Linux-Misc Digest #948, Volume #27 Sat, 26 May 01 17:13:04 EDT
Contents:
Re: Switch from Gnome to KDE ("bowman")
Re: RH7.1 in endless loop: & mouse not detected ("hassard")
Re: permission confusion (faeychyld)
FAT32 mounting problem....please help (Mark)
Re: LILO limit still there? ("hassard")
problems compiling glibc 2.2.3 ("Thomas Hoell")
Re: fetchmail + deliver to several local user (Claus Atzenbeck)
Re: Standard input and output problem ? (Vilmos Soti)
Re: strange partition problem ("Oliver")
iBCS 2.1 and Kernel 2.4.x (Bob Schreibmaier)
Re: HELP formated drive (Robert_L)
Re: linux freezin (Jacob Kristensen)
Dissapearing GNOME taskbar ("Stefan Viljoen")
Installing Linux on a Laptop (Sumit Dhar)
Re: strange partition problem (Svend Olaf Mikkelsen)
Re: Weird problem - Eterm and rxvt use one minute to start up, xterm does not (Wim
Koorenneef)
Trouble booting: time for BIOS update? (Andrew Purugganan)
error conditions (Allan Adler)
Re: HELP formated drive (Nils Holland)
Re: permission confusion ("Garry Knight")
----------------------------------------------------------------------------
From: "bowman" <[EMAIL PROTECTED]>
Subject: Re: Switch from Gnome to KDE
Date: Sat, 26 May 2001 08:24:39 -0600
"Buck Turgidson" <[EMAIL PROTECTED]> wrote in message
news:R0MP6.70137$[EMAIL PROTECTED]...
> How would I go about switching from Gnome to KDE on an RH 6.1 machine? It
was
> installed with the former, but I'd like to try the latter.
If KDE is on the machine, put an .xinitrc file in your home directory with
a line
startkde
in it before any gnome-session or other gnomish stuff. when you do a
'startx' from the console it should find and start KDE.
If KDE was not installed, you can try installing KDE2.0. I don't know if the
situation has gotten any better since it was first released, but be prepared
for a session of rpm hell.
------------------------------
From: "hassard" <[EMAIL PROTECTED]>
Subject: Re: RH7.1 in endless loop: & mouse not detected
Date: Sat, 26 May 2001 10:26:21 -0400
hi-
The "Device or Resource busy" message suggests an interrupt conflict.
To check, boot as you've been able to with "Linux single" then look at the
file
/proc/interrupt. Look for what's MISSING:
if a card in your PC is not properly recognized by RH7.1
(in my case it was an ISA netword card, a Netgear EA201)
then it won't be allocated an interrupt and won't appear in /proc/interrupt
as it should.
What happens in this case, is that the system mistakenly allocates the
interrupt
(in my case IRQ 12) that this card should have, to another device such as
the mouse.
When this happens, when gpm tries to start up, it attempts to "talk"
simultaneously
to the mouse and to the PC card: this IRQ conflict produces the "Device or
Resource busy" message.
Assuming all this theory is correct, here is a procedure to try to resolve
the conflict.
(1) under Windows: Start -> Settings ->Control Panel
Click on "System". Select "View Devices by type".
For each device, select "properties", then look at "Resources".
Make a list of the cards that use IRQs.
(2) boot into RH7.1 using "Linux single". Look at the file
"/proc/interrupt"
for cards on the list from (1), but that are not mentionned as having
been allocated IRQs in the file /proc/interrupt.
(3) Remove the suspect cards from your system and see if the mouse
will then work under RH7.1: try
/usr/sbin/gpm
This should continue to give the "Device or Resouce busy" message
while the
IRQ conflict continues, but start up normally when the conflict is
resolved.
Hope this helps. In my case, the "Device or Resource busy" message
disappeared
(and the mouse worked normally) after removing the Netgear EA201 (ISA) card.
There seems to be a problem of some type with RH7.1 "Plug and Play", at
least
for this particular ISA card.
Micro Hue wrote in message <[EMAIL PROTECTED]>...
