Re: Very weird stuff on ftp.debian.org
I think it's an old BSD bug. The directory /debian.org, before anything else is mounted on it, has odd permissions. You have to unmount what's on top of it to see that, but I bet that's it. Bruce
Bug#1496: dpkg returns to dselect on SIGSTOP
Ian Murdock writes ("Bug#1496: dpkg returns to dselect on SIGSTOP"):
>Date: Thu, 28 Sep 95 20:28 BST
>From: Ian Jackson <[EMAIL PROTECTED]>
>
>> I was using dselect for the first time (for real), and it is very,
>> very nice. However, while it was upgrading my bash.deb, it stopped to
>> query about the confile '/etc/profile'. Fair enough, it was
>> different. I chose the 'Z' (?) option to suspend and investigate for
>> myself. dpkg was then stopped, but instead of starting a shell, it
>> returned to the dselect menu, after which I couldn't do anything, as
>> dpkg was locked.
>
>I had a bug report like this before. Can you reproduce it ? If you
>have two .deb files which have different versions of a conffile in it
>you can keep getting dpkg to prompt by editing the conffile once and
>then never answering `y' as you install them alternately.
>
> Were you doing this in the dselect automatically started after
> installing the base system? In this case, dselect is started from a
> shell script (namely, /root/.bash_profile). Could this be the
> problem?
Aaah, the light dawns. Ooops.
Ian M.: please arrange for the environment variable "DPKG_NO_TSTP" to
be set to some non-empty value (like "yes") when dselect is invoked in
this way. This will cause dpkg to invoke "$SHELL -i" instead (or "sh
-i" if SHELL isn't set).
This feature will be in 0.93.79.
Of course, any conffiles prompts at this stage reflect bugs in the
base disks.
Ian.
Re: ghostview and xxgdb
> > Did anyone volunteer to update ghostview and xxgdb before the release? > These are the last two packages that use the old "R6" naming scheme > for X11 packages. I'd like to update these packages before release. > > If nobody wants to do this, I will, but I already have too much to do. > I'd appreciate it if someone with a few hours in the evening could do > this. (It shouldn't take long.) I'll do it. tomorrow there will be new ones. Helmut -- Helmut Geyer[EMAIL PROTECTED]
Re: virtual packages and X11 vs. X11R
Ian Murdock writes ("Re: virtual packages and X11 vs. X11R"):
> Well, the X11R5 and X11R6 libraries weren't compatible, and it is
> likely the X11R6 and X11R7 libraries won't be, either. In this
> case, the packages will have to be updated. We don't want people
> thinking they can use an X11R6 package under X11R7, or vice versa.
But they *can*. They have to install the R6 xlib package, and the R7
server and xbase. Perhaps every X application should say
Depends: xlibraries-R6
Recommends: xbase
rather than
Depends: X11R6
?
Ian.
Re: Very weird stuff on ftp.debian.org
Ian Murdock writes ("Re: Very weird stuff on ftp.debian.org"):
> I noticed this today, too, but I forgot about it until now.
> Basically, after logging in and noticing "Permission denied", I did
> a "cd /", and then a "cd", and everything appeared to work normally
> after that.
>
> It shouldn't be happening, of course, but this is how I got around it.
Hmm. Unfortunately I'm trying to use rcp, which doesn't seem to be
able to cope.
Bruce Perens writes ("Re: Very weird stuff on ftp.debian.org "):
> I think it's an old BSD bug. The directory /debian.org, before anything
> else is mounted on it, has odd permissions. You have to unmount what's
> on top of it to see that, but I bet that's it.
Could the superuser not cd to /debian.org and chmod .. ?
I suppose that taking the machine down to fix the underlying fs is
probably not advisable.
Also, the /debian.org directory (rather than the mountpoint) ought not
to be world-writeable, really ...
PS: re the tarring and/or gzipping of the whole archive: can you not
disable .tar.gz without disabling plain .tar (the latter is quite an
efficient way to do things if you really want the whole archive) ?
