun2.011758.11551@frsvnsvn.cts.com> Keywords: Novell NCP errors Organization: Little to None Date: Wed, 2 Jun 1993 01:17:58 GMT I've been having problems with a NeXT (turbo slab) connected to a Novell (3.11, Arcnet/10baseT) network. When using GNU tar, and other 'directory intensive' tasks the server gives errors: date: error User (username) has exceeded outstanding NCP directory search limit. Or something very close to that. Sometimes the NeXT just does not receive some of the directory information, and sometimes it hangs the connection to the Novell Network. The connection eventually comes back, though I'm not sure what triggers the return... Anyone have similar problems/advice on this one? Thanks, Kurt
From: maverick@halcyon.com (Thomas M. J. Fruchterman) Newsgroups: comp.sys.next.bugs Subject: next solaris nfs bug? Date: 1 Jun 1993 18:26:25 -0700 Organization: "A World of Information at your Fingertips" Sender: news@nwfocus.wa.com Message-ID: <1ugvk1$b6h@nwfocus.wa.com> I've just been reading about the NeXT mach kernel bug that interferes with interaction with Solaris machines. I'm having a different problem that I think may be related. Basically the opendir/readdir/closedir library suite gives different output for next mounted and solaris mounted filesystems. What follows is an illustrative program. run it on a next directory and then on a solaris directory. The d_namlen entry == strlen(d_name) on the next and strlen(d_name) (counting the null?) on the solaris. ------------------------------------------------------------ #include <sys/types.h> #include <sys/dir.h> void printDirInfo(char *dirname) { DIR *dirp; struct direct *dp; dirp = opendir(dirname); for (dp = readdir(dirp); dp != NULL; dp = readdir(dirp)) printf("d_name:%s d_namlen:%d strlen(d_name):%d.¥n", dp->d_name, dp->d_namlen, strlen(dp->d_name)); closedir(dirp); } int main(int argc, char **argv) { char *dir_name; dir_name = argv[1]; printDirInfo(dir_name); }
From: gad@eclipse.its.rpi.edu (Garance A. Drosehn) Newsgroups: comp.sys.next.bugs Subject: Sending compressed files via NeXTmail Date: 2 Jun 1993 02:59:02 GMT Organization: Rensselaer Polytechnic Institute, Troy NY, USA Distribution: world Message-ID: <1uh51m$apm@usenet.rpi.edu> Depending on your perspective this may not really seem like a bug, but I thought I'd write it up just so more people are aware of it. I know many people go thru the trouble of sending me compressed files (thinking it will reduce the transmission time), and this is a message I started to write up for one of them. Figured I might as well spread the word out a little more. While compressed files sound like an awfully good idea, they are pretty lousy things to send thru email. The problem is that NeXTmail itself wants to use tar and compress, and doing that to a tarred and compressed file makes the actual message (the # of characters going over the wire) significantly larger. Example: I take the file database.compressed, and I see that it's: (17) echoes/gad # ls -l database.compressed -rw-r--r-- 1 gad 554727 Apr 13 16:54 database.compressed I drag that into a message and send that as NeXTmail to myself. What ends up in my spool directory is: (18) echoes/gad # ls -l /usr/spool/mail/gad -rw------- 1 gad 1004165 May 16 00:13 /usr/spool/mail/gad Notice that it's increased the size by quite a bit. I then expand the original compressed file to get the directory "database": (19) echoes/gad # du -s database 2149 database I take that folder, and send that in an email message to myself (after deleting the previous message, of course!). The result is: (20) echoes/gad # ls -l /usr/spool/mail/gad -rw------- 1 gad 825355 May 16 00:19 /usr/spool/mail/gad Notice that the message actually sent when posting the plain folder is about 18% smaller than what you send by posting the compressed version of it. And the uncompressed folder is pretty easy to use (nothing new for a user to learn). If you're willing to go with something where the user does have to know a little bit more, there's always gzip. The people who are setup with gzip can use it with no trouble, but those who don't know about it would probably be annoyed by getting gzip'd files. Still, just for completeness sake, here's the result of the gzip example: (21) echoes/gad # gnutar cf database.tar database (22) echoes/gad # ls -l database.tar -rw------- 1 gad 2140160 May 16 00:20 database.tar (23) echoes/gad # /usr/local/gnu/bin/gzip -9 database.tar (24) echoes/gad # ls -l database.tar.gz -rw------- 1 gad 272932 May 16 00:20 database.tar.gz I sent that in a NeXTmail message, and got: (25) echoes/gad # ls -l /usr/spool/mail/gad -rw------- 1 gad 505939 May 16 00:27 /usr/spool/mail/gad -- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu ITS Systems Programmer (handles NeXT-type mail) Rensselaer Polytechnic Institute; Troy NY USA
Newsgroups: comp.sys.next.bugs From: sanguish@digifix.com (Scott Anguish) Subject: Re: Sending compressed files via NeXTmail Message-ID: <1993Jun2.054328.6336@digifix.com> Sender: sanguish@digifix.com (Scott Anguish) Organization: Digital Fix Development References: <1uh51m$apm@usenet.rpi.edu> Date: Wed, 2 Jun 1993 05:43:28 GMT Garance A. Drosehn writes > (17) echoes/gad # ls -l database.compressed > -rw-r--r-- 1 gad 554727 Apr 13 16:54 database.compressed > > I drag that into a message and send that as NeXTmail to myself. What > ends up in my spool directory is: > > (18) echoes/gad # ls -l /usr/spool/mail/gad > -rw------- 1 gad 1004165 May 16 00:13 /usr/spool/mail/gad It would appear that you are not taking into account that the file in your spool directory is also uuencoded. That does cause files to expand dramatically. A fair comparision would include uuencoding the original database.compressed file. -- - Scott Anguish - sanguish@digifix.com (NextMail) next-announce@digifix.com (comp.sys.next.announce submissions)
Newsgroups: comp.sys.next.bugs From: sanguish@digifix.com (Scott Anguish) Subject: Re: Sending compressed files via NeXTmail Message-ID: <1993Jun2.063133.6685@digifix.com> Sender: sanguish@digifix.com (Scott Anguish) Organization: Digital Fix Development References: <1993Jun2.054328.6336@digifix.com> Date: Wed, 2 Jun 1993 06:31:33 GMT Scott Anguish writes > It would appear that you are not taking into account that > the file in your spool directory is also uuencoded. That does cause files > to expand dramatically. A fair comparision would include uuencoding the > original database.compressed file. > > Garance was quick to point out that the comparison that I noted was not the one that he was trying to point out. Infact once he actually 'splained it fully, it does appear that there is an increase in size. I don't think that this could be considered a "bug" as proper, but it is definately something to consider when sending large NextMail attachments. Unfortunately gzip isn't the answer considering that it isn't standard, and that would make the transport of NextMail more akward. -- - Scott Anguish - sanguish@digifix.com (NextMail) next-announce@digifix.com (comp.sys.next.announce submissions)
From: maverick@halcyon.com (Thomas M. J. Fruchterman) Newsgroups: comp.sys.next.bugs Subject: Re: next solaris nfs bug? Date: 2 Jun 1993 11:17:14 -0700 Organization: "A World of Information at your Fingertips" Sender: news@nwfocus.wa.com Message-ID: <1uiqra$br@nwfocus.wa.com> References: <1ugvk1$b6h@nwfocus.wa.com> Arrgh. Screwed up my original posting. I meant to write that A. If you are logged on to a NeXT and use readdir on a directory that is mounted from a NeXT workstation, d_namlen == strlen(d_name). B. If you are on a NeXT, and you use readdir on a directory that is mounted via NFS from a Solaris 2.x system, d_namlen == strlen(d_name) + 1 Tom Fruchterman RAF Technology
Newsgroups: comp.sys.next.bugs From: kwang@data.acs.calpoly.edu (Kevin John Wang) Subject: Re: Novell Probs. (NCP errors) Message-ID: <1993Jun02.191346.198160@zeus.calpoly.edu> Sender: news@zeus.calpoly.edu Organization: Cal Poly State University, San Luis Obispo References: <1993Jun2.011758.11551@frsvnsvn.cts.com> Date: Wed, 02 Jun 1993 19:13:46 GMT In article <1993Jun2.011758.11551@frsvnsvn.cts.com> kurt@frsvnsvn.cts.com (Kurt Werle) writes: > I've been having problems with a NeXT (turbo slab) connected to a > Novell (3.11, Arcnet/10baseT) network. > When using GNU tar, and other 'directory intensive' tasks the server > gives errors: > > date: error User (username) has exceeded outstanding NCP directory > search limit. > > Or something very close to that. Sometimes the NeXT just does not > receive some of the directory information, and sometimes it hangs > the connection to the Novell Network. The connection eventually > comes back, though I'm not sure what triggers the return... > > Anyone have similar problems/advice on this one? I vaguely remember that there was a patch for this on the Novell side. If not, you can expand the "NCP dir search limit" on the server. Expand them to the point that you won't get any more errors, and all should be fine. It could very well be a NeXT bug, because what the NCP dir searches are, are "file open" requests, and I think that they might not close them immediately after they are done. If you need help, ask you Novell admin, or email me. (I am a Novell Admin 8> ) - Kevin Wang
Newsgroups: comp.sys.next.bugs From: eps@futon.SFSU.EDU (Eric P. Scott) Subject: @encode(long long) Message-ID: <1993Jun3.084357.372@csus.edu> Sender: news@csus.edu Organization: San Francisco State University Date: Thu, 3 Jun 1993 08:43:57 GMT This in partial response to the question someone e-mailed me "why does class-dump 1.2 blow up on Xanthus CraftMan?" It turns out to be a nasty ol' bug (verified for 1.0 and 2.1) that wasn't fixed in 3.0. /* cc -o longlong -object -s -O longlong.m */ #include <stdio.h> main() { (void)printf("@encode(int)=¥"%s¥"¥n", @encode(int)); (void)printf("@encode(long)=¥"%s¥"¥n", @encode(long)); (void)printf("@encode(long long)=¥"%s¥"¥n", @encode(long long)); (void)printf("@encode(unsigned int)=¥"%s¥"¥n", @encode(unsigned int)); (void)printf("@encode(unsigned long)=¥"%s¥"¥n", @encode(unsigned long)); (void)printf("@encode(unsigned long long)=¥"%s¥"¥n", @encode(unsigned long long)); exit(0); } [incorrect] output: @encode(int)="i" @encode(long)="l" @encode(long long)="" @encode(unsigned int)="I" @encode(unsigned long)="L" @encode(unsigned long long)="" Would someone with 3.1 for NeXT Computers (Motorola, not Intel) please post what the above test program produces? -=EPS=- -- Yes, there will be a class-dump 1.3 RSN.
From: charlie@snowflake.az.stratus.com Newsgroups: comp.sys.next.bugs Subject: Re: @encode(long long) Date: 3 Jun 1993 12:41:51 GMT Organization: Stratus Computer Inc, Marlboro MA Message-ID: <1ukrif$rvt@transfer.stratus.com> References: <1993Jun3.084357.372@csus.edu> In article <1993Jun3.084357.372@csus.edu> eps@futon.SFSU.EDU (Eric P. Scott) writes: > This in partial response to the question someone e-mailed me > "why does class-dump 1.2 blow up on Xanthus CraftMan?" > > It turns out to be a nasty ol' bug (verified for 1.0 and 2.1) > that wasn't fixed in 3.0. > @encode(int)="i" @encode(long)="l" @encode(long long)="" @encode(unsigned int)="I" @encode(unsigned long)="L" @encode(unsigned long long)="" -- Charles Spitzer charlie@snowflake.az.stratus.com Telecom Customer Assistance Center Stratus Computer, Inc. Phoenix, AZ 85018
Newsgroups: gnu.gcc.bug,comp.sys.next.bugs From: eps@futon.sfsu.edu (Eric P. Scott) Subject: @encode(long long), @encode(unsigned long long) Message-ID: <1993Jun4.004113.5887@csus.edu> Organization: San Francisco State University Date: Fri, 4 Jun 1993 00:41:13 GMT Apparently-To: gnu-gcc-bug@prep.ai.mit.edu Distribution: world Followup-To: gnu.gcc.bug,comp.sys.next.bugs These produce empty strings in NeXT's gcc. A quick peek at gcc-2.4.3/objc-act.c suggests that the same problem exists here as well [encode_type()]. Recommendation: Coordinate with NeXT and Stepstone to reserve two additional type characters--'q' and 'Q'--for 64-bit integers. -=EPS=-
Control: cancel <1993Jun2.011758.11551@frsvnsvn.cts.com> Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: kurt@frsvnsvn.cts.com (Kurt Werle) Subject: cmsg cancel <1993Jun2.011758.11551@frsvnsvn.cts.com> Message-ID: <1993Jun4.005940.5250@frsvnsvn.cts.com> Organization: Little to None References: <1993Jun2.011758.11551@frsvnsvn.cts.com> Date: Fri, 4 Jun 1993 00:59:40 GMT This message was cancelled from within rn.