>G'day all,
>
>I have been trying to upgrade from Redhat 6.2 to 7.1 but the
>the installed system chokes on my Logitech Wheelmouse when I boot.
>
>I am relatively new to Linux (occasional use on my dual boot PII 333
>64MB box) and would like some pointers please on my tale of woe...
>
>My mouse has been working fine with RH 6.2 and still does since Ive
>had to revert to 6.2 after many abortive attempts to get RH7.1 going.
>
>On install (or upgrade) of RH7.1, the mouse is recognized as Generic 3
>button mouse (PS/2) - same as RH 6.2 does. Mouse works fine during
>install.
>After install exits, I reboot. Error message, something like:
>
>"gpm: oops() invoked from gpm.c[978]
>Device or Resource busy"
>
>Then the computer goes into an endless loop between a text login
>prompt which disappears when (presumably) X tries to start and
>reappears briefly a few seconds later only to disappear before I have
>a chance to log in. Repeat ad nauseum. No keypresses seem to stop it.
>I've tried numerous keys: Alt-Ctl-F2, Alt-Ctl-Back, A-C-Tab etc.
>Nothing works but A-C-Del which only reboots into the same problem. (I
>use Loadlin to boot linux)
>However the shut down produces a message:
>"Shutting down console mouse services [FAILED]"
>
>How do I stop the startup looping?
>How do I fix the mouse?
>Why does RH7.1 choke on a simple mouse that RH6.2 happily swallows?
>
>I've tried reinstalling using every combination of Generic PS/2 mouse,
>even the Logitech PS/2 ones. All give same result.
>
>I even tried the floppy boot disk using 'Linux single'.
>At least I got a prompt.
>Ran 'mouseconfig'
>"There was an error reading file /etc/sysconfig/mouse
>Would you like to create a new configuration?
>(Would I ever!) OK
>Mouseconfig can update Xfree86 configuration file?
>YES
>
>Reboot. Same problem.
>
>It must be something simple I've overlooked...
>
>Tony
>
------------------------------
Date: Sun, 27 May 2001 00:51:58 +1000
From: faeychyld <[EMAIL PROTECTED]>
Subject: Re: permission confusion
The disk is formated ext2 .
But you are right - back to the man mount pages.
Bob Hauck wrote:
>
> On Sat, 26 May 2001 19:58:10 +1000, faeychyld <[EMAIL PROTECTED]> wrote:
>
> > Permissions always throw me a curve. I thought
> > if I mounted a drive ( ls120 n this case )
> > with the user option eg; fstab entry
> >
> > /dev/hdd /mnt/ls120 auto noauto,user,rw 0 0
> >
> > then I would own the drive and have rw access.
>
> Not necessarily. That must means you can mount it without being root.
>
> > Not so, the drive is still owned by root with 777 permissions.
>
> What filesystem is on the ls120? Sounds like it might be FAT or VFAT.
> Those don't grok permissions or ownership, so the kernel fakes them when
> you mount the drive.
>
> Look at the mount man page for info on the "uid", "gid", and "umask"
> options.
>
> Also, the mtools allow you to manipulate FAT disks without mounting
> them. There are analogs of the usual DOS commands, prefixed by "m", as
> in "mdir b:", "mcopy a:file file". Very handy. See "man mtools".
>
> --
> -| Bob Hauck
> -| To Whom You Are Speaking
> -| http://www.haucks.org/
--
-
-
-
Regards F
------------------------------
From: Mark <[EMAIL PROTECTED]>
Subject: FAT32 mounting problem....please help
Date: Sat, 26 May 2001 10:35:12 -0500
I am trying to mount a logical FAT32 partition in RedHat 7.1. It gives me
a somewhat cryptic error message when I try to do so.
mount -t vfat /dev/hda7 /mnt/data
mount: wrong fs type, bad option, bad superblock on /dev/hda7,
or too many mounted file systems
Here is my drive setup given by fdisk:
Disk /dev/hda: 255 heads, 63 sectors, 5005 cylinders
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System
/dev/hda1 * 1 765 6144831 b Win95 FAT32
/dev/hda2 766 770 40162+ 83 Linux
/dev/hda3 771 5005 34017637+ 5 Extended
/dev/hda5 771 1535 6144831 83 Linux
/dev/hda6 1536 1584 393561 82 Linux swap
/dev/hda7 1585 5005 27479151 b Win95 FAT32
I am able to mount the primary FAT32 partition (hda1) without a problem.
