Linux-Misc Digest #808, Volume #27 Tue, 8 May 01 09:13:02 EDT
Contents:
Re: WordPerfect wouldn't start the graphical installation on RH 7.1 ("Wayne Osborn")
pthread lib error (fow99)
Re: news/mail clients ("Wayne Osborn")
Re: How come... (Harold Stevens US.972.952.3293)
Re: database (Stefano Ghirlanda)
Re: wie funktioniert apt-ftparchive ? (SammyTheSnake)
2.4.3 ETHER_EXPRESS_PRO driver buggy? (Georg Schwarz)
Re: Restricting Shell from SSH login (Dhar)
Re: removing linux ("dark dragon")
RedHat 7.1 shared memory problem? (Chris Leahy)
Re: simple ownership and permission question (Lew Pitcher)
forcing a dynamic executable to use given shared library - how? (Michal Szymanski)
Re: PPP dialing program? (Pun Kuan Tou)
Re: cdrecord only works with SCSI drives? ([EMAIL PROTECTED])
Re: How come... (Jean-David Beyer)
Re: cdrecord only works with SCSI drives? (Michael Perry)
Did YOUR distro install Mesa or OpenGL? (Andrew Purugganan)
gnu find not pruning (Volker Apelt)
----------------------------------------------------------------------------
From: "Wayne Osborn" <[EMAIL PROTECTED]>
Subject: Re: WordPerfect wouldn't start the graphical installation on RH 7.1
Date: Tue, 08 May 2001 18:23:30 +0800
In article <[EMAIL PROTECTED]>,
"ThanhVu Nguyen" <[EMAIL PROTECTED]> wrote:
> Hello there, thanks alot - got it working
Ok. Let me know how you find WP8. One anoying bug I have not been able to
sort is the size of the WordPerfect opening dialog. It keeps resorting to
an unusable size.
--
Wayne A. Osborn, SCADA Engineer.[dnar AT iinet DOT net DOT au]
Registered Linux User #212818. [2.2.16-22-Win4Lin-686] [i686]
6:20pm up 1 day, 20:11, 1 user, load average: 1.00, 1.05, 1.31
...I've finally learned what "upward compatible" means. It means we get to
keep all our old mistakes.
-- Dennie van Tassel
------------------------------
From: fow99 <[EMAIL PROTECTED]>
Subject: pthread lib error
Date: Tue, 8 May 2001 11:28:36 +0100
Hi all,
I am having problem linking to the pthread library. I have a program which
uses pthread_create, pthread_mutex_init... thread_cond_destory and
pthread_mutex_destory. The other functions are linked fine. But gcc gave
out the following error in the end. The program used to be able to link
fine and worked fine. I really have no clue at all what is going on.
The following are the error messages:
=================
/tmp/cco6393p.o(.text+0x726): undefined reference to `pthread_cond_destory'
/tmp/cco6393p.o(.text+0x746): undefined reference to `pthread_mutex_destory'
------------------------------
From: "Wayne Osborn" <[EMAIL PROTECTED]>
Subject: Re: news/mail clients
Date: Tue, 08 May 2001 18:27:18 +0800
In article <[EMAIL PROTECTED]>, "andi"
<[EMAIL PROTECTED]> wrote:
> I'm in the process of a slow migration to linux - but I need
> recommendations regarding mail and news clients.
>
I use Spruce, but have also tried Balsa. Both are reasonable, but not
without problems.
--
Wayne A. Osborn, SCADA Engineer.[dnar AT iinet DOT net DOT au]
Registered Linux User #212818. [2.2.16-22-Win4Lin-686] [i686]
6:20pm up 1 day, 20:11, 1 user, load average: 1.00, 1.05, 1.31
...I've finally learned what "upward compatible" means. It means we get to
keep all our old mistakes.
-- Dennie van Tassel
------------------------------
From: [EMAIL PROTECTED] (Harold Stevens US.972.952.3293)
Subject: Re: How come...