Ian.
Re: diald for Debian
[EMAIL PROTECTED] writes ("Re: diald for Debian "):
> I haven't been able to upload diald yet. I see the server is back, but
> I see no Incoming directory (this is where the other Ian suggested I
> upload the package). Also, he mentioned using dchanges ... but I can
> see no dchanges in binary/misc (although it was announced on
> debian-something at some stage).
>
> I know there were problems with your machine and the server. If these
> quirks are going to "fix themselves" over the next few days ignore my
> message. If I'm missing something, please set me straight.
You're confusing me with someone else. I'm just the dpkg maintainer,
I'm nothing to do with the ftp site administration.
Ian.
Re: ghostview and xxgdb
From: Helmut Geyer <[EMAIL PROTECTED]> Date: Fri, 29 Sep 1995 08:24:49 +0100 (MET) > Did anyone volunteer to update ghostview and xxgdb before the release? > These are the last two packages that use the old "R6" naming scheme > for X11 packages. I'd like to update these packages before release. > > If nobody wants to do this, I will, but I already have too much to do. > I'd appreciate it if someone with a few hours in the evening could do > this. (It shouldn't take long.) I'll do it. tomorrow there will be new ones. Great! Thanks.
Bug#1508: Segmentation fault while installing...
Package: dpkg Version: 0.93.77 Hi... I installed dpkg 77 fine (before the R6 release) and installed a few packages no problem... Now, my mirror script deleted all the files on my hard drive and is re-mirroring debian again because the files have been moved to the dir debian-0.93 (but that's a different story). In any case on BOTH my machines now, I noticed I was still running adduser version 1.93. I noticed adduser 1.94 in base so I decided to install it, and here's what followed: [EMAIL PROTECTED]:2:/pub/linux/debian/debian-0.93/binary/base] dpkg -i adduser- 1.94-1.deb (Reading database ... 6380 files and directories currently installed.) Preparing to replace adduser (using adduser-1.94-1.deb) ... Unpacking replacement adduser ... Segmentation fault (core dumped) [EMAIL PROTECTED]:2:/pub/linux/debian/debian-0.93/binary/base] This happened on BOTH my machines which currently run Debian 0.93R5. I wondered if is was adduser itself, so I tried installing acct again from the admin directory - same error. It couldn't possibly be something wrong hardware wise becuase it's happening on both my PCs... Ideas? ...Karl -- | PO Box 828 Office: (09)316-3036 Fax: (09)381-3909 |OWER INTERNET SERVICES Canning Bridge After Hours: 015-779-828 WA, 6153 Sales Support: [EMAIL PROTECTED] Internet Service Providers and Networking Solutions
Bug#1505: setterm is missing
Bruce Perens: > I think there was a copyright problem with "setterm" that caused us to > remove it from the distribution a long time ago. If I recall correctly, > it didn't allow distribution for a fee, which is of course essential to > our CD-ROM redistributors. Hmm, setterm is distributed on countless Slackware CD-ROMs at least... Aren't we too paranoid about copyrights? Now that I have read the source, it says that the "conditions of use, modification and redistribution are contained in the file COPYRIGHT that forms part of this distribution" but I can't find this file anywhere. Has anyone tried to contact the author (Gordon Irlam [EMAIL PROTECTED]) about this? From a quick test using EXPN it seems that his e-mail address is probably still valid even though setterm was written in 1990... Marek
Re: Very weird stuff on ftp.debian.org
On Fri, 29 Sep 1995, Ian Jackson wrote: > > I suppose that taking the machine down to fix the underlying fs is > probably not advisable. > Already been done. > Also, the /debian.org directory (rather than the mountpoint) ought not > to be world-writeable, really ... Well we were tring to figure it out. > > PS: re the tarring and/or gzipping of the whole archive: can you not > disable .tar.gz without disabling plain .tar (the latter is quite an > efficient way to do things if you really want the whole archive) ? > yes .tar is nice it is the .tar.gz thats killing me. After the note last night the problem has stopped. Thanks Matthew S. Bailey [EMAIL PROTECTED]