Windows sees the logical FAT32 and can write to it just fine. The logical
FAT32 was created by the Linux fdisk. I notice that some others are having
this problem but can't find a solution. Any help would be greatly
appreciated.
------------------------------
From: "hassard" <[EMAIL PROTECTED]>
Subject: Re: LILO limit still there?
Date: Sat, 26 May 2001 11:56:52 -0400
There are various limits, not just the 1024 cylinder limitation which
modern BIOSes overcome.
The subject makes interesting reading, but here's experience
from last summer, installing dual-boot RH6.2 (15G) and Windows (15G)
in a 30G disk.
The major limitation was that "/boot" had to be in the first 8G.
We opted to make a single 15G "C:" drive for Windows, which had
come preinstalled taking up the entire 30G disk.
The only way to do make a contiguous 15G partition for Windows,
and yet to keep "/boot" in the first 8G, was to move the Windows partition
up so as to make space for "/boot" at the beginning of the drive.
We used Partition Magic, which moved the Windows installation
and kept it intact.
Vladimir Florinski wrote in message
<[EMAIL PROTECTED]>...
>I was installing RH7.1 for someone who has some version of Windows (60GB
>disk) and the installer would not go past the partitioning step, saying /
>is above 1024 cylinders, etc., you know the routine. My understanding is
that in
>LILO this limit was lifted quite a while ago. Is Red Hat so far behind the
>times? Or is there another reason?
>
>
>
>--
>
>
>Vladimir
------------------------------
From: "Thomas Hoell" <[EMAIL PROTECTED]>
Subject: problems compiling glibc 2.2.3
Date: Sat, 26 May 2001 18:12:22 +0200
Reply-To: [EMAIL PROTECTED]
Hi,
the last few days I was trying to properly compile glibc 2.2.3.
Unfortunally, something went wrong. I can't tell exactly what it is.
'make' works find. 'make check' ALWAYS fails in glibc-2.2.3/grp.
testgrp.out says: "Cannot find user entry for UID 0" However, when I
manually run the testgrp-prog, I get:
>root@sirius:/tmp/glibc-2.2.3/grp# ./testgrp
>My login name is root.
>My uid is 0.
>My home directory is /root.
>My default shell is /bin/bash.
>My default group is root (0).
>The members of this group are:
> root
[This works with other users as well]
I once installed the lib by accident, the result was that I couldn't log
on anymore, I couldn't even reboot the machine (CTL+ALT+BS => "I don't
know who you are, go away!"). It may be a problem with the crypt library,
but I'm not sure.
my system configuration:
Slackware 7.0
kernel 2.4.4
gcc 2.95.3
binutils 2.11
The strange thing is that it used to work! I compiled glibc 2.2.0, 2.2.1
and 2.2.2 without any problems, but even these versions won't compile
properly anymore. I also can't remember doing any changes to the system.
Any ideas what I may have done wrong?
Thanks in advance,
Thomas
--
Windows 9x: n. 32 bit extensions and a graphical shell for a 16 bit patch
to an 8 bit operating system originally coded for a 4 bit microprocessor,
written by a 2 bit company that can't stand 1 bit of competition.
GnuPG-Key 0x0FFE104B
------------------------------
From: Claus Atzenbeck <[EMAIL PROTECTED]>
Subject: Re: fetchmail + deliver to several local user
Date: Sat, 26 May 2001 18:17:22 +0200
Michael Heiming wrote on Samstag, 26. Mai 2001 14:56:
> "apropos procmail" for all man pages, sendmail uses procmail as default
> LDA on Linux, AFAIK postfix does the same.
This should work. Thanks a lot for your help!
Claus
------------------------------
Subject: Re: Standard input and output problem ?