Date: Tue, 08 May 2001 11:07:32 GMT
In <HUHJ6.20$[EMAIL PROTECTED]>, kalasend:
>How come until today, Linux (or in general Unix) still does not have the
>"undelete" feature?
This is a troll, right? If you're serious, consider this as well:
How come until today, Linux (or in general Unix) still does not have the
"defrag" feature?
Supposedly, *nix defrags exist, but I've not seen/required/used them. As
others suggested it's trivial to roll your own undelete scripts. So, how
do banks and other trivial pursuits do without "features" like this?
Undelete is for those who cannot manage their own backups. Defrag is for
file systems that cannot manage their own garbage. If you need them, you
should get an OS with them. Because *nix in general screams, and doesn't
have time for coddling. Business types seem to prefer this *nix style.
In short: A "feature" to some is a "PITA" to others.
--
Regards, Weird (Harold Stevens) * IMPORTANT EMAIL INFO FOLLOWS *
Pardon the bogus email domain (dseg etc.) in place for spambots.
Really it's (wyrd) at raytheon, dotted with com. DO NOT SPAM IT.
Standard Disclaimer: These are my opinions not Raytheon Company.
------------------------------
From: Stefano Ghirlanda <[EMAIL PROTECTED]>
Subject: Re: database
Date: 08 May 2001 13:18:48 +0200
Jef Peeraer <[EMAIL PROTECTED]> writes:
> To start with, MySql is only free for personal use
This is not correct, ASAIK. MySQL is covered by the GPL and LGPL
(depends which bits of it). As a commercial company, you *can*, if you
want, buy a commercial license from MySQL AB (the company) if you
want to use MySQL without adhering to the GPL and LGPL licenses. That
is, by buying the commercial license you will not have not disclose
your source code even if it includes MySQL code.
--
Stefano - Hodie postridie Nonas Maias MMI est
------------------------------
From: [EMAIL PROTECTED] (SammyTheSnake)
Subject: Re: wie funktioniert apt-ftparchive ?
Date: Tue, 8 May 2001 12:19:48 +0100
In article <9d6j28$rl1$00$[EMAIL PROTECTED]>, Thomas Weidner wrote:
>hi NG
>
>ich hab mich dazu in den man pages erkundigt,bin aber nicht schlau
>geworden,kann mir da jemand helfen ?
in english:
Subject: how does apt-ftparchive work?
I have looked for it in the man pages, but didn't get enlightened, can anyone
help me with it?
incidentally, I'm not sure what it's for myself, isn't it just a helper ap
for apt-get?
Cheers & God bless
SammyTheSnake(can't find my dictionary, pants!)
--
Sam.Penny @ Ntlworld.com | Looking for a computer related
Linux, Hardware & Juggling specialist :-) | job, if you can help, e-mail me :)
Wheels: bike, 'ickle bike, and unicycle. | /o \/ Working on 5 ball 1/2 shower
Boxen: K6-266@300, dual Celery500 & Nx486 | \__/\ & some 6 / 7 ball exercises
------------------------------
From: [EMAIL PROTECTED] (Georg Schwarz)
Subject: 2.4.3 ETHER_EXPRESS_PRO driver buggy?
Date: 8 May 2001 11:29:16 GMT
I'm running a Pentium 100 with an Intel EtherExpress Pro/10+ ISA card.
With Kernel 2.2.X (RH 7.0) there was no problem whatsoever. Now after
upgrading to RH 7.1 there are severe network problems, even after I have
built a custom 2.4.3 kernel.
The card is detected perfectly, and at first, after a reboot, everything
runs fine. Then after a few minutes (or a few hours) NIC performance
suddenly gets very bad. When I ping another host on the LAN the response
times I get from ping are quite exactly the delay times between each ping
(dfefault 1 s); e.g. if I'm using ping -i 0.1 then response times are
around 100 ms, etc. This seems to be a driver issue.