Newsgroups: comp.sys.next.bugs From: thf@zelator.in-berlin.de (Thomas Funke) Subject: Re: Sending compressed files via NeXTmail Message-ID: <1993Jun3.102518.771@gamelan> Sender: thomas@gamelan (thomas) Organization: NNU Corp. - NeXT is Not UN*X References: <1993Jun2.054328.6336@digifix.com> Date: Thu, 3 Jun 1993 10:25:18 GMT In article <1993Jun2.054328.6336@digifix.com> sanguish@digifix.com (Scott Anguish) writes: > Garance A. Drosehn writes > > (17) echoes/gad # ls -l database.compressed > > -rw-r--r-- 1 gad 554727 Apr 13 16:54 database.compressed > > > > I drag that into a message and send that as NeXTmail to myself. What > > ends up in my spool directory is: > > > > (18) echoes/gad # ls -l /usr/spool/mail/gad > > -rw------- 1 gad 1004165 May 16 00:13 /usr/spool/mail/gad > > > > It would appear that you are not taking into account that > the file in your spool directory is also uuencoded. That does cause files > to expand dramatically. A fair comparision would include uuencoding the > original database.compressed file. While sending big files via NeXTMail is easy for unexperienced users, the REAL WAY :-) to do this is different: use 'gzip -9' for compression use 'btoa' AND NOT 'uuencode' This saves a lot of transmission time. NeXT should include this in future releases ! -- ------------------------------------------------------------------ Thomas Funke ** Unix-Consultant ** thf@zelator.in-berlin.de The nice thing about standards is that there are so many of them to choose from. -- Andrew S. Tanenbaum
From: robert@steffi.demon.co.uk (Robert Nicholson) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: strftime query? Date: 4 Jun 1993 19:16:10 +0100 Organization: me organized? That's a joke! Distribution: world Message-ID: <1uo3ha$3is@steffi.demon.co.uk> Ok call me stupid but aren't these two strftimes the same? strftime(buf,DATESIZE+1,"%d/%m/%Y¥t%H:%m:%M",timebuf); strftime(buf,DATESIZE+1,"%d/%m/%Y¥t%X",timebuf); Observe the following program. #include <sys/time.h> #define DATESIZE 19 void main(argc,argv) { time_t now,now2; char buf[DATESIZE+1]; struct tm *timebuf,*timebuf2; char buf2[DATESIZE+1]; now=time(NULL); timebuf = localtime(&now); strftime(buf,DATESIZE+1,"%d/%m/%Y¥t%H:%m:%M",timebuf); printf("%s¥n",buf); now2=time(NULL); timebuf2 = localtime(&now2); strftime(buf2,DATESIZE+1,"%d/%m/%Y¥t%X",timebuf2); printf("%s",buf2); exit(1); } Now, the second buf2 doesn't get properly null terminated, should it? (gdb) print buf $11 = {"04/06/1993¥t19:06:12¥000"} (gdb) print buf2 $12 = {"04/06/1993¥t19:12:33 "} (gdb) BTW: Black h/w and release 3.0 -- Real programmers don't create classes, they build hierarchies. (me) "If it doesn't compile and run emacs, it's not a computer." (Erik C. Sowa)
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: jfr@RedBrick.COM (Jon Rosen) Subject: Re: strftime query? Message-ID: <1993Jun5.030855.723@RedBrick.COM> Sender: usenet@RedBrick.COM (Net News Account) Organization: Red Brick Systems, Los Gatos, CA References: <1uo3ha$3is@steffi.demon.co.uk> Date: Sat, 5 Jun 1993 03:08:55 GMT In article <1uo3ha$3is@steffi.demon.co.uk> robert@steffi.demon.co.uk (Robert Nicholson) writes: >Ok call me stupid but aren't these two strftimes the same? >[munched most of program] > now2=time(NULL); > timebuf2 = localtime(&now2); > strftime(buf2,DATESIZE+1,"%d/%m/%Y¥t%X",timebuf2); > printf("%s",buf2); >Now, the second buf2 doesn't get properly null terminated, should it? > >(gdb) print buf >$11 = {"04/06/1993¥t19:06:12¥000"} >(gdb) print buf2 >$12 = {"04/06/1993¥t19:12:33 "} >(gdb) According to the ANSI C specificion, %X is a locale-specific time and it therefore not length-defined. You would have to find out what NeXT was doing with %X in order to know if you needed more than 19 characters to represent your format. I assume if might be that is it is putting out something like 19:12:33 PDT and may even be doing "bad" things to memory outside the array bounds. Make the buffer bigger and try it again. Jon Rosen
From: Jacob Gore <jacob@blackbox.enmu.edu> Newsgroups: comp.sys.next.bugs Subject: 3.0 /usr/lib/NextPrinter/Server/lpd.comm is hosed Date: 5 Jun 1993 23:57:12 GMT Organization: Eastern New Mexico University, Department of Mathematical Sciences Distribution: world Message-ID: <1urbsoINN3oa@almond.enmu.edu> Stumbled upon it while trying to print a man page using troff -t -man manpage.1 | lpr -t Looks like this one missed QA by a mile, folks... Hope it's been caught by 3.1 release. Could somebody check? Patch for /usr/lib/NextPrinter/Server/lpd.comm: *** lpd.comm.orig Fri Oct 25 13:33:20 1991 --- lpd.comm Sat Jun 5 17:44:12 1993 *************** *** 8,24 **** export PATH # Cannot pass file parameter through to prserver. It will be hooked up to stdin. ! while "x$1" != x do case "$1" in -f) FILE=$2; shift;; -p) PRINTER=$2; shift;; -n) USER=$2; shift;; ! -h) HOST=$"; shift;; *) ;; shift done ! PRSERVER="/usr/lib/NextPrinter/Server/prserver -p $PRINTER -n $USER -h HOST -f -" FILTER=`basename $0` case $FILTER in pstf) pscat | $PRSERVER;; --- 8,25 ---- export PATH # Cannot pass file parameter through to prserver. It will be hooked up to stdin. ! while [ "x$1" != x ]; do case "$1" in -f) FILE=$2; shift;; -p) PRINTER=$2; shift;; -n) USER=$2; shift;; ! -h) HOST=$2; shift;; *) ;; + esac shift done ! PRSERVER="/usr/lib/NextPrinter/Server/prserver -p $PRINTER -n $USER -h $HOST -f -" FILTER=`basename $0` case $FILTER in pstf) pscat | $PRSERVER;; Jacob --- Jacob Gore, Eastern NM U. Jacob@BlackBox.ENMU.Edu | Jacob@Gore.Com Member of the League for Programming Freedom (LPF)--lpf@uunet.uu.net
Newsgroups: comp.sys.next.bugs From: marcel@nice.usergroup.ethz.ch (Marcel Waldvogel) Subject: Re: Sending compressed files via NeXTmail Message-ID: <1993Jun11.142756.12502@bernina.ethz.ch> Sender: news@bernina.ethz.ch (USENET News System) Organization: NiCE - NeXT User Group, Zuerich References: <1993Jun2.054328.6336@digifix.com> <1993Jun3.102518.771@gamelan> Date: Fri, 11 Jun 1993 14:27:56 GMT In article <1993Jun3.102518.771@gamelan> thf@zelator.in-berlin.de (Thomas Funke) writes: >While sending big files via NeXTMail is easy for unexperienced users, the REAL >WAY :-) to do this is different: > > use 'gzip -9' for compression > use 'btoa' AND NOT 'uuencode' > >This saves a lot of transmission time. NeXT should include this in future >releases ! I _REAALY_ hope that NeXT will do this when switching to MIME (since the format then becomes totally incompatible with the old one, they can easily add another incompatibility...) -Marcel
Newsgroups: comp.sys.next.bugs From: uv94002@black.ox.ac.uk (Matt Millar) Subject: controlling terminals and rlogin Message-ID: <1993Jun13.094531.17187@black.ox.ac.uk> Originator: uv94002@black Organization: Oxford University Computing Services, Oxford, U.K. Date: Sun, 13 Jun 1993 09:45:31 GMT I have noticed a lack of controlling terminal when using rlogin from a DEC 5500 running Ultrix 4.3 to a NeXT (3.0) This doesn't occur when using the supplied csh, but does with our own compiled tcsh and bash. Also it doesn't happen with telnet! Anyone else experienced this? Anyone got any ideas for fixes? -- /¥/¥att /¥/¥illar uv94002@black.ox.ac.uk Full signature available on request in any of the following formats: TeX, PostScript, LaTeX, nroff, troff, groff, ascii, plot(5), rtf and handwritten.
From: izumi@pinoko.berkeley.edu (Izumi Ohzawa) Newsgroups: comp.sys.next.software,comp.sys.next.bugs Subject: Virtuoso 1.0 doesn't clip EPS at BoundingBox Date: 13 Jun 1993 11:20:10 GMT Organization: University of California, Berkeley Distribution: world Message-ID: <1vf2ha$7t7@agate.berkeley.edu> Keywords: Virtuoso 1.0, EPS, Clipping, BoundingBox Is there a way around the following bug in Virsutoso 1.0 for NeXT? When the original EPS file is cropped by making the size of %%BoundingBox smaller, Virtuoso 1.0 doesn't give me WYSIWYG for the imported EPS file. It crops correctly on the screen and only things within the %%BoundingBox is visible, but when the document is printed, objects outside the %%BoundingBox get printed. This makes it really difficult to edit my drawing because it totally destroys WYSIWYG. Am I missing something obvious? I don't want to be doing "paste inside". This is the only drawing app that I know of that does not crop at the BoundingBox. Thanks for any help. Izumi -- Izumi Ohzawa [ 大澤五住 ] USMail: University of California, 360 Minor Hall, Berkeley, CA 94720 Telephone: (510) 642-6440 Fax: (510) 642-3323 Internet: izumi@pinoko.berkeley.edu (NeXTMail OK)
From: izumi@pinoko.berkeley.edu (Izumi Ohzawa) Newsgroups: comp.sys.next.software,comp.sys.next.bugs Subject: Re: Virtuoso 1.0 doesn't clip EPS at BoundingBox Date: 14 Jun 1993 01:33:36 GMT Organization: University of California, Berkeley Distribution: world Message-ID: <1vgkhg$elf@agate.berkeley.edu> References: <1235@rtbrain.rightbrain.com> In article <1235@rtbrain.rightbrain.com> glenn@rightbrain.com (Glenn Reid) writes: Izumi Ohzawa writes >> >> When the original EPS file is cropped by making the size of >> %%BoundingBox smaller, Virtuoso 1.0 doesn't give me WYSIWYG for >> the imported EPS file. > >What do you mean "by making the size of %%BoundingBox smaller? Are >you editing it by hand? Of course it doesn't crop to the bounding >box if you change the bounding box. It's just a comment, after all. Yes, that's what I'm doing. It's just a comment all right, but in the world of PostScript, comments have become as important as the executable code itself. Things don't work without them. It somehow reminds me of a joke about a "comment compiler" which ignores all code and compiles the comments because they better reflect the intent of the programmer. >It's also impossible, by definition, to make the bounding box smaller >than the marks on the page. It is no longer a "bounding" box if it >doesn't bound the items on the page. > >Unless I'm misunderstanding something fundamental, there is no bug, >you're just making false assumptions. There is no reason, nor any >decree, to clip to the bounding box, since by definition there's >nothing outside the bounding box. Yes, there is a big reason: program's robustness to bad input. Because it does destroy WYSIWYG if the %%BoundingBox doesn't actually reflect the marking bounds of a drawing. On the screen, it will get cropped because the importing apps typically create an NXImage or cache of the size of %%BoundingBox. I only say the following, as a user, and will buy apps with features implemented as I want them. I expect NeXT apps to give me WYSIWYG even when imported EPS files are misbehaving. It is a matter of the robustness of a program to bad input. Clearly, in this case of Virtuoso, it fails to guarantee WYSIWYG for bad EPS, and that's bad -- bad enough to make me want to switch to another app that is more rubust, unless they will soon change the "feature" to the way I want it. So my big question now is: Does Adobe Illustrator clip imported EPS at the BoundingBox? Thanks. -- Izumi Ohzawa [ 大澤五住 ] USMail: University of California, 360 Minor Hall, Berkeley, CA 94720 Telephone: (510) 642-6440 Fax: (510) 642-3323 Internet: izumi@pinoko.berkeley.edu (NeXTMail OK)
Newsgroups: comp.sys.next.software,comp.sys.next.bugs From: lloyd@world.std.com (Chris Lloyd) Subject: Re: Virtuoso 1.0 doesn't clip EPS at BoundingBox Message-ID: <C8LGqo.G3s@world.std.com> Organization: The World Public Access UNIX, Brookline, MA References: <1235@rtbrain.rightbrain.com> <1vgkhg$elf@agate.berkeley.edu> Date: Mon, 14 Jun 1993 04:35:12 GMT In article <1vgkhg$elf@agate.berkeley.edu> izumi@pinoko.berkeley.edu writes: >In article <1235@rtbrain.rightbrain.com> glenn@rightbrain.com (Glenn Reid) writes: >Izumi Ohzawa writes >>> >>> When the original EPS file is cropped by making the size of >>> %%BoundingBox smaller, Virtuoso 1.0 doesn't give me WYSIWYG for >>> the imported EPS file. >> >>What do you mean "by making the size of %%BoundingBox smaller? Are >>you editing it by hand? Of course it doesn't crop to the bounding >>box if you change the bounding box. It's just a comment, after all. > >Yes, that's what I'm doing. It's just a comment all right, but >in the world of PostScript, comments have become as important as >the executable code itself. Things don't work without them. >It somehow reminds me of a joke about a "comment compiler" which >ignores all code and compiles the comments because they better >reflect the intent of the programmer. >[...] Page 725 (Guidelines for Importing EPS Files) of the Level 2 Red Book, paragraph "Set Up a Clipping Path". The importing application should set up a clipping path around the imported EPS file. This can be accomplished by setting a clipping path that corresponds to the bounding box of the imported EPS file after making the PostScript coordinate system transformations or by allowing the user to optionally supply an arbitrary clipping path for special effects. It then goes on to describing what to inject in the print stream, and it mentions clipping path again... Shrug, -- :: Christopher Lloyd :: Yrrid Incorporated :: lloyd@world.std.com :: :: I'm sure I said something wrong, so go ahead, flame away! ::