Re: Very weird stuff on ftp.debian.org
On Fri, 29 Sep 1995, Ian Jackson wrote: > Hmm. Unfortunately I'm trying to use rcp, which doesn't seem to be > able to cope. > [..] > > I suppose that taking the machine down to fix the underlying fs is > probably not advisable. > Give your cron a shot again, Ian. I had to fix a few things and hope everything else is also fixed as result..(fingers crossed:)) From what I can tell, I see no problems. I tried an rsh from a remote site and was able to get through no problem but have not had a chance to try an rcp from remote. I would think it "should" work but that was "should"..:) > Also, the /debian.org directory (rather than the mountpoint) ought not > to be world-writeable, really ... > That was a temp. test to see if that would fix things but didn't change a thing so it has already been changed. Right now I have it set with world read and execute and no_write. > PS: re the tarring and/or gzipping of the whole archive: can you not > disable .tar.gz without disabling plain .tar (the latter is quite an > efficient way to do things if you really want the whole archive) ? > Having the .tar.gz option was thought of as a way that we can provide people that don't already have gzip to retreive a tar.gz as a .tar file instead. Running a virtual ftp server as ours is means that no longer does ftp use the gzip found in ftp/bin, but the system level gzip. Personally I don't really want to live without gzip on the system level. :) As we are running a virtual ftp server removing just gzip would mean that the system level gzip program would have to be removed. Going virtual means that no longer does ftp just execute gzip from the ftp/bin directory but the real system level gzip. David David Brinks - Assist Sys Admin. [EMAIL PROTECTED] 422 Pearce HallMail maintainer of: archive, ftpadmin Computer Science Department www, and sometimes other accounts Central Michigan University used for admin purposes. ---All opinions, and possibly some of the facts, are strictly of my own.
Sizes and Packages in dselect
Hello developers, * diskspace requirements (displayed/registered) I was wondering what happens when someone is running dpkg / dselect to install some packages and is running out of diskspace. Perhaps it is usefull to add a system able to calculate/display the required diskspace for the selected packages. Is it there allready and I never noticed because I never run in a situation with not enough diskspace? This will become more difficult when the different directory's where parts of the packages are going to are mounted on seperate drive. Most of the people will probably have a big drive where /usr is mounted on and it will probably do, but with more and more packages coming up and networking setups are being considered this might be worth a thought. * dselect creating it's list from several package files dselect is building it's selection list from a file (Packages or something like that). What about making it possible that dselect builds it selection list from different Packages files or optionally from all the Packages files it recursively finds from some root directory. Perhaps this implemented now and I'm not aware. If so please ignore this, otherwise please comment on this one. At this moment a handy solution is probably catting all Packages files into one. I have to admid that the clear line between distribution packages and non-free packages is becoming more vagely and this is perhaps undermining the original concept of Debian GNU/linux. If implementing something like this is going to happen an extra notifier of what is in the distribution and what not might be added within dselect/dpkg? BTW: good thing that non-free has a ms-dos (why not just dos/fat?) subdirectory for downloading the packages according to the eight dot three naming scheme. Erick Very off topic/PS: someone ever used snns (stuttgart neural network simulator), I'm trying to get version 3.0 running. Sorry I ever asked, please don't flame.