It can also appear that ping don't go through at all any longer. Using
tethereal one sees that even the ARP request stays without reply.
Running tethereal on other hosts on the LAN it seems that the ethernet
packets on the machine in question do not make it onto the ethernet, i.e.
they're not seen from the other machines (they might as well be damaged).
On the other hand, the other machines' traffic can be sniffed on the
EtherExpress card. Rebooting reliably does away with the problem for at
least a short time. Sometimes the problem goes away on its own after a
while. Could it be a buffer issue in the driver somewhere?
ifconfig shows thousands of TX "carrier" counts on the EtherExpress
interface.
Any idea what could be wrong? Could it be a douplex/simplex issue?
--
Georg Schwarz http://home.pages.de/~schwarz/
[EMAIL PROTECTED] +49 177 2437545
------------------------------
From: Dhar <[EMAIL PROTECTED]>
Subject: Re: Restricting Shell from SSH login
Date: Tue, 08 May 2001 11:30:04 -0000
> books, but don't know where to begin. So, is it possible to give these > users a
>shell that can run only two or three commands? Any guidance > would be great.
>Thanks. Here is what you can try. In the passwd file, against the user give the
>shell /bin/bash2 -r. This is a restricted shell. Then set the path of the user to
>a directory which holds just the commands you want them to execute. So ideally what I
>would do would be have a .bashrc which looks like this: cd imprison export
>PATH=/home/guest/restricted Then I would create the directories imprison and
>restricted. The directory restricted would hold the executables I want them to
>execute. Remember dont put in executables like vim/pine etc. They can be used to
>break out of a restricted shell... Regards Sumit Dhar PS: In case of further
>doubts please mail me at dhar at sawaal.com
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: "dark dragon" <[EMAIL PROTECTED]>
Subject: Re: removing linux
Date: Thu, 3 May 2001 16:38:01 -0400
Crossposted-To: alt.os.linux,comp.os.linux.setup
what linux CD? i'm guessing maybe the CD got scratched or something. when
you get a new CD, it should be a simple matter to use your old linux
partitions for the new install
"Kenny@BUI" <[EMAIL PROTECTED]> wrote in message
news:UtfI6.331$[EMAIL PROTECTED]...
> hello,
> our box had rh7.0 on it. my buddy used dos diskettes to book the computer
> and format the drive. Linux will not boot up now. we tried booting with
both
> Linux and dos boot\installation disks. our goal is to remove Linux and do
a
> fresh install.
> the cd drive kicks the cd out all the time. it continuously says that
Linux
> is not found on the cd and ejects the cd.
>
> thank you,
> Kenny.
>
>
>
>
------------------------------
From: Chris Leahy <[EMAIL PROTECTED]>
Subject: RedHat 7.1 shared memory problem?
Date: Tue, 08 May 2001 07:53:18 -0400
==============4B197205A8C6CCAC66D7967B
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Hello all,
I have a newly installed RedHat 7.1 system with a custom built 2.4.4
kernel.
I have noticed serious performance degradation after a relatively short
uptime < 1 day
Ive noticed that the system starts swapping FAR FAR sooner than the 6.2
System with 2.2.18
kernel I upgraded it from.
Previously, it never really touched the swap space and now it routinely
goes 150 - 200+ MB into the swap partition.
The following is output from top 3 minutes after a fresh boot. You can
see that there is "0k shrd"
Did I leave something out of my kernel for shared memory or is top just
not working properly with the 2.4.x kernel?
This figure remains 0 no matter how long the system is up.
Mem: 126672K av, 69420K used, 57252K free, 0K shrd, 4340K
buff
Thanks,
Chris
--
==========================================================
Christopher Leahy | [EMAIL PROTECTED]
UNIX System Administrator | http://www.zoltanium.com
Zoltanium (aka /dev/null) | Voice (610)408-0151
==========================================================
==============4B197205A8C6CCAC66D7967B
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Hello all,
<p>I have a newly installed RedHat 7.1 system with a custom built 2.4.4
kernel.