From: Vilmos Soti <[EMAIL PROTECTED]>
Date: Sat, 26 May 2001 16:46:26 GMT
"Eric Chow" <[EMAIL PROTECTED]> writes:
> Would you please to show me a simple example ?
>
>> How to send the INPUT parameter to a program when it ask for input ?
>> for example, just like "ftp", when I type "ftp localhost", it will ask for
>> username and password.
>> How to send the username and password to that ?
man bash, and search for "here document".
Vilmos
------------------------------
From: "Oliver" <[EMAIL PROTECTED]>
Subject: Re: strange partition problem
Date: Sat, 26 May 2001 18:54:41 +0200
> If data currently is lost, you should not write to the disk at all
> until the exact nature of the problem is known.
>
> The cause of the problem is that the type of the extended partition is
> wrong. It should be 0F.
thx for the tips, but didn't work
------------------------------
From: [EMAIL PROTECTED] (Bob Schreibmaier)
Subject: iBCS 2.1 and Kernel 2.4.x
Date: Sat, 26 May 2001 17:29:25 GMT
Has anybody gotten ibcs-2.1-981105 to compile under any of
the 2.4.x kernels? The kernel documentation hints that they
might still have some work to do, but if anyone has any
secrets to making it work now, I would greatly appreciate it.
Thanks in advance.
Bob
--
+------------------- \-\-\-\ ----------------------------+
| Bob Schreibmaier K3PH | E-mail: [EMAIL PROTECTED] |
| Kresgeville, PA 18333 | http://www.qsl.net/k3ph |
+--------------------------------------------------------+
------------------------------
From: Robert_L <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: HELP formated drive
Date: Sat, 26 May 2001 13:47:13 -0400
Reply-To: [EMAIL PROTECTED]
John Hasler wrote:
> Robert writes:
>> Anyway, someone here will probably know a better way using linux.
>
> If _all_ you did was delete a partition or alter the partition table in
> some way you can recover your data by simply recreating the partition with
> Linux fdisk. Messing with the partition table does not touch the data in
> the partition.
Well, all I know is that the disk was fdisked, and the MRecover program
found the original partition table and restored it , along with ALL of my
data.
I'll leave the details to the experts.
Robert
--
remove spamfree when replying directly
------------------------------
From: Jacob Kristensen <[EMAIL PROTECTED]>
Subject: Re: linux freezin
Date: Sat, 26 May 2001 19:58:45 +0200
alik blochin wrote:
>
> Hi thanks for your help
> i will try some of the tips you gave me..
> by the way on my system
> eth0 and soundcard are indeed share one IRQ
> i know it from windows...
>
> question:
> where do i set the option PNP enabled to yes ?
In ur BIOS settings. When u boot ur machine hit DEL, F1 or
something(read the manual that came with ur machine) to enter BIOS
settings.
> and second what is the sysRq magic key ? :)
> can you explain me what is all about :
>
> "Before hitting the Reset - try the SysRq magic... I *think* the
> sequence would be Ctrl-Alt-SysRq-t (I know it is t for terminate, the
> CAS is a different matter) followed by CAS-s to sync the disks. If
> that works, all is well. If not, CAS-k for kill, then CAS-s to sync,
> and finally CAS-r to reboot... At least you'll get a clean disk..."
A kernel can be compiled with the "magic sysRq key" turned on or off. As
far as I know it is turned OFF by default(hence - the kernel that came
with ur dist probably dos'ent have this feature), but I'm not sure. The
idea is that, if a system is in a freeze the SysRq will still work. One
can(if it works) make the system sync(write buffers to disk) and other
neat technical stuff. Read the Kernel-HOWTO or the docs in
/usr/src/linux/Documentation. I never used it...
>how i change the interrupsts between sound and ethernet ?
Don't know....Ask the NG hopefully somebody knows...
BTW: Ur emailaddr does'ent work...
Jacob
--
There he goes... One of Gods own prototypes.
A high-powered mutant of some kind,
never even considered for mass production.
Too weird to live, and too rare to die.