From: foster@seismo.CSS.GOV (Glen Foster) Newsgroups: comp.sys.next.software,comp.sys.next.bugs Subject: Re: Virtuoso 1.0 doesn't clip EPS at BoundingBox Date: 14 Jun 1993 11:37:01 GMT Organization: Center for Seismic Studies, Arlington, VA Distribution: world Message-ID: <1vhnst$reg@seismo.CSS.GOV> References: <1235@rtbrain.rightbrain.com> <1vgkhg$elf@agate.berkeley.edu> In article <1vgkhg$elf@agate.berkeley.edu> izumi@pinoko.berkeley.edu writes: >So my big question now is: >Does Adobe Illustrator clip imported EPS at the BoundingBox? Illustrator ver. 3.0.1 does not. It does, however, use the Bounding Box information as the handles for the imported image. The usefuleness of this is marginal. It appears at first blush that the image is cropped but any manipulation, including de-selection, reveals the entire image. I edited the BB information by hand to ascertain this. I have a basic understanding of PS but I am no guru. Glen Foster foster@css.gov
Newsgroups: comp.sys.next.bugs From: nico@imani.cam.org (Nicolas Dore) Subject: Strange Workspace behavior Message-ID: <1993Jun14.122825.6044@imani.cam.org> Sender: nico@imani.cam.org Date: Mon, 14 Jun 1993 12:28:25 GMT Hi, folks! Well, my WM seems to have a mind of it's own. Up until a few moments ago, I _could not_ drag windows. If I clicked on the Title bar, the window would go to the Background. Clicking elsewhere would make it come back, but I still wouldn't be able to move it! I could hide and make visible all windows, click buttons and such (I started this post under this behavior), but not move windows. Now, without rebooting or even logging out, it's back to normal. Everything works A-OK. Kind of scary thought. When is 3.2 shipping again? Ciao -- Nicolas Dore nico@imani.cam.org - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - >"Si la classe ouvriere ne nous satisfait plus, nous la destituerons< > et en elirons une autre" B. Brecht (ne parlans pas d'informatique)<
From: alex@cs.umd.edu (Alex Blakemore) Newsgroups: comp.sys.next.software,comp.sys.next.bugs Subject: Re: Virtuoso 1.0 doesn't clip EPS at BoundingBox Message-ID: <68467@mimsy.umd.edu> Date: 14 Jun 93 16:58:59 GMT References: <1235@rtbrain.rightbrain.com> <1vgkhg$elf@agate.berkeley.edu> Sender: news@mimsy.umd.edu Followup-To: comp.sys.next.software Organization: U of Maryland, Dept. of Computer Science, Coll. Pk., MD 20742 [regarding whether an app should clip to the bounding box of an EPS file] Glenn Reid (or perhaps an imposter 8^) : >> There is no reason, nor any decree, to clip to the bounding box, >> since by definition there's nothing outside the bounding box. Izumi Ohzawa: > Yes, there is a big reason: program's robustness to bad input. The red book (2nd ed) _recommends_ that apps clip EPS files at the bounding box. see appendix H.3 "Guidelines for Importing EPS Files" -- Alex Blakemore alex@cs.umd.edu NeXT mail accepted -------------------------------------------------------------- "Without an engaged and motivated human being at the keyboard, the computer is just another dumb box." William Raspberry
Newsgroups: comp.sys.next.software,comp.sys.next.bugs From: lorinr@altsys.com (Lorin Rivers III) Subject: Re: Virtuoso 1.0 doesn't clip EPS at BoundingBox Message-ID: <1993Jun14.164933.684@altsys.com> Organization: Altsys Corporation, Richardson, TX References: <1235@rtbrain.rightbrain.com> <1vgkhg$elf@agate.berkeley.edu> Date: Mon, 14 Jun 1993 16:49:33 GMT To the world: I'm trying to find out what Izumi needs here. He is the only person to mention this as a problem. Virtuoso does parse EPS to prevent "bad" EPSs from being placed into Virtuoso. When I find out what the deal -- Lorin Rivers Lorin_Rivers@altsys.com Altsys Technical Support 214.680.2518 269 W. Renner Parkway NeXT Mail Expected Richardson, Texas 75080 I said it, not my boss
From: annard@theborg.stack.urc.tue.nl (Annard Brouwer) Newsgroups: comp.sys.next.bugs Subject: Re: Strange Workspace behavior Date: 14 Jun 1993 19:21:23 GMT Organization: the Borg Distribution: world Message-ID: <1vij3jINNd5@theborg.stack.urc.tue.nl> References: <1993Jun14.122825.6044@imani.cam.org> In article <1993Jun14.122825.6044@imani.cam.org> nico@imani.cam.org (Nicolas Dore) writes: > Well, my WM seems to have a mind of it's own. Up until a few > moments ago, I _could not_ drag windows. If I clicked on the > Title bar, the window would go to the Background. Clicking > elsewhere would make it come back, but I still wouldn't be able > to move it! I could hide and make visible all windows, click > buttons and such (I started this post under this behavior), but > not move windows. > > Now, without rebooting or even logging out, it's back to normal. > Everything works A-OK. Kind of scary thought. Well, what program(s) did you run before this happened? In my experience, this behaviour is to be expected when you mess around with the WindowServer. It happened to me when I was hacking some ugly postscript code... Good luck, Annard -- Annard Brouwer annard@stack.urc.tue.nl (NeXTmail appreciated) People? You can forget it.
Newsgroups: comp.sys.next.bugs From: Karl_Kraft@ensuing.com Subject: Re: Strange Workspace behavior Message-ID: <C8MrG8.111@ensuing.com> Sender: karl@ensuing.com (Karl Kraft) Organization: Ensuing Technologies Inc. References: <1993Jun14.122825.6044@imani.cam.org> Date: Mon, 14 Jun 1993 21:24:12 GMT In article <1993Jun14.122825.6044@imani.cam.org> nico@imani.cam.org (Nicolas Dore) writes: > Hi, folks! > > Well, my WM seems to have a mind of it's own. Up until a few moments ago, I _could > not_ drag windows. If I clicked on the Title bar, the window would go to the > Background. Clicking elsewhere would make it come back, but I still wouldn't be This is the result of the Window Server not detecting that you have released the command key. Just press and let go of the command key to return things to normal. In case you didn't know, command-clicking a window's title bar sends it to back but leaves it the key window. -- _________ Karl Kraft Karl_Kraft@ensuing.com (NeXT mail)
Newsgroups: comp.sys.next.bugs From: djc@puck.fnbc.com (Dan Crimmins) Subject: Re: Strange Workspace behavior In-Reply-To: nico@imani.cam.org's message of Mon, 14 Jun 1993 12:28:25 GMT Message-ID: <DJC.93Jun14163821@puck.fnbc.com> Sender: news@fnbc.com Organization: First National Bank Of Chicago, Chicago, IL References: <1993Jun14.122825.6044@imani.cam.org> Date: 14 Jun 93 16:38:21 nico@imani.cam.org (Nicolas Dore) writes: > Well, my WM seems to have a mind of it's own. Up until a few moments ago, I _could > not_ drag windows. If I clicked on the Title bar, the window would go to the > Background. Clicking elsewhere would make it come back, but I still wouldn't be > able to move it! I could hide and make visible all windows, click buttons and such > (I started this post under this behavior), but not move windows. this is the behavior you would expect if your command key was stuck down. > Now, without rebooting or even logging out, it's back to normal. Everything works > A-OK. Kind of scary thought. likewise, the problem goes away if the command key is released. you may want to see if your keyboard needs cleaning. --dan. -- dan crimmins djc@fnbc.com
From: dkk@cbnewsl.cb.att.com (david.k.kallman) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs Subject: 3.1 Mailer and symbolic links Message-ID: <C8n9By.5wE@cbnewsl.cb.att.com> Date: 15 Jun 93 03:50:18 GMT Followup-To: comp.sys.next.sysadmin Organization: AT&T Bell Laboratories Folks, Just upgraded to 3.1 for Next hardware and am having a problem with mail. My mail actually goes to a Sun system, and I have /var/spool/mail on the Sun mounted as /Net/sun/var/spool/mail on the NeXT. (There are several users on the NeXT and most of them get mail locally on the NeXT so I couldn't just mount /var/spool/mail from the Sun on /usr/spool/mail on the NeXT.) Now Mail (note capital M), autoretrieves mail from the Sun, but does not delete the mail on the Sun after autoretrieving it. So every autoretrieve period the mail just keeps getting duplicated. The worked fine in 3.0, and also works fine with mail (small m, i.e., deleting the mail and quiting in mail). Has anyone experienced this? Also would the problem go away if I rearranging things and mounted /var/spool/mail on the Sun on /usr/spool/mail on the NeXT? I'd experiment, but I'd prefer not taking the chance of losing mail. Thanks. Dave Kallman -- ---- Dave Kallman, AT&T, 480 Red Hill Rd., Middletown, NJ 07748 d_k_kallman@att.com, (908)615-2989, fax: (908)615-2507
Control: cancel <C8n9By.5wE@cbnewsl.cb.att.com> Newsgroups: comp.sys.next.sysadmin,comp.sys.next.bugs From: dkk@cbnewsl.cb.att.com (david.k.kallman) Subject: cmsg cancel <C8n9By.5wE@cbnewsl.cb.att.com> Organization: AT&T Date: Tue, 15 Jun 1993 03:57:33 GMT Message-ID: <C8n9ny.62A@cbnewsl.cb.att.com> References: <C8n9By.5wE@cbnewsl.cb.att.com> Sender: dkk@cbnewsl.cb.att.com (david.k.kallman) <C8n9By.5wE@cbnewsl.cb.att.com> was cancelled from within rn. -- ---- Dave Kallman, AT&T, 480 Red Hill Rd., Middletown, NJ 07748 d_k_kallman@att.com, (908)615-2989, fax: (908)615-2507
From: dkk@cbnewsl.cb.att.com (david.k.kallman) Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin Subject: 3.1 Mailer and symbolic links Message-ID: <C8nA5L.6B2@cbnewsl.cb.att.com> Date: 15 Jun 93 04:08:08 GMT Followup-To: comp.sys.next.sysadmin Organization: AT&T Bell Laboratories Folks, Just upgraded to 3.1 for Next hardware and am having a problem with mail. My mail actually goes to a Sun system, and I have /var/spool/mail on the Sun mounted as /Net/sun/var/spool/mail on the NeXT, with a symbolic link from /usr/spool/mail/dkk -> /Net/sun/var/spool/mail/dkk, and also a .forward in my home directory on the Next to dkk@sun - this handles local mail. (There are several users on the NeXT and most of them get mail locally on the NeXT so I didn't just mount /var/spool/mail from the Sun on /usr/spool/mail on the NeXT.) Now Mail.app (note capital M) autoretrieves mail from the Sun, but does not delete the mail on the Sun after autoretrieving it. So every autoretrieve period the mail just keeps getting duplicated. The worked fine in 3.0, and also works fine in 3.1 with mail (small m, i.e., deleting the mail and quiting in mail). Has anyone experienced this? Also would the problem go away if I rearranged things and mounted /var/spool/mail on the Sun on /usr/spool/mail on the NeXT? I'd experiment, but I'd prefer not taking the chance of losing mail. Thanks. Dave Kallman -- ---- Dave Kallman, AT&T, 480 Red Hill Rd., Middletown, NJ 07748 d_k_kallman@att.com, (908)615-2989, fax: (908)615-2507
Newsgroups: comp.sys.next.bugs From: nico@imani.cam.org (Nicolas Dore) Subject: Re: Strange Workspace behavior Message-ID: <1993Jun15.025140.7634@imani.cam.org> Sender: nico@imani.cam.org References: <1vij3jINNd5@theborg.stack.urc.tue.nl> Date: Tue, 15 Jun 1993 02:51:40 GMT In article <1vij3jINNd5@theborg.stack.urc.tue.nl> annard@theborg.stack.urc.tue.nl (Annard Brouwer) writes: > In article <1993Jun14.122825.6044@imani.cam.org> nico@imani.cam.org (Nicolas > Dore) writes: > > Well, my WM seems to have a mind of it's own. Up until a few > > moments ago, I _could not_ drag windows. If I clicked on the > > Title bar, the window would go to the Background. Clicking > > elsewhere would make it come back, but I still wouldn't be able > > to move it! I could hide and make visible all windows, click > > buttons and such (I started this post under this behavior), but > > not move windows. > > > > Now, without rebooting or even logging out, it's back to normal. > > Everything works A-OK. Kind of scary thought. > > Well, what program(s) did you run before this happened? In my experience, this > behaviour is to be expected when you mess around with the WindowServer. It > happened to me when I was hacking some ugly postscript code... Actually, I wasn't doing much. From answers I've received, it seems the command-key got stuck or something. Holding it down and trying to grab windows reproduces the problem. Still don't know what caused that, though. Ciao, and thanks > Good luck, > Annard > -- > Annard Brouwer annard@stack.urc.tue.nl (NeXTmail appreciated) > People? You can forget it. -- Nicolas Dore nico@imani.cam.org - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - >"Si la classe ouvriere ne nous satisfait plus, nous la destituerons< > et en elirons une autre" B. Brecht (ne parlans pas d'informatique)<