Bug#1510: prermission denied /usr/doc
Package: several packages Version: ? Libc: libc.so.4.6.27 Kernel etc: Linux eb 1.2.10 #1 Tue Jun 13 18:37:28 EST 1995 i486 Reporter: repair 0.1 Subject: accessing /usr/doc/subdirs Following command showed this, indicating that these subdirectory's don't have the correct permissions set. $ ls /usr/doc/ -l | grep drw- drw-r--r-- 3 root root 1024 Apr 10 23:29 acs drw-r--r-- 2 root root 1024 Apr 11 01:11 pcb drw-r--r-- 2 root root 1024 Jul 16 00:21 spice $ dpkg --status acs Package: acs Status: install ok installed Priority: extra Section: electronics Maintainer: Bruce Perens <[EMAIL PROTECTED]> Version: 016 Revision: 1 Description: Al's Circuit Simulator $ dpkg --status pcb Package: pcb Status: install ok installed Priority: extra Section: electronics Maintainer: Bruce Perens <[EMAIL PROTECTED]> Version: 1.1.1.2 Revision: 2 Description: Printed Circuit Board Design Program $ dpkg --status spice Package: spice Status: install ok installed Maintainer: Bruce Perens <[EMAIL PROTECTED]> Version: 3f4 Revision: 0 Description: Circuit Simulator
Bug#1509: error 139 with jgraph/core dump
Package: jgraph Version: 83-5 Libc: libc.so.4.6.27 Kernel etc: Linux eb 1.2.10 #1 Tue Jun 13 18:37:28 EST 1995 i486 Reporter: repair 0.1 Subject: Error 139 with jgraph I wanted to test jgraph with the examples which came with it and did the following: $ cp /usr/doc/examples/jgraph/* . $ gzip -d * $ make jgraph < cube.jgr > cube.eps make: *** [cube.eps] Error 139 So I tried manually $ jgraph < cube.jgr > cube.eps Segmentation fault (core dumped) I tried another file with same segmentation fault.
Re: FTP archive
On Thu, 28 Sep 95 19:20 EST, Ian Murdock <[EMAIL PROTECTED]> said: >auto-pgp is not pgp. It is still under GPL. Please move it back > into the distribution. Ian Jackson is the author. You can ask him > yourself if you are in doubt. :-) > Actually, Ian asked me to move it to non-free until there is a > replacement PGP, since it depends on PGP. Really? That's strange. I wonder why. Ian? --Mike
Bug#1505: setterm is missing
Hm. There seems to be some confusion about who the author is. Unfortunately, we can't be _too_ careful about copyrights. For the CD-ROM distributiors, there is a significant risk of getting sued (and losing the lawsuit) if we haven't verified that we have the right to distribute these programs. The publicity would not be pretty. As a fledgeling CD-ROM distributor myself, I have an incentive to get this stuff right. I can't presently afford to get sued. As to the program existing in Slackware, etc., for countless distributions, perhaps they got individual permission from the author, or perhaps they overlooked the problem or they don't care. We choose to operate differently. I do know that there are some very sloppy CD-ROM manufacturers out there. I'm not saying who they are. They are definitely selling things that they have no legal right to sell. Thanks Bruce
Re: diald for Debian
ftp://ftp.debian.org/debian/private/project/Incoming seems to be OK now. Also, there is a copy of the "dchanges" package in it. Thanks Bruce -- -- Attention Radio Amateurs: For information on "Linux for Hams", -- read the WWW page http://www.hams.com/LinuxForHams, -- or e-mail the word "help" to [EMAIL PROTECTED]
Bug#1511: Packages vs. packages for ms-dos compatability
PACKAGE: dpkg (and the ftp site) The "Packages" and "Packages.gz" files do not have ms-dos compatible filenames. Depending on the tools used, manipulating these files on, MSDOS filesystems can be a problem. I suggest that these files be renamed to "packages" and "packages.gz", respectively. they should be placed on the ftp site with these ms-dos compatible names. dselect should probably look for the files first under these names and, if that fails, under their current names. Files intended to have ms-dos compatible filenames should have names with an 8.3 format, and using only the chars [a-z0-9] and '-', '_', and possibly a few other non-alpha chars. Two things known to cause problems are names exceeding 8.3 limits and names having uppercase characters.
Bug#1512: sendmail man pages missing.
Package: sendmail Version: sendmail-8.6.12-7.deb "dpkg --listfiles sendmail" shows no man pages in the latest binary package. And in fact, when I install the package, "man sendmail" doesn't work. --Rob.