------------------------------
From: "Stefan Viljoen" <[EMAIL PROTECTED]>
Subject: Dissapearing GNOME taskbar
Date: Sat, 26 May 2001 20:22:31 +0200
Hi
I booted X recently and the GNOME taskbar was gone - no changes to
configuration or to any element of the system. I haven't heard of this or
seen it somehwere else - any ideas?
Thanks!
Stefan
------------------------------
From: [EMAIL PROTECTED] (Sumit Dhar)
Subject: Installing Linux on a Laptop
Date: 26 May 2001 11:38:58 -0700
Hello Everyone...
I need to install Linux on a 486 Laptop which has just 500 Mb hard disk and
no CD ROM drive. I initially thought that I will do it via NFS. So
made a bootable floppy using the standard bootnet.img that comes with
RHL 6.2
Now, this was the first time I was installing on a laptop. It asked me
for a driver disk for the PCMCIA card. Pulling out the PCMCIA card, I
could see it was Longshine. (Nothing more than that could be
discerned)
I downloaded the drivers.img file from the Redhat Site and continued
further. It proceeded further till it reached the stage where it asked
me for the method of installation. Here I chose NFS and when I did
that, it asked me to chose a driver. Unfortunately none of the drivers
seemed to work.
But this machine is already running RHL 4.2 and hence I know,
everything it has is supported.
My questions are:
1. What can I do to get the driver disks.. (Longshine site does not
provide them for Linux)
2. Given that I have a running system, can I probe it for irqs, io
ports and use them in some way?
3. Can I some how create a driver disk from the running system that I
have?
If yes, how... ? If I am missing some information, please ask me.. Any
answers would be greatly appreciated.
Regards
Dhar
PS: I have already gone thru the Linux Laptop Howto, the 4 Mb Laptop
Howto, the PCMCIA Howto... any Howto that I missed???
~
------------------------------
From: [EMAIL PROTECTED] (Svend Olaf Mikkelsen)
Subject: Re: strange partition problem
Date: Sat, 26 May 2001 20:00:13 GMT
"Oliver" <[EMAIL PROTECTED]> wrote:
>> If data currently is lost, you should not write to the disk at all
>> until the exact nature of the problem is known.
>>
>> The cause of the problem is that the type of the extended partition is
>> wrong. It should be 0F.
>
>thx for the tips, but didn't work
The wrong partition ID was the cause of the problem. I did not say
correcting it would bring the data back. The data is written to an
unknown area of the disk inside one of your other partitions. Writing
anything to the disk, until the exact nature is known, will reduce the
chance of recovery.
--
Svend Olaf
------------------------------
From: Wim Koorenneef <[EMAIL PROTECTED]>
Subject: Re: Weird problem - Eterm and rxvt use one minute to start up, xterm does not
Date: Sat, 26 May 2001 20:36:10 GMT
Rune Jacobsen wrote:
> I'm having a weird problem. When I try to start an Eterm or an rxvt (without
> any special options) on my Linux box, weird things happen. It takes about a
> minute to start them up!
I'm having the same problem. It doesn't take a whole minute though, more
like 15 seconds. Running Gnome (redhat 7.1) on a PIII 733 and 256MB.
If you hear of a possible solution outside of this newsgroup, please let
me know.
--
Wim Koorenneef <[EMAIL PROTECTED]> Boxtel, the Netherlands
------------------------------
From: [EMAIL PROTECTED] (Andrew Purugganan)
Subject: Trouble booting: time for BIOS update?
Date: 26 May 2001 14:10:02 GMT
I have a 586ipvg with Award BIOS revision A 1997
with a Voodoo1 graphics card, a P233, & 64M RAM
During BIOS boot I make it detect the hard drives but instead of LBA I
set it to normal so that it will correctly show me that I have a 2.1G
hdd. (If I set it to LBA, it tends to show me a 540M hdd instead, even
tho it detected correctly the 2.1G)
I get one of the following
LI
LIL?
LLLLLLLLLLLLLLLLLLLLLLLLLLLL
and also
LILO: linux
Uncompressing Linux..........
bad gzip magic number
OR
LILO: linux
vesa modes not detected....please e-mail [EMAIL PROTECTED]
OR
LILO: linux
You entered a bad mode number. Please enter the etc. etc.