Newsgroups: comp.sys.next.bugs From: meyergru@Informatik.TU-Muenchen.DE (Uwe Meyer-Gruhl) Subject: CC error in NS 3.0 Keywords: CC 1.93 Followup-To: comp.sys.next.bugs Originator: meyergru@hpradigc.informatik.tu-muenchen.de Sender: news@Informatik.TU-Muenchen.DE (USENET Newssystem) Organization: Technische Universitaet Muenchen, Germany Date: Tue, 15 Jun 1993 09:30:47 GMT Message-ID: <1993Jun15.093047.9210@Informatik.TU-Muenchen.DE> Hi Folks, I stumbled upon this one last week while trying to adapt 'hpcdtoppm 0.5' on my NeXTstation. No need to inform the author, I have done so already. Everything works fine with newer versions of GCC, by the way. cheers, Uwe Uwe Meyer-Gruhl Confucius say: Lehrstuhl Informatik IX "Unlucky to play leapfrog with unicorn" Technische Universitaet Muenchen email:MeyerGru@Informatik.TU-Muenchen.DE Orleansstr. 34, D-8000 Muenchen 80 tel: ++49 89 48095-209 --- /* This program yields three different outputs, depending on the optimization switches used. (Results given for NeXTSTEP 3.0) cc test.c -> signed, signed (correct) cc -O test.c -> unsigned, unsigned (wrong) cc -O2 test.c -> unsigned, signed (wrong) */ #include <stdio.h> int main(void) { char c; int i; c = 126; c++; c++; if (c > 126) printf("unsigned.¥n"); else printf("signed.¥n"); i = c > 126; if (i) printf("unsigned.¥n"); else printf("signed.¥n"); return 0; }
From: annard@theborg.stack.urc.tue.nl (Annard Brouwer) Newsgroups: comp.sys.next.bugs Subject: Re: Strange Workspace behavior Date: 15 Jun 1993 18:43:30 GMT Organization: the Borg Distribution: world Message-ID: <1vl58iINN8u@theborg.stack.urc.tue.nl> References: <1993Jun15.025140.7634@imani.cam.org> In article <1993Jun15.025140.7634@imani.cam.org> nico@imani.cam.org (Nicolas Dore) writes: > Actually, I wasn't doing much. From answers I've received, it seems the > command-key got stuck or something. Holding it down and trying to grab windows > reproduces the problem. Still don't know what caused that, though. > Well, this was not my problem since I couldn't move icons anymore (and this can also be done with a stuck command key). So I do not believe the "stuck command key" solution completely. Then again, I'm famous for finding weird bugs... :-))) Good luck, Annard -- Annard Brouwer annard@stack.urc.tue.nl (NeXTmail appreciated) People? You can forget it.
Newsgroups: comp.sys.next.bugs From: greyham@research.canon.oz.au (Graham Stoney) Subject: Re: Strange Workspace behavior Message-ID: <C8oy2o.LK0@research.canon.oz.au> Sender: news@research.canon.oz.au Organization: Canon Information Systems Research Australia References: <1993Jun14.122825.6044@imani.cam.org> <C8MrG8.111@ensuing.com> Date: Wed, 16 Jun 1993 01:42:23 GMT Karl_Kraft@ensuing.com writes: >In article <1993Jun14.122825.6044@imani.cam.org> nico@imani.cam.org >(Nicolas Dore) writes: >> Hi, folks! >> >> Well, my WM seems to have a mind of it's own. Up until a few moments >ago, I _could >> not_ drag windows. If I clicked on the Title bar, the window would go to >the >> Background. Clicking elsewhere would make it come back, but I still >wouldn't be >This is the result of the Window Server not detecting that you have >released the command key. Just press and let go of the command key to >return things to normal. In case you didn't know, command-clicking a >window's title bar sends it to back but leaves it the key window. ... and note that bringing up the NMI monitor is one cause. Graham -- Graham Stoney Canon Information Systems Research Australia Ph: + 61 2 805 2909
From: sears@tree.egr.uh.edu (Paul S. Sears) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.hardware,comp.sys.next.bugs Subject: [long]New Installation solutions for Gateway 2000 4DX2-66V with X30 Chip Set (recent models) Date: 16 Jun 1993 23:06:44 GMT Organization: University of Houston Message-ID: <1vo924$hnk@menudo.uh.edu> Previously I had made a posting about the problems we were having with installing NEXTSTEP/fip on some Gateway 2000 4DX2-66V machines. Specifically, we were having problems with the keyboards not responding (all types of keyboards). More specifically, this relates to Gateways with the revised Micronics System Boards u).sing the X30 Chip Set (Phoenix "80486 Rom BIOS Plus ver 0.10 GJX30-01d". Below are my notes for finally getting everything to work. These notes will be forwarded to NeXT, Inc. and Gateway 2000 as it seems to have something to do with a possible timing problem with the PS2Keyboard driver. NeXT and Gateway 2000 Tech Support have been outstanding and even though they couldn't solve the problem, they tried everything they could (NeXT even got a hold of one of the newer Gateway 2000 machines to try to solve the problem). Anyway, these notes explain installation problems and solutions that _should_ work, but there is no promise. Please read them carefully. I am sorry if they ramble on at times... :-) ************************************************************************* Gateway 2000 4DX2-66V Installation Notes (Please read carefully!) [Revised notes are at the bottom of the text. I have included all my notes so that others may possibly benefit from them or figure some more information out...] For Installation: Short Answer: Do not boot the system with the keyboard plugged in!!! If you are installing, then plug the keyboard in _only_ when it asks you questions like what language, what partition, etc. After the initial "mini" install, you can complete the entire installation with just the mouse plugged into COMA (COM1). Once you have NEXTSTEP installed and working, and then you do the final reboot into NEXSTEP, you can plug in the keyboard (In other words, don't plug in the keyboard until you are logged into the "me" account. Long Answer: Problem: NeXTSTEP/fip will not install with the keyboard plugged in while the machine is booting. This seems to be because of the conflict with the BIOS. Problem occurs during the following part of the boot process (using the Gateway.config): ISA bus Registering: PCKeyboard0 <- PS2Keyboard Registering: PCPointer0 <- PS2Mouse It was suggested that there was a possible IRQ/Device conflict between the keyboard and the Microsoft Serial mouse. The mouse and keyboard were moved to different interrupt vectors (IRQ 3 and IRQ5 respectively) but the problem still persisted. The following things were attempted to remedy the problem: Changed BIOS settings: The final decided BIOS settings were: Diskette A: 3.5", 1.44 MB Diskette B: Not Installed Cyl Hd Pre LZ Sec Size Hard Disk 1: Auto Config 1010 12 0 0 55 325 Hard Disk 2: Not Installed Base Memory: 640 KB HD1 Block Mode: Enable Extended Memory: 15360 KB HD2 Block Mode: Disable Video Card: VGA/EGA IDE Speed: Fast Keyboard: Not Installed System BIOS Shadow: Disable Internal Cache: Enable Video BIOS Shadow: Disable External Cache: Enable E8000h 32K Shadow : Disable F0000h-FFFFFh : Non Cacheable E0000h 32K Shadow : Disable C0000h-C7000h : Non Cacheable D8000h 32K Shadow : Disable On board COMA : COM1 D0000h 32K Shadow : Disable COMA_IRQ : IRQ-4 C8000h 32K Shadow : Disable On board COMB : Disabled Turbo : Enable COMB_IRQ : Disabled Bus Clock : 1/4 Clock On board LPT : Disabled Processor Type: 80486 Coprocessor Type: 487 or compatible Option ROMs Found: Reserved Memory: 3072k Segment Size BIOS Version #: 0.10 C000 32 k Video Mode: 03h DC00 16 k Serial Ports: 03F8h 02F8h Printer Ports: 03BCh The system could be used only if the following was done: 1) Disconnect the keyboard and boot the system 2) Connect the keyboard when prompted for question or need to log in This problem does not effect the DOS partition in any way. DOS can be used normally. More specifically, the keyboard works under NEXTSTEP for the initial boot process (i.e, it can be used to select partition to boot and for NEXTSTEP boot parameters). After it registers the PCKeyboard0 and PS2Mouse devices, the keyboard hangs, even if the keyboard is immediately disconnected after a response is entered in the boot manager. The only way to that NEXTSTEP will boot up and the keyboard will work under NEXTSTEP is to make sure the keyboard is disconnected during the entire boot phase. Probable cause for problem: NeXT had certified the Gateway 2000 systems to be "NEXTSTEP Compatible." The system that was certified was using a Phoenix BIOS with the MCM2 Chip Set and an older version of the Micronics system board. The Gateways that we received have a new version of the Micronics system board with a Phoenix "80486 Rom BIOS Plus ver 0.10 GJX30-01d". In otherwords, it was shipped with the X30 Chip Set instead of the MCM2 Chip Set. From what we can tell, there seems to be a slight, but serious, difference between the two chip sets and the way they handle the keyboard. It seems that during the NEXTSTEP boot process, there is a polling or resetting of the keyboard which is indicated by the "NUM LOCK" led being turned off (it is usually turned on when the system does its self diagnostics). It seems that the PS2Keyboard driver, which is the only one that NeXT includes with NEXSTEP 3.1 has a problem with the difference. It is possible that the problem is that the keyboard has been relocated to a new port (not likely since Gateway says it should be at port 0x60h which is the same as NeXT) or that a few initialization commands have been altered... Whatever the case, the keyboard (not just the 124-key "AnyKey" Gateway keyboard, but also an IBM, Zenith, etc) will not function if the system is booted into NEXTSTEP while it is connected to the computer. If it is connected after the boot process, everything functions normally. Solution found!!!! (well, kinda sorta as we are 1 for 2 on this....) Short answer: 1) In the file /usr/Devices/System.config/Instance0.table remove the PS2Mouse from the "Boot Drivers" line. [Note: it turned out that step #2 was not needed. It was actually the Turbo button... See below...] 2) In the file /usr/Devices/SerialPorts.config/Instance0.table change the IRQs from "4 3" to "4 9" (in two separate places.) The first change removes the poll for the PS2Mouse. Not sure why, but this definitely causes problems with the Gateways, even though it uses IRQ 12 and we are using Microsoft Serial Mice.... The second changes changes the IRQ for COM2. NOTE: An indication that the keyboard will work is if the Num Lock does not turn off during the boot phase. This should happen automatically when NEXTSTEP does the following: Registering: PCKeyboard0 If it doesn't then you most likely have a conflict with a driver... On another note: What is interesting about the whole mess is that we still can't get the keyboard to work on the system with the built in CD-Rom drive. The only difference between the two systems is the CD-Rom card, and even when that card was removed, the system still would not recognize the keyboard Well after more experimenting we stumbled on the real truth (I guess we were mostly correct..) The real answer: In addition to removing the PS2Mouse driver from the "Boot Drivers" line in System.config/Instance0.table, you need to make sure the turbo button is disabled during the boot/install process!!!!!!!!!! Honest!!!! This was the reason why we couldn't get the other machines to work. They had "Turbo" enabled via BIOS (which is ok!!) but it was also enabled via the front panel button. Disable Turbo during boot with this button, then enable it after NEXSTEP boots up. The keyboard will work!! [It seems to have something to do with timing for the keyboard. Maybe with the new BIOS, the keyboard doesn't respond fast enough???] ************** So here is the rule of thumb for Installing and running on the Gateway 2000 4DX2-66V with the Micronics System Board with the X30 Chip Set: For Installation -> do not plug in the keyboard until you are asked questions. Then plug it in and begin the installation. When you are told to "Eject the Floppy Disk and Press Return", eject the floppy, press return, then before the boot process begins, disconnect the keyboard. Do not connect it. You can complete the installation with only a mouse. When installation is completed, you will be asked to restart the system. Click on "RESTART" but LEAVE THE KEYBOARD UNPLUGGED! Now that the system is installed and you have restarted, wait until you are logged into the "me" account. Then plug in the keyboard, become root and then remove "PS2Mouse" from the file /usr/Devices/System.config/Instance0.table in the line "Boot Drivers". Now log out and reboot. When you reboot, you can leave the keyboard plugged in, but make sure the turbo light is off (disabled). To verify that your keyboard will work, watch to see if the "NUM LOCK" led stays lit. If so, then something is wrong and you will be without a keyboard... -- Paul S. Sears * sears@uh.edu (NeXT Mail OK) The University of Houston * suggestions@tree.egr.uh.edu (NeXT Engineering Computing Center * comments, complaints, questions) NeXT System Administration * DoD#1967 '83 NightHawk 650SC >>> SSI Diving Certification #755020059 <<< "Programming is like sex: One mistake and you support it a lifetime."