<br>I have noticed serious performance degradation after a relatively short
uptime < 1 day
<br>Ive noticed that the system starts swapping FAR FAR sooner than
the 6.2 System with 2.2.18
<br>kernel I upgraded it from.
<br>Previously, it never really touched the swap space and now it routinely
goes 150 - 200+ MB into the swap partition.
<br>The following is output from top 3 minutes after a fresh boot. You
can see that there is "0k shrd"
<p>Did I leave something out of my kernel for shared memory or is top just
not working properly with the 2.4.x kernel?
<br>This figure remains 0 no matter how long the system is up.
<br>
<p>Mem: 126672K av, 69420K used, 57252K
free, 0K shrd, 4340K
buff
<br>
<p>Thanks,
<br>
<p>Chris
<br>
<br>
<br>
<pre>--
==========================================================
Christopher
Leahy |
[EMAIL PROTECTED]
UNIX System Administrator | <A
HREF="http://www.zoltanium.com";>http://www.zoltanium.com</A>
Zoltanium (aka /dev/null) | Voice (610)408-0151
==========================================================</pre>
</html>
==============4B197205A8C6CCAC66D7967B==
------------------------------
From: [EMAIL PROTECTED] (Lew Pitcher)
Subject: Re: simple ownership and permission question
Reply-To: [EMAIL PROTECTED]
Date: Tue, 08 May 2001 12:11:40 GMT
On Tue, 08 May 2001 16:34:17 +1000, David <[EMAIL PROTECTED]>
wrote:
>TRY wrote:
>
>> Kirk R. Wythers wrote ...
>>
>>> I'm trying to set the permissions in a mounted dos directory (/dose). It
>>> is owner and group root. I want to allow users to save downloads to a
>>> subdirectory /dose/tmp.
>>>
>>> I tried to chmod (as root). It doesnt have any affect. I also tried
>>> chown, and chgrp, they both report "Operation not permitted" .
>>>
>>> What am I missing here?
>>>
>>> thanks,
>>>
>>> Kirk
>>
>>
>> As DOS (i.e. FAT/VFAT) partitions do not hold any user information, they are
>> collectively set for all files by the mount command. If you do not specify
>> any options, the owner/group for all files are taken from the process (i.e.
>> user) that runs the mount command.
>>
>> The default behaviour can be changed with the following options: uid=UserID
>> and gid=GroupID, like "mount -t vfat -o uid=100,gid=500 /dev/fd
>> /mnt/floppy". Observe that uid= and gid= does not take user or group names
>> (you cannot say uid=root). If you mount through fstab (like "mount
>> /mnt/floppy"), you can specify the options in the fourth field of
>> "/etc/fstab".
>>
>> /TRY
>
>I have a similar problem trying to set up a user to access
> a Zip drive - any ideas on that - I just get a "Root mount
>permission error"
add 'umask=0' to the mount options...
mount -t vfat -o uid=100,gid=500,umask=0 /dev/fd0 /mnt/floppy
or in /etc/fstab
/dev/fd0 /mnt/floppy vfat uid=100,gid=500,umask=0 0 0
Lew Pitcher, Information Technology Consultant, Toronto Dominion Bank Financial Group
([EMAIL PROTECTED])
(Opinions expressed are my own, not my employer's.)
------------------------------
From: [EMAIL PROTECTED] (Michal Szymanski)
Subject: forcing a dynamic executable to use given shared library - how?
Date: 8 May 2001 12:14:51 GMT
Hi,
I've got into following problem: RH6.2 installs by default the package
"libungif" which is used by a dozen or so other packages. The package
contains shared libraries, including "/usr/lib/libgif.so.3".
Now I need to use utilities from "giflib" package, e.g. "raw2gif".
This package also includes "libgif.so.3" but these two libraries are
really different. I've installed it relocated to /usr/local/lib.