Any ideas?
--
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??
------------------------------
Subject: error conditions
From: Allan Adler <[EMAIL PROTECTED]>
Date: 26 May 2001 16:50:48 -0400
I'm running RedHat 7.2 on a Dell Latitude CSx. When I mount /a or /c
as msdos file systems (which they are) I have trouble reading them.
Specifically, executing "ls /a" or "ls /c" gives me segmentation
faults. If instead I execute "ls /a/wqrszt" (there is no file /a/wqrszt)
instead the system logs me out. I gather that the error condition provoked
by "ls /a/wqrszt" is much more serious than the error condition provoked
by "ls /a".
Can anyone guess what the difference might be? I don't know much about
error conditions.
Allan Adler
[EMAIL PROTECTED]
****************************************************************************
* *
* Disclaimer: I am a guest and *not* a member of the MIT Artificial *
* Intelligence Lab. My actions and comments do not reflect *
* in any way on MIT. Morever, I am nowhere near the Boston *
* metropolitan area. *
* *
****************************************************************************
------------------------------
From: Nils Holland <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: HELP formated drive
Date: Sat, 26 May 2001 22:54:02 +0200
John Hasler wrote:
> Robert writes:
>> Anyway, someone here will probably know a better way using linux.
>
> If _all_ you did was delete a partition or alter the partition table in
> some way you can recover your data by simply recreating the partition with
> Linux fdisk. Messing with the partition table does not touch the data in
> the partition.
Yes, but if mke2fs already started to mess with the partition in question,
then a good deal of data on there may already have been deleted and
recreating it is probably not so easy. In fact, I don't know if it's at all
possible to recreate the data stored on a partition that you have run
mke2fs on...
Greetings
Nils
--
==========================================================
Nils Holland - [EMAIL PROTECTED]
NightCastle Productions - Linux in Tiddische, Germany
http://www.nightcastleproductions.org
"They asked me where this earthquake would begin,
I offered to let them feel my pulse."
==========================================================
------------------------------
From: "Garry Knight" <[EMAIL PROTECTED]>
Subject: Re: permission confusion
Date: Sat, 26 May 2001 15:39:45 +0100
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, "faeychyld"
<[EMAIL PROTECTED]> wrote:
> Permissions always throw me a curve. I thought if I mounted a drive (
> ls120 n this case ) with the user option eg; fstab entry
>
> /dev/hdd /mnt/ls120 auto noauto,user,rw 0 0
>
> then I would own the drive and have rw access. Not so, the drive is
> still owned by root with 777 permissions.
>
> Where have I gone wrong?
I've never used an LS120, but I would have thought you'd need a partition
number in the device name, e.g. /dev/hdd1, rather than just 'hdd' on its
own.
In any case, you didn't say which filesystem is on the LS120, but from
the behaviour you describe I assume it's DOS, FAT or VFAT, since you can
only specify one user ID and group ID for these filesystems.
The way you specify the UID and GID is by using the uid and gid options
in /etc/fstab. For example, here's an entry from my /etc/fstab which
gives ownership of my /mnt/windows partition to user 'garry', group
'garry' with default perms of 770.
/dev/hda1 /mnt/windows vfat user,uid=501,gid=501,umask=007 0 0
You can find out your own UID and GID when you're logged in by using the
'id' command in a console. Root's UID and GID are both 0.
And to set the default perms on the partition you specify a umask. In the
example above, I've specified a umask of 007, which gives default perms
of 770, i.e. read, write and execute perms for user and group only. It
should be obvious why I've chosen these perms.
In order to work out which umask to specify in /etc/fstab you need to be
able to work in octal, so it might be best if you posted back here with a
permission mask such as 'rwxrw-r--' and I (or someone else) will give you
the corresponding umask.
The line you'll end up with in /etc/fstab will end up looking something
like this:
/dev/hdd /mnt/ls120 auto noauto,user,rw,uid=501,gid=501,umask=007 0 0
--
Garry Knight
[EMAIL PROTECTED]
------------------------------
** 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
******************************