Newsgroups: comp.sys.next.bugs From: luke@research.canon.oz.au (Luke Kendall) Subject: Re: CC error in NS 3.0 Message-ID: <C8sKus.AMD@research.canon.oz.au> Sender: news@research.canon.oz.au Organization: Canon Information Systems Research Australia Date: Fri, 18 Jun 1993 00:47:16 GMT Verified on my 3.0 system; it's been fixed in the 3.1 beta. One assumes that it's still fixed in 3.1 proper. Luke Kendall, Senior Software Engineer. | Net: luke@research.canon.oz.au Canon Information Systems Research Australia | Phone: +61 2 805 2982 P.O. Box 313 North Ryde, NSW, Australia 2113 | Fax: +61 2 805 2929 -- Luke Kendall, Senior Software Engineer. | Net: luke@research.canon.oz.au Canon Information Systems Research Australia | Phone: +61 2 805 2982 P.O. Box 313 North Ryde, NSW, Australia 2113 | Fax: +61 2 805 2929
From: dbbrown@turtle.mrj.com (Dan Brown) Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs Subject: NXInvalidation Archiving problems. Date: 18 Jun 1993 09:55:15 -0400 Organization: MRJ, Inc./Oakton, Virginia, USA Distribution: world Message-ID: <1vshg3$2i3@turtle.mrj.com> Programmers, NeXT, and Netters in General, I don't know if this is a known bug but I am experiencing the following problem when Archiving NXInvalidationNotifier Objects in 3.0. In the following uuencoded sample code, I have an object that references an NXInvalidation object. The NXInvalidation object contains the referencing object in its funeralList so when the NXInvalidation object is freed it will notify the referencing object. All this works fine and perfect when the environment is set up directly. But if the same object is Archived the NXInvalidation object does NOT deArchive correctly. The funeralList and listGate variables are nil, and when the NXInvalidation object is freed, the following error appears in the CONSOLE/ gdb screen...."Unknown error code 10020 in NXReportError". I have tracked this down in gdb to the call to [super free] in my NXInvalidation Object subclass. I have tried a few work arounds such as archiving the listGate and funeralList directly, but this causes a "mutex_lock()" error. I am at a loss as to what to do. Is this a known bug? If so, is it fixed in 3.1? If not, is there a work around to get the desired results I need? I am currently waiting for my 3.1 upgrade to come in the mail, so I can't test it in 3.1. yet. To get a better Idea of what I am doing, here is some code to show what is going on. Sorry for the extra bandwidth. Help would be much appreciated, I strongly need this capability. Thanks in advance, Dan Brown (dbbrown@mrj.com) ---------------<uuencoded file starts below this line.>----------------- begin 644 NXInvalTest.compressed M'YV03K`D<6,G#!LJ9>;0>0&@H<.'$"-*G$BQHL6+%4%HO%&C!@@`&D'(L!'C M8T@0,628/,D2AL8:-&+`F$%CY@P;&E/.O`&R!L:?0(,*'2JQCL(P<C0"("-& MC)PW=]P0C7C4C)FI6+-JW<JUJ]>L`0<6/)AP81$W9]BDF8/&!1LX3]4P_$HW M:$B.'D%J'%E2;TZ5?EEJ=`D"ILP9,V[(@$$C)XV./@O7G4RYZ-&D'YDZA2IU M:]6KE4.+'DTZ=%B"!A$J?'$V[=JV;^.^.#U6-1T7;M*(F5MZZ-V.*T62#(X2 ML."3A`W;5,PX)¥>9/7M+QVJ4#E*EFI]&Y?IYNO?OX*733EV6-5JU;-W"?2-W M/%F%N'6_(!/&NHLQ;,+,F9,PO%¥;-#3F%U_$I43<236I5%A,RRW6F$B+R<`3 M"##X9Z%EUF&V5%/:=:95=Q>&*.*(%AF1AAP*29$0'&^X04892?4`PAX*G!3$ M$%0D¥8034X`@(XTG];%#C2%-404414@Q!!-!3-&CC$^(H4898]`QI)`*#-$B M'4^QP0:,/LY(I$8WYKCCDV(F4`8>8;0!QY=7I$$'&D'(,08::=B1!EHZ#)D` M?F4@U:<""80!!QQ$I$'&0'(.JB:;;L(I)QI.O%'GG7GN>8:C=^Q)!E1Q>CD$ M&V_P-VB00Y[T1!54,%$$%6C2F``=:UHY)@A8GF0DDDHRZ62844Y9Y95#DFCL ML<B*)Q!J[YEUWFOJR>:>;?'M1I]]N8DA8D@`"A@2@8$9&%A(`3[V$H.(,?>@ M##'$P!%(%2;[77779<8A9]Q99Y6¥_/8[&@$+T)$''&60H9`<@;81R`"B$!(` M$(1$S$`20@1+)1U$U!<&`!$;8'&5`!1"B`!`I!$Q(0P,4=T;;7Q,!P"4C*P" MQ(0L$,2A:HU17QHM%F+(!2,#03/*393A1AT(15I?&0)0(HG(`EBU0"`!O$+( M`3///,;6$2¥P;5DG&]!$?7*D@4<!$1^@I1M<OL$&QX008$4:9=P!=P(JSL&B MBS#"_+3#5¥5ME54!`&`&'J,T=/C3?Q,`Q'Y`%!((!@V)/$`:CX<<"#8`,%Z( MYU<#H4/):91^,@%,K$5'`#$'D(8!<6]A`Q!=G(Q`T4</48:7`42L@!!UT$%' MB[KS[GL05?+L1O%L]!XW¥ZS'/?-^:0]DQALN9'^Z$5LV'7<)9E"?`!*[VU$& M'6GHS```@2`B@"&!&-#0WP*DH4/((^¥7"`#(9.!*(9(30"HX1C(SH*)S(@N` M#GQ&B%&X(F($*`(9Y(2*^,V/$)F0!"8VL;]Q4,T>G"#$+0C!B:>)`@"C*)P4 MH)`X`*S0<Z20'`*`$+)2:(YSGO.<*0@1"E00('8TH%T#'TB(`:B,#A64'P)G M$3):[*]__PM@*@*!!P#4XH!/LP4#'0A!+<$A#TF<7R&86`@G¥L]_``R$``,Q M!BMBL1!:-,00(U8`*.B'5F%<8A.?B$8I!L(.;D1@'.=("`5,87<7`T$0O)3' MIY'1C%!,XQHW=D5!^LP0`/S'/P@0@$U(@0*;:`@A$&`5H8U`!SJ8@R8!.`I) M<BP`8Q"`+]X'0=S5H1`.,003">&`.=1!#&TP6AV0A[X6Z0"2?53C`3-QL@.$ MI0UA.`/38B:`.:@@?G#H6D"">;0Z:0>3DF#%(+E(B`*,*E!R($0J8B&+]04" M$.^SH![+R,<HJG&*:["B+RPI1W(6P#9S¥(4¥)8%!#6XB#E3[1PB+04(3HK!P M^W,%"EH8413`4'*4JZ'FD('`'!9BAST40.QD($1R9F!-;7I3&4#0J3F!P`UO M``%2,*4GM`A4B8[<XQGM*4!]¥I.0%D!II%;:4C3(U$YXJND9;BK&1]93D@.L MQ3ZS>,E,_D,`G8PH*!O2BU46HI4!!,`L9RG'7^"R(;H$`#"0R=-EYN)SP1@G M$0F`!$65H9%CU&DD_8B&0%*UGW.-0ATH.-"¥TG.G4`U$'/P*QZH60I,&Z*09 MM@J`KOZ#E:X<*RW-FLM*9.(7F¥B$,D*["4QDPA:N8*;OIC"&+K%A;G4+@"<( M.MM"G#"%"5`C^P*`"XH*8+>X@"$/4=$[`FPA!K1[AB$J9XC;-D2W5,N%;W>; M"TE$XVFD&*X`C&&[4:4!#K"]PP!F^XS;5D``5$L'U6I!T0"H-P"UD,0SL#O< M`##C=`C!P¥MF2PWG[O:]['VN>]<K"6I@5W*Y8%¥"P/"YK_Z6:N58;¥W&4+K! MC6%P<W#(`QH2@,(-P@>J4,4J`C&)*&QB$'%(@CR4@6(5CS:]ZST@*@2`BCL0 MP@#'C4$.QM"%/7!!#G0P`PRX$`8WS"$-/KX>V^@@!C9P0<A/GD.G]@,"¥K'! M?.C3V0[ZH``N0%,.::E)E[]¥!CF(^<E#%L.0TS#D.K`!IFXH`Q?BX.3PR8`& M3V8S%]0RY#*'(0]#-L,8AGSA-L-AR&1P@TL4P&5#P"¥#[%,"_@8`A/!=(P#O M),`L"1HQ!`1$92AZ@QQ@M@D"X"-_-"2$+H2!S;0%1#<)<[0<.¥U:U_:M$_*] MK0`$(('UMK#7¥)7O@0.1B<K94')0Z-PS!&>&5(J2`5_X5*W=]B4YZ"`0$'@? M_!(1:?P90&B5#E¥W"D$-!2@@$(0`@"&BD0E9I/88PU`M+^/D(J@D[4U+8]TH MJ-8!JK$`NKPM9.E*IX*"/PX(!A!``031D'=X(&(B<(^<=M:B2J'/#'1+BFU` M¥(3@P2%XA`!%Q^C]J3M$(Q!@P$,A]KE/0WA6%*&-QC>8:8E+^&YM;?,2C'SF M656$=AC!R$0P?I$)7H26&:&%16IID=IK/",3Q8BW*UB1"5:H`A4,")H*5"%" M;¥""$+GP!B8Z/;8]V=(;#8N;]=[@#5L0@A?>^`;9¥T!RJ'B#%H38A3<J(3$3 M?>D$<^!X5&#D#5FHVAO*($0RO'%?97B#'(3H!G$E¥0V"]C#K!3@N26L7,0E0 M#.==6EN<JR1JN"W@¥RT:_?!&'3-*`T$%RN![.?'T(AW(D1[F8(;L%4`K.;1A M3TNS/2%P#PN1Q6T,P1,^[F5A_`*,X0U?5+XY:&%¥`¥#ACF60OA9]QY¥O56F1 M;!!^Q")`L8[3X4MT$/W%2E¥/<Y#C&A"DE7[E*`!S?`/RHWS1E¥ZP-/J;HQ+D M4`"W9P[%@'¥7(%0J%2=S4BF7DE2:(GW'@'¥5@("2,B<-F"E¥,H"¥@'¥&`"B" MXFA/LUS^,H(D*!I?LQJM@1ZPL1[ML2RU41;5X@>¥<2R_D1??,ASA8AS',1CG M<AB)$0,!XAPWP!C148*C02¥:DAWXXAGZ`AI&^(10B!$GZ"RND1ZQP1ZSX8+D M`1_9(H/7$@;WD1_[T1_@P2U!."`X>!+BLH,)TH/+`83K(@,R,`,50B%16!E( MB!WWLAU,&`;[<H>`&(@-82(H0@=YLS<O$B-B8B,XHB,¥$B9`$A*Y4B1'DB1+ MTB1HXC+$DB5;8FN*&(EDTHAG`HF$0H%EH(!T@E08N"E^XH'6YB>&@BB*PBAT MX"BFB(H,J(I*Q2F>`BII("JD8BJI(HG#J!&KTBJO$BN$(G^V@BJWLBN6Z"N9 M*"47LXF">(W86!I3:!Y5N(+2HH7-$H/SH3'5$B)FZ"U[D88AL8:"42X>L2`^ M>`-PN!<QD!@V`"_9V!5Y:"^;P8<?TH3Y&)#]`C`"0S`&PR4)LS`-¥S`G,S$5 M0XU5DC'6`3<>`Y$O`S4ETY`JHQ`LXS(P(S,T8S,XDSX4YP8^`S0D,S0,8$OW MEA^TTC3T(S548S58¥WI`L#5CH$W@:!MA,S9<8C9H<S6@YS9P(S=T8S<1@S<K MTB*)Z#<)%#@$,#AF4#B'TT*+TV!Q¥SAS$#F34SE%A#D9)CDXU&"@<P"B0SJF M`T&IHQ#1XSJP8URSPWFC9$O0XSO`(SS$LSO-<SS)DY?&¥SQZ&3T$,#US4#UN M<#W9XP+;TSTQ0P#@(SY6AF4DZ4[N`S¥X!37V@S_5-`=/)44$5&EOY#`+!