I'm trying to force linux loader to use /usr/local/lib/libgif.so.3
for raw2gif instead of /usr/lib/libgif.so.3. I've already tried
setting LD_LIBRARY_PATH, LD_RUN_PATH to /usr/local/lib. No success.
Also putting /usr/local/lib in the first place of "/etc/ld.so.conf"
does not help although, AFAIK the default directories (/lib, /usr/lib)
should in principle be searched for a library *after* those listed
in /etc/ld.so.conf
Any hints how to make it work (other than removing the library from
/usr/lib - that surely *does* work).
regards, Michal.
--
Michal Szymanski ([EMAIL PROTECTED])
Warsaw University Observatory, Warszawa, POLAND
------------------------------
From: [EMAIL PROTECTED] (Pun Kuan Tou)
Subject: Re: PPP dialing program?
Date: 8 May 2001 12:13:27 GMT
Reply-To: [EMAIL PROTECTED]
On 7 May 2001 21:15:31 GMT, Ken Arromdee <[EMAIL PROTECTED]> wrote:
>I installed Mandrake 8 and couldn't find a graphical PPP dialing program
>included. I'm not running Gnome/KDE. (I tried kppp anyway but I got some
>odd problems and it ultimately didn't connect.) Has anyone got any
>recommendations for good programs? Typing 'ifup ppp0' in a root shell every
>time is getting annoying.
a nice tools in text mode is call "wvdial"
--
============================== ______ _____ __ __
Email: [EMAIL PROTECTED] | || || | |
ICQ: 9416756 `-| |-'| | || | |
OS: GNU/Linux debian 2.2r3 | | | | || | |
============================== `--' `-----'`-----'
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: cdrecord only works with SCSI drives?
Date: Tue, 08 May 2001 12:31:38 GMT
Ryan Joseph <[EMAIL PROTECTED]> writes:
> I recently bought an HP 9100 series cd-rw which connects via IDE.
> Now, when I downloaded cdrecord and a bunch of different front-ends,
> all of them (the front ends) reported that they couldn't open
> '/dev/pg0' and then said something about SCSI drives!
> I thought that cdrecord was supposed to work with almost all types
> of ATAPI/IDE drives as well - or did I just misread somewhere?
cdrecord only knows SCSI, which means you have to set up your system
to access the IDE CDROM via the SCSI emulation subsystem.
I've always done so by creating a custom kernel that has IDE CD
support turned off and SCSI emulation turned on, thus:
CONFIG_BLK_DEV_IDESCSI=y
# CONFIG_BLK_DEV_IDECD is not set
It is reportedly also possible to handle this via some custom boot
parameters, which requires no kernel recompiles.
--
(concatenate 'string "cbbrowne" "@ntlug.org")
http://vip.hyperusa.com/~cbbrowne/resume.html
"Sigh. I like to think it's just the Linux people who want to be on
the `leading edge' so bad they walk right off the precipice."
-- Craig E. Groeschel
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: How come...
Date: Tue, 08 May 2001 08:34:56 -0400
kalasend at YAHOO dot COM wrote:
>
> How come until today, Linux (or in general Unix) still does not have the
> "undelete" feature?
It does not need one. Do not delete files you wish to keep. If you
find yourself typing rm file_to_keep too often, you can alias rm to rm
-i and it will ask you if you really mean it before each delete. On my
machine, for example, the .bashrc for root contains (among other
things):
alias rm='rm -i'
alias cp='cp -i'
alias mv='mv -i'
I do not do this for my users' accounts because I find it a nuisance.
In the rare cases where you delete a needed file, or where you
overwrite, one by mistake, you can always restore it from your backup.
My backups are done every morning at 1:03AM (or so), so I can always
restore to something less than 24 hours old.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey http://counter.li.org
^^-^^ 8:30am up 6 days, 23:41, 3 users, load average: 2.09, 2.15, 2.10
------------------------------
From: [EMAIL PROTECTED] (Michael Perry)
Subject: Re: cdrecord only works with SCSI drives?