Bug#1505: setterm is missing
For what it's worth: I've been pretty strung out lately (illness, deadlines, lack of sleep). Please take anything I mailed out yesterday with at least a grain of salt. [Not that you shouldn't in any event, but especially for Sep 28 95]. Sorry about that. -- Raul
Re: Very weird stuff on ftp.debian.org
David Brinks writes ("Re: Very weird stuff on ftp.debian.org"):
> Give your cron a shot again, Ian. [...]
It succeeded some time earlier today, but I've only just got around to
answering my mail. Thanks for getting it fixed.
The new trn, dpkg and texinfo packages are now available in
project/Incoming, or in the binary/* directories if Ian M. has moved
them out already.
Ian.
Re: FTP archive
Michael E. Deisher writes ("Re: FTP archive"):
> On Thu, 28 Sep 95 19:20 EST, Ian Murdock <[EMAIL PROTECTED]> said:
> >auto-pgp is not pgp. It is still under GPL. Please move it back
> > into the distribution. Ian Jackson is the author. You can ask him
> > yourself if you are in doubt. :-)
> > Actually, Ian asked me to move it to non-free until there is a
> > replacement PGP, since it depends on PGP.
>
> Really? That's strange. I wonder why. Ian?
Because if you try to select auto-pgp dselect will confuse the user by
moaning about the dependencies on non-free, and then dpkg will confuse
them even more by failing to configure.
My objection to its inclusion in the main archive is nothing to do
with my status as its author. With my "Auto-PGP author" hat on I'm
perfectly happy for Debian to include it - indeed, I'd like to see it
included.
However, with my "dselect maintainer" hat on I have to ask that it not
be put in the main archive until we have a version of PGP that can go
there too. If you think about it this is a perfectly sensible
requirement of our package management scheme - packages shouldn't
depend on things which are "less available" or "of lower priority"
than they are.
Ian.
Re: Bug#1508: Segmentation fault while installing...
Karl Ferguson writes ("Bug#1508: Segmentation fault while installing..."):
> [...]
> [EMAIL PROTECTED]:2:/pub/linux/debian/debian-0.93/binary/base] dpkg -i
> adduser-
> 1.94-1.deb
> (Reading database ... 6380 files and directories currently installed.)
> Preparing to replace adduser (using adduser-1.94-1.deb) ...
> Unpacking replacement adduser ...
> Segmentation fault (core dumped)
> [EMAIL PROTECTED]:2:/pub/linux/debian/debian-0.93/binary/base]
You're running NIS, aren't you ? Read on ...
> This happened on BOTH my machines which currently run Debian 0.93R5. I
> wondered if is was adduser itself, so I tried installing acct again from the
> admin directory - same error. It couldn't possibly be something wrong
> hardware wise becuase it's happening on both my PCs...
I asked Karl for a copy of the coredump, which he kindly sent me:
chiark:d> gdb ../../things/debian/dpkg-0.93.77/main/dpkg core
GDB is free software and you are welcome to distribute copies of it
under certain conditions; type "show copying" to see the conditions.
There is absolutely no warranty for GDB; type "show warranty" for details.
GDB 4.14 (i486-debian-linux), Copyright 1995 Free Software Foundation, Inc...
warning: exec file is newer than core file.
Core was generated by `dpkg'.