%@0 M)$&$A5,%M4$=]$$A-$(E9%L/Y4(LU!`OU&`Q%`@SI%%BV5%D^5'#]4/&%42U M0TY&%#QXY52(Y4=55$E_14@$X$5@5%C(N5?WQ$:,)5=T9$<*<5?2J5?)M$:` MQ)R-19J%=$C>1P>*Q$C=>5C4.4G7Z5B:Q$F>1%FC5$I`<$JIY%5@I5L.$TMD M54O"=%;JMDN]]$O<-$Q]Z0;'U)G*!`#RYDP"`4W2Y#W5=$W6IY,'ZDU0`4[B MM$5$9$Y?@A3JQ$[N!$^6V53>R5/XY%/-Z4¥`Q52=DYH'E5`+U5"PF4+L`P`2 M15$Z:E&VB5'&ME&¥J4/:-5(E140G!2DJQ5*3¥E(Q-5,.:%/KR590Q:+C"52F MR*0N!:6K"*,YQ9[?&553-9Z8]%A7E54Z2EF6A5EAI5EE):!IM58,2F,.^E:2 M$%<>"D%U]2+'F:*)U5?BB9UQ(UBGB:)@JJ**]9Z.9E61M0F3%4J5I9^9)4N; M!:>>!5JB15JFA5KR9DBM16WA)5NTY5"X!7"]]5P/QEO"U4/%=5S))8(^XU¥` M)UVH2EW615¥AQ5V=YEW@=93CE6LH=%XP!E_M!6#"5@C9Q:KW!4'YM5¥%)JL# M1JP"9JP&AJP(IF`,]C2CD*H1!E¥35F%6<6%6D6$-L6$`T&$`¥&$A-F(E=F(I MMF(MMF(<%*V5-&,U=F,YMF,]]F-!-F1%=F1)MB5,5F>!)F5K$7B1>3XDN65C MAA1A!@,-"V9F!K%HQ@5JQ@5ZYF9P)F=T%F5WEF=#QF=<X&>`]F2#Q@6%Q@5U M<&A<D&B+UFB/UFV64VES<&F9MFFV¥VEU$&JL5VJG5DVIMFJM=C6O)@:QYFBV M,VTZ-VJX5EXHM&O`%F``$+7'>IO%IIN!D&SRQ6S.QC'0)FV?NK37EFV.%@C< M%@B2)C+?)C3C.F[E=F[IMF[M]F[Q=C(.4'=WT)+Y1@G[!@#Z0#77EJJX('`# M5W#3@W`*QW``X'`0)W'644QN8'%I@'%@LG'F]W%T$'(CUXLFAW(JQW+J]G(Q M-W.$4',WUXG4MG,NEPD^EPE`)W1$9W29@'29H'29P'29X'10)W549W58IW5< M=PM>!W9B1W9A8';"A'80M'9M]W9Q-W=X>W=YMW=]]XME`'B"%V=R4'B'EWB+ MUWB/%WF35WG#A7F:)T2=AWIL$WJIMWZCUC7JJWJEUWI"`WNR5P"TEWT#J'N^ MTWN_YP;!-X#%!T'(5XL#R'QT]'S1-X#4US'7MYW:9WP*T'T7`W[B1PCD5S'! M@W[J1WJCUG[O%W^UXG_WUVGZ5P;¥1RO^!X`".'P$:("W."FY2%,/.(`1&#$3 MJ*05F(HTG($NO($=XXJV9P@A*)!&;(3;"`5"X`(L>#%&?([!`2YJJ(,L(1,W M,`-NF"YRB,4H41,I@8]'/!7[N"']Z"%9`2)AG,86D@0¥0@5!P`1,0`1)(`5A M(@(D@`)(¥`1-4`0I¥`(W`P=S(`)#<@1%X`1)$@144`1-$`1L'"994`13,"1& MD`2N,@5N+`2N0HJJ0@5(D"1?P`04@R8H4#03%`9?,`<L``).4`980`6GG`+% M"`*7Z"20'"8H,)1+ZP)MD,JVX3)SH,NP?"M/P,F>/`6KHB2U+".C'`9K4`88 M]R7!?!)(¥`637,FVC,O5Y@)HP,MEX<O:',TAP<:)+`5&<"/)#`(H,(75`L[& M2,Q2¥,ELO`1%0`2VS,X@T`343,GGG,X[619?`$U[`LR;2`1/,`1),`0[4LW[ M',Q!``50,,MH¥L<Y4Y)#0M!#4`2M7,A3X(BB',R+S,9.0#%A,H5#`@52¥`1* M4`0X0@59@"1A(M$D";DE?=(IC2-6D"0;O2-A$@,N$`-#P@0%_<9)H`5!@,E% MD,_6_".WHL[9LHDFC=(J305.$`1[/-+]K!!#,@598,F*C-&)S",<7<¥_'01. M<`15$`2$'"8I^!K%HL9N_=9P'=>4L8U3^,_'ZP:Z'(A0C(9],<4'$A(UL!A9 M/(?N4A(H`1.(`<9R;1%CK(3^>,8`N=B231DOH`(@<`1&`R-+0P8@(`9Y``)S MLE*KW,H@``5QD4C`¥XN)"`)CX@3#7`0Z``)$$%.N304@<"=%)DV@C2>!]¥PK MU0(M4-JG724@D-ILL-H`S39W'7ARX@(*H`(O8&XCD`9N(FKHR0.QN`9RXL<C MJ3.0J¥T^8&YV¥`:*`@+)C0)[@IX..P:IC-M)H0(.:P=;T`4I(":WL@4P[=W* M¥U)UTP6Q/+GHO`4!¥<<@0"IA0`9.H!N'E*`Z$'%7?1O9(@(@P!DPH@,#?BA: M2BEMDA`6_@3T32@G(>!80.!R4`=NX-¥@'A+W?>%P``)F@#!E@.(GL29R@@(P M$,Q<-MEQ3=</[@+ITR)H$"A-^81[?8-]O8Y47,4PD!SH0MCRF!,V,!+W:(<Z M/A&-O8=FC!5H7.5<GA5&D`#J'(M@_N"%<B@*¥.53..9BL85T4.9PH`!='N=R M/N=0N(UCT¥R^'<9%GHY'_A=_?=@UP.3Q*(</4H¥U$.6*/>=77L;YXH=.2.>0 M7A$CH``C<-F9+0>;W=F?'=JJS,JV;=KL@=J#==PPXMR37NFU#=NR3=O#?-MH MD-LK-2=KX>+6"P+`+=RA3MS&C=QW;1U[PMRW0>F47MK!(U->,N%F``)Y¥`8[ M>]LK¥WMZ4)+,[0:T_B6!IS,ZQ]EW[LS6R¥0(TR9-5@;#7F3:SLS<_B5,7"K6 MT0;ACLYB¥`8N!7V0:Q`ID,KD?L_F[MM,>NQ[@A]U¥"*[709M8.KF-M55+2-3 M:&Y/7=-4<--2D--.L-,]K0!,4M9GG=8RLM9L86Y`/01O_`52`,G'?-%HPM2Z M86X0?<[87.IM``*];)&_W`;FA@0*C28K+P?:[/+=#//:;&Y-4/-6O>;-8M<! M+?,*,,R=_/!*@B;;[MOFI@"++,¥*+<=T+".S42M$4`;F0RH$(P<OT/36.P<O M$(L*P,:6_,9Q/,=A<L=YO,=]_,=S4/9MC/9&P"1'@"8M,`:V'G@MT/)X84@C MK¥A/0`1%$";E8FX'?291'=8]D``MP!]5TEJ!0BL@¥`5?D/@1'XL@¥`)&X?5J ML1NKK%]30"MP¥`)7(&IKH#=A,`9EX`*Q./;=79(N<'%FP/&A'"8MP`:DG`:F MS/=L,-JN'/>$+P15<`3P+`1H<L>@C/PIH`!/;03'G_PHL/Q3T/SFU@+]S@;_ MOE)@C^YPD3#A;F[9O_T@<,=17P13/¥=]'(NZ;.[7/_X`W_VMSR)'P>Y?¥OYN MX._QG^_=_B($PS?YCV[$O4A'(I+8$CL#3&%/R(DGIA&Z11131WYN7*PC&X`3 MX%&#N#,Y(3'0@`D1+^C<HNL0C>X/$<`1"!%.G:737IG.LP6¥3D?:0)VP0$^[ MKM0).U3WVF+;;%-EK0ZWH858Q]NJW6¥+;BY0U*DV,)'<?)V1`0'-3=CIB;IA M;DB@$WR":FP;W3QM%I#VG'#H<¥7ASX&`&W`#*J!R0`PT8`8$.L=0$VA`HI-S M'W`)_2-'!P6=X'2K;D`&!&"W0Z'=%D)V:VYH(+R]0181!T7`RWN!OPP-B`!S M4S+8!HPP`ZMOI=P¥'>`R%``H0H)D(`',BEK1BEZ='$@`8J`.F($M0/R,P$;3 M`BB.RRB`RJ;S%()Y.Q]HX`T8#!``W:X?"(AA"¥A2Z*('Q!_XAAP8$L$-%O*P M*+4I:F$BPH4@0)K0`2I0XL;`&@"&<"&]'0&D$$VR3VA;AIA.F@##.1`&S`<T M;(8ZX!DR0VFH`((;POA)6J¥,7$-N.`HMV¥T[A7-"%0:>5M@-9<JA,!IDH%GI M`!3P?(P,>G)OK#`%A('15S;0`C`48KX01@P)!0`$X"$A3&^',!'*,M2UM%!` M?BM)66__+`WK%]QB4:)8%+FA%@7$6T@(#2)!1(AR`!&ROH6XOE*7'$`!>"LB MHK")Z`H[1;VY`Z&"#8R*4I%]-N)`+(@N`LX!`"G($+-9&ZB"#?`,&;D"D>1R MPI*;0UG,'M$`!0&'EN(9C'-I¥+%IN<C6!@G@'K1N($`$3$%!>!"'"C=Y7/OM MYID;4KAQ@DDZ7(7L,!?JL%,DPV1A#^N%¥!!&.$)"H2@DH3:,AF7`3]3%!-`& M¥@`QK`/&T$¥0"E)(B$PA%"AV??$O&D,DR#9BREULABZ`%4:W!+",@MPX7"DR M8@N`@"UP&8T=J>`QC%%.H#B^Z!>+X1H($UN@^R0[8:@8U¥!HW`*/4;?Q!SH0 M!*Q`&<`3@$(')$;3.!H'HV5+%'HC/WRV._#JT-.LVQ,K¥#)&QE9(&1-`:MP= MR2X6P4-YF!6I0)"[;-MPI1BMZX$P5B"7`(RG¥0[H!Q#`D;X$7DM,7,`-I$<1 M¥!I5(PA(AFSC,F9#RY@=>V,"((6C+U"P`>]H&B4CH2`46V`W@D<7!^/L(RD$ MCOAF.!9')#CMDF-V7(Z3$4"Z1^GH(JBC'[R.G=$/]1[^."#%8^`ICT8C,;F` M]+@>VR-T?(_]D`[(Q]@8XP2C<V21!+(,M,C_F`"^X<ZB=JKQ2KA"77B!=M%& MG(L)8"^R2+T8"4EC:W21A/%$&$;$6!H'9'ISC/01+T)(&@DC,^-FS)!>X@V` M1@5(!T:C@%R,F=$]LD;>Z"=@8Y1LAN3Q?-3&VY@^OH1N;)+&T$!:MA[)`U>* M<O2/S?$YLH%D1PVM87:<C^*P/B))RZ8BVL`;,!^>$00@##,`(XP&ZPL¥P¥-! MXD69XB*0(!W`7N8#-Z([/"DA3R2%C(>U(K:)@$)Y*#6%I$23]ZXU$DB6<0H- MI1S(`^C1#;!'%VDF`^6D?'$R¥C5^R=.(*V=D<R2%E>(.*,KS43;"X:G4;:FR M/[XX5LGIN!2B]&U3,D].R'=8(4-E5E01X%"I',N5DBP'Y+)L>6'`!¥;*64DE MSZ1N"Y,G¥AL62S^)%[TDG#R-F1%&A@';J"E]I6^4;6LA0;*4!7D<J1VGNY/, ML5/N27?X_T"E?A&5UM%.9L?.YLQ$39T$;?W10Y+'X6$>122)E)4F<F#"1Q7Y M)UFDG`0!^=$@<,C%&#"=XZZ,D2U2(Y!"A,8CGH"KB&U5P`VL`9@2%5ZA''@* M2>'Y`#R9``,$&W(,""J"#]*!(G`S14VTO)<(4CCNR_K`(!GCMCR:SG%:%DP+ MF3"QXZ34D&""TWW'Q2@Q0>1YS!X7DUQ&S1.Y,5>DN;27+_)LHDS[:"/E`(Z$ MCCJR'8Y)¥`@DZV(8.)(T¥F[V1RNY!52E0?B,72`T=LDRJ3<'Y&S<=@V.>¥@! M,B`VMT#A7(R'4Q7NC@8G!6"='!";A*)M4KO'>0P9C2LLFQWS;-;-(AG:Z&6; MS(OEL@R83D"!&N7EZOP2&3,Z4DN#68NR8GK$.4:##@2>-Y#L_N6#M)CJ$6/J MQ62'`DHGFP04S4¥2ZHT4:0900"8T`ZE,!'`!.O`Z,:,/``$E(/!DS["9RFIE M]31OYJX+I#)"(0F]Y_'¥$N9-<K(!^N8G=@=_()_+,[TUS^<9/:=GU>R,LT[@ MP0&!$0+"9C"3A`)3=D[-4/D¥V>;YN)%I<D_&S>#6)VTE-F210#(@4(&!43#X M8<)(`"K@#I0-6N%W4.>?L(22,`'X-@`63#AA%?"$0¥T^QD^VT3Q'Z(:+GMGS M!5Q&$9#*`@(28!EE(%$@#`^6!U!`"OB?*S3(.$]-6#_=0!`0'OH3?:`%T!93 M&JC51)/(4=_U@.MY`K+G"5B/J<R%!I-@%D`)YG2TE@5T("8`#2HGRD`'M6I/ MX/]-4(-D0=N$XNR@*$"+EH$:B@6^P!60`DD@D>T()I`%_B>`0P%DE(/6NA"@ MS"0H!24#;;3EJ8#ZY@2J`!Q+GK+B/L[)B?GMP&6E/*.O+O`8+:-QV[X=K>!L MW'$%HH?;@">;XYB[`ANT#$B!-P#O7`8@3:4=-)6QR/¥YYF8B?UBC%10AM8E7 M6D8[*"TEGWI2@'[1@UD=,>3"S*0,DY,V4#+0/X4GH>`R:H(-O,¥$$$F#Z/PD MHK:3VB6)DT;'4$#V3`&Q;7O*RBQJO4CH'*6D,`)GHH!T"E-^J)¥8FP/S4¥K# M,/I,::3F3*!F8($.RVSI0*5)!&UEAS218M!O1P8Z:"5$"B%4CII0%`H*B60( MG15GTT5:TSA:3E]H5HRA,Y2.WM!@HD/=5P]UIX1BHM+/;'I$:86;4*)G@(D. MRP/G`WT$%;6B6-0'FE,N"D^%:;4DIO0T<PJY,XKPL$`:-1JX%)'JTC;P1JT7 M136/%C4@@+PB$`0(FA/@HWZ4>!+4,TI(T9DA9:-#-1^J,D?*!"`I^4R2Z!.9 M;E)J1U`98Z3D=`<C841+4UHE-0+0%'*N=*I:+UKZ:8)1&0BJB10%R%5H]DZ! MJ=0<IK7S0BK,22E69219)99TPWPPTQ+I)Z"I^Q1WU)1¥MM7<P`;>J4C%IM)3 MFTH!;HK.OFDXG0,Q58ZFLK1:-$¥B.WT#(+5¥_M6;6HOH:0*`ICJ5#B#0(>DY MRR%)S#G5!AVF0K08W=JA/`V5]+!%F$)¥N$CWX4]""W63;4A"¥W@&7$J¥K!6@ MK58P`:/17-&`?82-T'4VFB<=P%Q="G>EKG/B-3*C88EO6-]V-:[]¥`R(U^P: MMMB`>:(";V!N')EPMU[U"PB8(,'QSQA0V>HV¥:D^!8AQ40X`2<V(7>TK!?L^ M7B*50,?