Date: Tue, 08 May 2001 12:42:46 -0000
On Tue, 8 May 2001 00:40:27 -0400, Ryan Joseph <[EMAIL PROTECTED]> wrote:
> Hello!
>
> I recently bought an HP 9100 series cd-rw which connects via IDE. Now, when
> I
> downloaded cdrecord and a bunch of different front-ends, all of them (the
> front
> ends) reported that they couldn't open '/dev/pg0' and then said something
> about
> SCSI drives!
>
> I thought that cdrecord was supposed to work with almost all types of
> ATAPI/IDE
> drives as well - or did I just misread somewhere?
>
> Thanks for the help!
>
> R.Joseph
>
I have one of these drives as well. It works with cdrecord but you have to
do a few things first. See the cdrw howto for using ide cdrw w/Linux. Its
not particularly tricky but without undertaking the steps cdrecord will not
work correctly. I also use a variety of gui front ends like Xcdroast,
gcombust, and gtoaster.
Suffice to say, If I had it to do over again, I woulda, shoulda bought a
scsi one.
--
Michael Perry
[EMAIL PROTECTED]
==================
------------------------------
From: [EMAIL PROTECTED] (Andrew Purugganan)
Subject: Did YOUR distro install Mesa or OpenGL?
Date: 8 May 2001 12:49:02 GMT
Would like to hear from your experiences if your distro detected your
graphics accelerator card and maybe automatically installed these goodies:
Mesa or
Open GL
AFAIK, or as far as I can tell, it seems to be an optional preference,
and the most you would get is the latest XFree, and as far as gaming or
graphics goodness goes, you're on your own: you'd have to get them
yourself either from the Mesa website or the gaming vendor e.g. Loki
It would be nice if it were as simple as getting GLSetup for Windows, a
one-stop shop nomatter which card you own, but hey this is Linux. "Good
Luck boy! And tell me how it goes!"
--
jazz
Registered linux user no. 164098 +--+--+--+ Litestep user no. 386
Doesn't it bother you, that we have to search for intelligent life
--- OUT THERE??
------------------------------
From: [EMAIL PROTECTED] (Volker Apelt)
Crossposted-To: gnu.utils.help
Subject: gnu find not pruning
Date: 8 May 2001 14:57:48 +0200
I don't understand why /proc and /data are listed
by the find expression below. They are not
suid/sgid and they are explicitly pruned,
so they should be excluded.
The expression is part of a suid/sgid searching and
reporting script. It all files and directories on nfs
filesystems or in /proc and lists all files, if they
have a suid/sgid bit set. The resulting list is compared
to the previous list and a report is mailed to root
if the list has changed.
In addition to the suid/sgid files (ERROR!) the mountpoints them self
are listed, although they are directories!
This is problem for /proc because it changes its inode count
depending on the number of processes! this causes false
alerts if the previous dump is compared to the current one.
eg:
Because /data is a mount point to an NFS filesystem
${CURR_OUT} will contain a line describing /data,
the same happens to /proc.
Can you explain this error ?
================================== part of script
#!/bin/sh
CURR_OUT=`mktemp /tmp/find.suids.XXXXXXXX`
find / \
-path "/proc" -prune -o \
-fstype nfs -prune -o \
-type f \( -perm -04000 -o -perm -02000 \) | \
xargs --no-run-if-empty ls -ldis --full-time \
> ${CURR_OUT}
# [snip rest]
# ... if not empty send mail to root ...
# ... compare to last dump and send diff with an alert message
#
rm -f ${CURR_OUT}
--
Volker Apelt Group of Prof. Dr. Ch. Griesinger
Johann Wolfgang Goethe Universitaet
Frankfurt am Main (Germany)
[EMAIL PROTECTED] (use va@ instead of ...@ )
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.misc.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