#0 0x60014516 in end ()
(gdb) where
#0 0x60014516 in end ()
#1 0x76b00 in end ()
#2 0x60014361 in end ()
#3 0x1767f in DecodeTarHeader (block=0xb670 "", d=0xb64c)
at tarfn.c:61
#4 0x179be in TarExtractor (userData=0x24294, functions=0x7354) at tarfn.c:110
#5 0x8ffa in process_archive (
filename=0xbeb8
"/pub/linux/debian/debian-0.93/binary/base/adduser-1.94-1.deb") at
processarc.c:481
#6 0x723f in archivefiles (argv=0xbe78) at archives.c:556
#7 0x251d in main (argc=3, argv=0xbe78) at main.c:293
(gdb) up
#1 0x76b00 in end ()
(gdb) up
#2 0x60014361 in end ()
(gdb) up
#3 0x1767f in DecodeTarHeader (block=0xb670 "", d=0xb64c)
at tarfn.c:61
61 struct group * group = getgrnam(h->GroupName);
(gdb) print h
$1 = (TarHeader *) 0xb670
(gdb) print *h
$2 = {Name = '\000' , Mode = "\000\000\000\000\000\000\000",
UserID = "\000\000\000\000\000\000\000",
GroupID = "\000\000\000\000\000\000\000", Size = '\000' ,
ModificationTime = '\000' ,
Checksum = "\000\000\000\000\000\000\000", LinkFlag = 0 '\000',
LinkName = '\000' ,
MagicNumber = "\000\000\000\000\000\000\000",
UserName = '\000' , GroupName = '\000' ,
MajorDevice = "\000\000\000\000\000\000\000",
MinorDevice = "\000\000\000\000\000\000\000"}
(gdb) quit
chiark:d>
I think this is another user who has been bitten by Bug#1278: getgrnam
dumps core if you are using NIS and feed it an empty string (NB, not a
null pointer, an empty string).
Can the libc maintainer please fix this ? Pretty please ?
This bug probably makes dpkg well-nigh unuseable on systems using NIS.
Karl, the solution is to take the `+' out of your group or passwd file
(I don't know whether you have to take it out of just one or out of
both) and then to try it again.
If you're *not* running NIS please let me know, as then I have more
investigating to do.
I'm closing this bug report, as I think it's a duplicate of #1278.
Ian.
Bug#1509: error 139 with jgraph/core dump
> So I tried manually
> $ jgraph < cube.jgr > cube.eps
> Segmentation fault (core dumped)
I wonder if someone with better and/or more recent low level
debugging skills than I could take a quick look at this.
Jgraph appears to be getting a segfault before reaching
main(). It looks like either I'm missing something really
obvious, or there's a wider problem here.
I installed the last known working version of jgraph, and
it runs fine. I rebuilt from those sources, and it segfaults.
I rebuilt using -g and ran it under gdb, and it gets a segfault
before reaching main(). It also segfaults before reaching
main() if built from current sources, which haven't changed
substantively from the last known working sources.
Here's a log of my gdb session:
Script started on Fri Sep 29 17:58:24 1995
[?1h=bash# gdb jgraph
[?1l>GDB is free software and you are welcome to distribute copies of it
under certain conditions; type "show copying" to see the conditions.
There is absolutely no warranty for GDB; type "show warranty" for details.
GDB 4.14 (i486-debian-linux), Copyright 1995 Free Software Foundation, Inc...
(gdb) break main
Breakpoint 1 at 0xd44b: file jgraph.c, line 332.
(gdb) run
Starting program: /home/debian/packages/work/jgraph-83/old/jgraph-83/jgraph
Program received signal SIGSEGV, Segmentation fault.
0x62f013be in end ()
(gdb) where
#0 0x62f013be in end ()
#1 0x0 in ?? ()
(gdb) list
323 int i;
324 int show, pp;
325 int landscape;
326 int comments;
327 int nfiles;
328
329 #ifdef VMS
330 IOSTREAM = stdin; /* for token.c -hdd */
331 #endif
332 show = 0;
(gdb) list 318
313 newg->epifile = 0;
314 if (first(gs) == nil(gs)) newg->page = 1; else newg->page =
last(gs)->page+1;
315 insert(newg, gs);
316 }
317
318 main(argc, argv)
319 int argc;
320 char **argv;
321 {
322 Graphs gs;
(gdb)
323 int i;
324 int show, pp;
325 int landscape;
326 int comments;
327 int nfiles;
328
329 #ifdef VMS
330 IOSTREAM = stdin; /* for token.c -hdd */
331 #endif
332 show = 0;
(gdb) quit
The program is running. Quit anyway (and kill it)? (y or n) y
[?1h=bash#
Script done on Fri Sep 29 17:59:37 1995