`><)HA8+=D[WQ)-S3'%E;<2(9Z(HJY2N6I-OJB1QB[(-<*C&%E0&* M2#`O(BU2L+9PP&[&¥>H22T[@A";-;`F4@3P0!%S$$UB<,,((/`6VH0,H:WU% M3P>6-B98U;A?$2B%%846%L-^"0T+N3BL242)G"O$LL1VV&)]$3"BB2@V$0U8 M%A==?0_P<;`^UD5.6+A980UB53RS:#;-(HMM]`<O1B#,1U90BB&Y/]<NS(4% M1`P.XAT!H<!6ASJ@HKL,>HC1]2$1J&;CW%6,@W,0#M1!V*=H¥:`>I&Y",RMN MQ4%(Z5X$QHDS(),*/+RA5@0DH9A1`.X-$VK"+5`#4@**^XB&,"0J1%49VR3. MM5`>DHMR"5A9$6IA;!G8`I9LTVJ!(C`::ZWZQ+6:]A/R6AT9W+CD-#P?B#.X MVD-7=QT6J4S9=L!0V0Y72ZA5;>VQI0--0'W.PWHX;9MM?0L#V5;_1%MN>P^I MK;,UE.*V'?9*8$A0Y^%5S:655*L6)'B;,(!A(,T^_`RO9E5G.V^%:B6]B1ZQ M$/8>D3A8T61L<QER,"`<$EN8!.:`T%,4X4U6[,5Y>3Y1IRADH&JR>L8VXUDO M@2'%[;CM<",VW(=+!F*;352W,([="CEW"U"Q:KSEMP%UJ-K;5,IR]>W+K6_] M-I$"W)Q8:'NNSP5$;';G`<*¥EHWB+`3,@A(0!-`$CI`4K]APZ&*([2EVN:B8 MY:;"EONYD^W0HB<_*'3=K#:CM"7#*^;.#=L:72&7'+"^I.L)SMB9)K&MN1.5 M,=/2[HF"T3C=8^34/W"7;UQ:NEMD^^N1=86'¥^U*VW+K;9^M^Y,5!R/_?1$4 M8&OMW;9+9;E6V+:`&/`_QZP"K:W!;?`RV_=6WVPMD+RGFI>X=EYSIT_O[N0, MO>;VVX;;W0$D$^¥86+SH=G<X7O4)>8,MIYV¥E?>`^ETR>W'1&;G=O-56?7Y> MW4OM4&_AC;V5%?.BS-$I";^AF#T?J#'M@HE>.1K_Z#=,GC62^/K74B1-(6OV MY:_4+C6N+##Q(OZFEEP:H=`5MMM¥&W-O+FAKOPD#[8[?I-!N<^Y0'8UO]<!Y MDS]3/.%OFXB>8X"&9EI=6P12F:W]G_F7#.S?'FI_*RD`%L"1E],:8/7Y/V.K MD86;E+'WWMN:ZW¥5*<[MP/)7[6[@!IPP¥"¥60*5E=`'W7Y>;,!YP[27`!MC< M(>`3G$I5,`G^OUDQ`+]@83N!]4¥%_KX7^/(BV8`+:9?LV&VREQ'P9MR*NW&# M7/6LF¥33<7Y/A*$WF-(<@*¥Z@'P^#D3BP5"`B4,89T!UY%A10W)+4JR-:<I# M!Z2`#Q=9R^?'/9T@X`NC,#$L!Q1G&89<9UB_&1,BVUAIY!¥UG_5R6%9A%W&% MWT`6EH1;^#R)&KTZA[=3'2;#S$)1F&%X-[G0L#%9PUZU?')<.`R&Z;`=CL2O MMN)08HRCAQ44'ZZGS?$-LTX9L8D!Q?,%O]NW]ZKB^R<K[NDLMKA*F.$Z7%!L M<@,LD+S%88*RON(@G$^5;Z^$P@$.&%-A1%2(#W$"2,3NRPN[@4[LB#_Q6)#$ M>'@46V(%A8D)1225PA4W#D_C1MQ[K+%!P,:P5AN7X@=K!D8C;"T4WU-&"&-[ MJGW%K]JEOOIT_;Y;?YLPY"T(EA7V&$S4WPX¥&H%Q3U41!¥Z5WN`V¥(,M;S'6 MP#27_;+@-N&/)[+,`¥C2-RF,8(+L)P(""J85"KD#V[OJV9"U[]¥EB&86ZZKD ME<R26[)+?LDP.2;+Y)E,DVNR3;[).#DGZ^2=S)-[LD_^R4`Y*`OEH4R4B[)1 &/LI(640` ` end
Newsgroups: comp.sys.next.bugs,comp.sys.next.sysadmin From: sxuwan@fnma.com (William P. Nixon) Subject: NIS Password Map Problems with 3.1 Message-ID: <1993Jun18.221218.591@almserv.uucp> Followup-To: comp.sys.next.sysadmin Sender: usenet@almserv.uucp Organization: FannieMae, Washington, DC Date: Fri, 18 Jun 1993 22:12:18 GMT We are having problems using NIS with NeXTSTEP 3.1. We create a user on our NIS server and make the maps. We go to the NeXT machine to login as this new user and fail. If I login as root and su to the new user, everything is fine, even with "su -". I can ypcat and ypmatch the passwd file and everything looks good. If I su to the new user not as root so I am forced to eneter the password, I get "login failed.". It looks as if the NeXT machine is not matching the encrypted password correctly. I wrote a one line c program which uses the crypt() function to verify that it works and the encrypted string matches the passwd map. I even tried from the console window in text mode just to see errors and it just fails. Any ideas? Background: We have a RS/6000 running AIX as our NIS master server. We have one 486 running NeXTSTEP 3.1 (Gold) and one NeXTstation Turbo running NeXTSTEP 3.1 (PR1). The two "NeXT" machines are bound to the NIS domain. -- bnixon@fnma.com or uunet!fnma.com!bnixon (NeXT Mail Okay) Bill Nixon, Advanced Technology FannieMae, 3900 Wisconsin Ave NW, Washington, DC 22016 (202) 752-5468
Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer From: ian@mindvox.phantom.com (Ian Bainbridge) Subject: Serial port Crashes System In-Reply-To: <1vl58iINN8u@theborg.stack.urc.tue.nl> Message-ID: <X1JJ6B1w165w@mindvox.phantom.com> Sender: mindvox@mindvox.phantom.com (MindVox) Organization: [Phantom Access] / the MindVox system Date: Tue, 22 Jun 1993 10:22:56 GMT Since I use the NS/FIP machine I am on to access a larger system using SLIP it is excrutiatingly frustrating when the entire system decides to just freeze up ever couple of hours for no apparent reason. This never happens when the serial ports are not in use. I know this is listed in the Release Notes, but how about a FIX sometime soon, this renders the entire system I am using nearly useless since I can't connect it to remote networks using a modem, without making it hang. To top it off, once hung it is REALLY hung, typing alt-numlock as suggested does nothing, the only thing to do is hit the reset button, which is getting a little hard on the hard disk after doing it two or three times a day. Running fsck after this circumstances produces this message whenever I try to reboot or log out of the root account: Raise RDP exception 6, code 3, subcode 0, C to continue of coures typing C or c, produces an endless scroll of the above, with the only way to break out of it, again, hitting the reset button. *sigh* ian ################# @ #Ian Bainbridge # mindvox. # ############################################### phantom. # I am not responsible for my opinons, I don't know or care! # com ###############################################################
From: zryx0376@awssg6.rus.uni-stuttgart.de (Markus Wenzel (Hiwi bei R.Rabenseifner)) Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: Re: Serial port Crashes System Date: 22 Jun 1993 13:06:56 GMT Organization: Visualization Group Comp.Center (RUS) U of Stuttgart, FRG Message-ID: <20705g$qgm@info2.rus.uni-stuttgart.de> References: <X1JJ6B1w165w@mindvox.phantom.com> In article <X1JJ6B1w165w@mindvox.phantom.com> ian@mindvox.phantom.com (Ian Bainbridge) writes: > >Since I use the NS/FIP machine I am on to access a larger system using >SLIP it is excrutiatingly frustrating when the entire system decides to >just freeze up ever couple of hours for no apparent reason. > >This never happens when the serial ports are not in use. I know this is >listed in the Release Notes, but how about a FIX sometime soon, this >renders the entire system I am using nearly useless since I can't connect >it to remote networks using a modem, without making it hang. This is exactly the same experience as I had. >To top it off, once hung it is REALLY hung, typing alt-numlock as >suggested does nothing, the only thing to do is hit the reset button, >which is getting a little hard on the hard disk after doing it two or >three times a day. For this reason I temporarily stopped my uucp feed. With tears in my eyes... Linux did a good job.... ..... :-(((( >Running fsck after this circumstances produces this message whenever I try >to reboot or log out of the root account: > >Raise RDP exception 6, code 3, subcode 0, C to continue Rebooting with fsck at least finds no irrepairable damages in the file system. But I am sick of rebooting several times a day. Desperately, Markus. -- /dev Markus Wenzel, University of Stuttgart /usr/spool/mail wenzel@rus.uni-stuttgart.de ‾/.ircrc/nick Marsu /etc/motd NeXTSTEP for Intel - Come on and join!
Newsgroups: comp.sys.next.bugs From: ecw-m!jr Subject: Simon Say's and NS 3.1 on black hardware Message-ID: <1993Jun22.134939.1838@dcm.com> Keywords: Simon Say's and NS 3.1 on black hardware Sender: jr@dcm.com Organization: DCM, Inc. Distribution: usa Date: Tue, 22 Jun 1993 13:49:39 GMT I just upgraded to 3.1 and my Simon Say's from HSD no longer works. Here are the error messages that it generates. I emailed HSD but have not received any reply. Any one else experience this? Jun 21 20:02:36 nsm08 SimonSays[568]: %%[ Error: invalidaccess; OffendingCommand: put ]%% Jun 21 20:02:36 nsm08 SimonSays[568]: DPS client library error: PostScript program error, DPSContext 30988 Jun 21 20:02:36 nsm08 SimonSays[568]: %%[ Error: invalidaccess; OffendingCommand: aload ]%% Jun 21 20:02:36 nsm08 SimonSays[568]: DPS client library error: PostScript program error, DPSContext 30988 Jun 21 20:02:36 nsm08 SimonSays[568]: %%[ Error: invalidaccess; OffendingCommand: put ]%% Jun 21 20:02:36 nsm08 SimonSays[568]: DPS client library error: PostScript program error, DPSContext 30988 Jun 21 20:02:36 nsm08 SimonSays[568]: %%[ Error: invalidaccess; OffendingCommand: aload ]%% Jun 21 20:02:36 nsm08 SimonSays[568]: DPS client library error: PostScript program error, DPSContext 30988
From: dbbrown@turtle.mrj.com (Dan Brown) Newsgroups: comp.sys.next.bugs Subject: Re: NXInvalidation Archiving problems. Date: 22 Jun 1993 11:34:11 -0400 Organization: MRJ, Inc./Oakton, Virginia, USA Distribution: world Message-ID: <2078pj$c2p@turtle.mrj.com> References: <1vshg3$2i3@turtle.mrj.com> I must apologize for the previous post. I just found a solution to this problem. I will summarize for anyone interested. What is required for the NXInvalidationNotifier Subclass's read: and write: methods is to implement the following..... - write:(NXTypedStream *) typedStream { [super write:typedStream]; NXWriteTypes(typedStream, "ic", &refcount, &isValid); NXWriteObject(typedStream, funeralList); /* Archive object specific instance variables here. */ return self; } - read:(NXTypedStream *) typedStream { [super read:typedStream]; NXReadTypes(typedStream, "ic", &refcount, &isValid); funeralList = NXReadObject(typedStream); listGate = [ [NXLock alloc] init]; /* de-Archive object specific instance variables here. */ return self; } It guess it takes a few days of sleep to come up with a solution. I still don't know if it is a bug or a feature though. Dan Brown.
Newsgroups: comp.sys.next.bugs From: thompson@netcom.com (Eric Thompson) Subject: Re: Simon Say's and NS 3.1 on black hardware Message-ID: <thompsonC91qx6.Hr4@netcom.com> Keywords: Simon Say's and NS 3.1 on black hardware Organization: Netcom - Online Communication Services (408 241-9760 guest) References: <1993Jun22.134939.1838@dcm.com> Distribution: usa Date: Tue, 22 Jun 1993 23:36:41 GMT In article <1993Jun22.134939.1838@dcm.com> jr@dcm.com writes: > >I just upgraded to 3.1 and my Simon Say's from HSD no longer works. Here >are the error messages that it generates. I emailed HSD but have not >received any reply. Any one else experience this? Yes. I got mostly the same error messages... this really sucks. First Microphone (which is sorta better now).. now this. Eric > > > >Jun 21 20:02:36 nsm08 SimonSays[568]: %%[ Error: invalidaccess; >OffendingCommand: put ]%% >Jun 21 20:02:36 nsm08 SimonSays[568]: DPS client library error: PostScript >program error, DPSContext 30988 >Jun 21 20:02:36 nsm08 SimonSays[568]: %%[ Error: invalidaccess; >OffendingCommand: aload ]%% >Jun 21 20:02:36 nsm08 SimonSays[568]: DPS client library error: PostScript >program error, DPSContext 30988 >Jun 21 20:02:36 nsm08 SimonSays[568]: %%[ Error: invalidaccess; >OffendingCommand: put ]%% >Jun 21 20:02:36 nsm08 SimonSays[568]: DPS client library error: PostScript >program error, DPSContext 30988 >Jun 21 20:02:36 nsm08 SimonSays[568]: %%[ Error: invalidaccess; >OffendingCommand: aload ]%% >Jun 21 20:02:36 nsm08 SimonSays[568]: DPS client library error: PostScript >program error, DPSContext 30988
Newsgroups: comp.sys.next.bugs From: nico@imani.cam.org (Nicolas Dore) Subject: New one in Mail.app? Message-ID: <1993Jun24.030101.8433@imani.cam.org> Sender: nico@imani.cam.org Date: Thu, 24 Jun 1993 03:01:01 GMT Hi folks. This may be old news, but has anyone else noticed what happens when you enlarge a "Lip Service" panel in Mail.app? (didn't check with other apps) Seems the nice folks at NeXT didn't pay the "Size" attribute of the button group to much attention! Just a little, non-treathening, misbehaving, but just trhought I'd mention it. Ciao Nicolas Dore -- Nicolas Dore nico@imani.cam.org - - - - - - - - - CAREFUL!!! FRENCHIE ZONE!!! - - - - - - - - - - - - >"Si la classe ouvriere ne nous satisfait plus, nous la destituerons< > et en elirons une autre" B. Brecht (ne parlant pas d'informatique)<
Newsgroups: comp.sys.next.bugs From: jon@afs.com Subject: Re: New one in Mail.app? Message-ID: <1993Jun24.172840.1755@afs.com> Sender: jon@afs.com References: <1993Jun24.030101.8433@imani.cam.org> Date: Thu, 24 Jun 1993 17:28:40 GMT In article <1993Jun24.030101.8433@imani.cam.org> nico@imani.cam.org (Nicolas Dore) writes: | Hi folks. | | This may be old news, but has anyone else noticed what happens when you enlarge a | "Lip Service" panel in Mail.app? (didn't check with other apps) Seems the nice | folks at NeXT didn't pay the "Size" attribute of the button group to much | attention! What exactly is wrong with it? Perhaps they fixed it in 3.1 and I'm not seeing it. -- Jonathan It's a thousand classes, give or take a few, Hendry I'll be writing more in a week or two. <jon@afs.com> I can make it deeper if you like the style, I can change it round and I want to be an ObjectWare writer
Newsgroups: comp.sys.next.bugs From: peter@corsica (Peter Eisch) Subject: Re: New one in Mail.app? Message-ID: <C95GME.D88@news2.cis.umn.edu> Sender: news@news2.cis.umn.edu (Usenet News Administration) Organization: University of Minnesota References: <1993Jun24.030101.8433@imani.cam.org> Date: Thu, 24 Jun 1993 23:39:21 GMT fixed in 3.1 Nicolas Dore (nico@imani.cam.org) wrote: : Hi folks. : This may be old news, but has anyone else noticed what happens when you enlarge a : "Lip Service" panel in Mail.app? (didn't check with other apps) Seems the nice : folks at NeXT didn't pay the "Size" attribute of the button group to much : attention! : Just a little, non-treathening, misbehaving, but just trhought I'd mention it. : Ciao : Nicolas Dore : -- : Nicolas Dore nico@imani.cam.org : - - - - - - - - - CAREFUL!!! FRENCHIE ZONE!!! - - - - - - - - - - - - : >"Si la classe ouvriere ne nous satisfait plus, nous la destituerons< : > et en elirons une autre" B. Brecht (ne parlant pas d'informatique)< -- -- peter@tahiti.umhc.umn.edu (Peter Eisch) dig.
Newsgroups: comp.sys.next.bugs From: eps@futon.SFSU.EDU (Eric P. Scott) Subject: eqn broken in 3.1 Message-ID: <1993Jun26.180214.18109@csus.edu> Sender: news@csus.edu Organization: San Francisco State University Date: Sat, 26 Jun 1993 18:02:14 GMT eqn /usr/man/man1/eqn.1 dies with "Segmentation fault" on NEXTSTEP 3.1 (m68k) + workaround is to use eqn executable from Release 3.0 not tested on NS/FIP -=EPS=-
Newsgroups: comp.sys.next.bugs From: nico@imani.cam.org (Nicolas Dore) Subject: Re: New one in Mail.app? Message-ID: <1993Jun27.050446.3814@imani.cam.org> Sender: nico@imani.cam.org References: <1993Jun24.172840.1755@afs.com> Date: Sun, 27 Jun 1993 05:04:46 GMT In article <1993Jun24.172840.1755@afs.com> jon@afs.com writes: > In article <1993Jun24.030101.8433@imani.cam.org> nico@imani.cam.org (Nicolas > Dore) writes: > | Hi folks. > | > | This may be old news, but has anyone else noticed what happens when you > enlarge a > | "Lip Service" panel in Mail.app? (didn't check with other apps) Seems the > nice > | folks at NeXT didn't pay the "Size" attribute of the button group to much > | attention! Just noticing: I forgot to add "Make the wave view come out." _Then_ try resizing the panel. See it now? I thought so (many people have written to tell me it's fixed in 3.1. "There's one bug. Now let's move on to the XXXX next ones!"). 8^)> Ciao > What exactly is wrong with it? Perhaps they fixed it in 3.1 and I'm not seeing > it. > > -- > Jonathan It's a thousand classes, give or take a few, > Hendry I'll be writing more in a week or two. > <jon@afs.com> I can make it deeper if you like the style, > I can change it round and I want to be an ObjectWare writer -- Nicolas Dore - CAREFUL! POLITICAL COMMENT ZONE!! - nico@imani.cam.org > "If I can't dance, I don't want to be in your revolution!" < > Emma Goldman (to Vladimir Ilitch Oulianov [Lenin], 1917) <
Newsgroups: comp.sys.next.bugs From: peter@tahiti.umhc.umn.edu (Peter Eisch) Subject: Re: New one in Mail.app? Message-ID: <C9AuAo.KB6@news.cis.umn.edu> Sender: news@news.cis.umn.edu (Usenet News Administration) Organization: University of Minnesota References: <1993Jun27.050446.3814@imani.cam.org> Date: Sun, 27 Jun 1993 21:26:42 GMT I just spent no less than 1 minute resizing the Lip Service panel over and over and over and... Note too that you can't resize the window unless the SoundView is visible. Software Version 3.1 (Lightning4H) peter Nicolas Dore writes > In article <1993Jun24.172840.1755@afs.com> jon@afs.com writes: > > In article <1993Jun24.030101.8433@imani.cam.org> nico@imani.cam.org (Nicolas > > Dore) writes: > > | Hi folks. > > | > > | This may be old news, but has anyone else noticed what happens when you > > enlarge a > > | "Lip Service" panel in Mail.app? (didn't check with other apps) Seems the > > nice > > | folks at NeXT didn't pay the "Size" attribute of the button group to much > > | attention! > > Just noticing: I forgot to add "Make the wave view come out." _Then_ try resizing > the panel. See it now? I thought so (many people have written to tell me it's > fixed in 3.1. "There's one bug. Now let's move on to the XXXX next ones!"). 8^)> > > Ciao > > > What exactly is wrong with it? Perhaps they fixed it in 3.1 and I'm not seeing > > it. > > > > -- > > Jonathan It's a thousand classes, give or take a few, > > Hendry I'll be writing more in a week or two. > > <jon@afs.com> I can make it deeper if you like the style, > > I can change it round and I want to be an ObjectWare writer > -- > Nicolas Dore - CAREFUL! POLITICAL COMMENT ZONE!! - nico@imani.cam.org > > "If I can't dance, I don't want to be in your revolution!" < > > Emma Goldman (to Vladimir Ilitch Oulianov [Lenin], 1917) <
From: madler@cco.caltech.edu (Mark Adler) Newsgroups: comp.sys.next.bugs Subject: ISO 9660 bug fixed in 3.1? Date: 28 Jun 1993 22:01:36 GMT Organization: California Institute of Technology, Pasadena Message-ID: <20npo0INN8t3@gap.caltech.edu> Has anyone checked to see if the ISO 9660 CD-ROM bug in 3.0 (which I noticed on NASA Voyager CD's) is fixed in 3.1? Mark Adler madler@cco.caltech.edu
Newsgroups: comp.sys.next.bugs From: hb@raven.ix.de (Henning Behme) Subject: NewsGrazer 72.2 won't load news.answers Message-ID: <1993Jun30.080801.6544@ix.de> Sender: news@ix.de Organization: Verlag Heinz Heise GmbH & Co KG Date: Wed, 30 Jun 1993 08:08:01 GMT Hi, it hasn't happened from the beginning, but recently, whenever I start to read news.answers (looking for FAQs, that is), NewsGrazer starts to read (in) what is available, but after a while - not having completed the task, apparently - vanishes from the screen. Has anyone experienced the same and/or got an idea about possible reasons? Henning Behme iX Multiuser Multitasking Magazin * Verlag Heinz Heise Helstorfer Strasse 7 * 30625 Hannover * Germany hb@ix.de * +49 511 5352374 * +49 511 5352361 (Fax) ------- Religion decays, the icon remains. (Julian Barnes) -------
These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Marcel Waldvogel and Netfuture.ch.