Linux-Setup Digest #983, Volume #20 Wed, 4 Apr 01 11:13:14 EDT
Contents:
Re: X Servers and Voodoo 3 (Martin Gregorie)
Re: Unkown users made by RedHat (Jimbo)
Re: hdb boot ("Andrey Shcherbina")
Re: Kernel compilation failed ? Redhat 7.0 (Steve Martin)
Re: upgrading rpm ("Tristan Day")
rquota question (Mohd-Hanafiah Abdullah)
Re: Memory pb : 384M real, 64M effective (Stephen Rank)
Re: Unkown users made by RedHat (H.Bruijn)
PTBL ("Ron Nicholls")
Re: Logrotate...or not? (ImaLuzer)
Re: ReiserFS for / (Rolf Magnus)
Re: update to glibc 2.2.2 ("Rex Dieter")
Re: Kernel compilation failed ? Redhat 7.0 ("Rex Dieter")
Re: hdb boot ("Eric")
eth0 initialization fail ("Marc Ulrich")
Re: Logrotate...or not? (H.Bruijn)
Re: problems w/ Mandrake install (Dave Bossie)
Re: problems w/ Mandrake install (Dave Bossie)
Re: problems w/ Mandrake install (Dave Bossie)
more info: eth0 initialization fail ("Marc Ulrich")
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Martin Gregorie)
Subject: Re: X Servers and Voodoo 3
Date: Wed, 04 Apr 2001 11:29:51 GMT
On Wed, 4 Apr 2001 01:12:00 +0100, Mat Grove <[EMAIL PROTECTED]>
wrote:
>Is there a better driver for my video card? Or is this what seems like lag
>to me between kde and the x server due to the nature of the setup - x runs
>as a sepearte program with the window manager just talking to it.
>
You could take a look at XIG (www.xig.com) who supply
accelerator-enabled replacements for xfree86. I've only looked at
their website, but it would seem to offer what you're wanting.
--
gregorie | Martin Gregorie
@logica | Logica Ltd
com | +44 020 76379111
------------------------------
From: Jimbo <[EMAIL PROTECTED]>
Subject: Re: Unkown users made by RedHat
Date: Wed, 04 Apr 2001 13:36:36 +0200
Reply-To: [EMAIL PROTECTED]
On 3 Apr 2001 20:32:41 GMT, [EMAIL PROTECTED] (H.Bruijn)
wrote:
>>daemon **Never logged in**
>>adm **Never logged in**
>>sync **Never logged in**
>>mail **Never logged in**
>>named **Never logged in**
>> [ snip: more ]
>>
>>Which of these am I allowed to just issue "userdel xx" and "groupdel
>>xx"?
>
>These are "system users" added to the system by certain pieces of
>software. They have no active password, ie the password field in
>etc/shadow only contains a * . You can't use them to log in.
>The reason they exist that for some software it is advisabe
>that it doesn't run as root, but as an user with limited permissions.
>
>The second are some groups which exist on a system. They are used,
>obviously, to give some groups of users some extra permissions and such.
Thx for the reply.
I've put * in the /etc/shadow for the users that don't need to log-in
(ie..every user that's not mapped to a 'human' person has a *).
I've figured it out for httpd, dnscache and all the others. But user
"sync", "adm", "shutdown" doesn't seem to be connected to any service.
What runs under "shutdown" & "sync"? Obvious shutdown has to deal with
shutting down the system but why should I use a specific user-account?
I can issue "halt" from the command-line as root!
Greetings,
Jim
------------------------------
From: "Andrey Shcherbina" <[EMAIL PROTECTED]>
Subject: Re: hdb boot
Date: Wed, 4 Apr 2001 07:58:27 -0400
Hi,
"Eric" <[EMAIL PROTECTED]> wrote in message
news:9aehpc$44r$[EMAIL PROTECTED]...
> > When I installed Debian, it asked me if I wish to boot from the HDD.
Sure
> I
> > want to. There were two choices, MBR and hdb3. MBR didn't work as I
> removed
>
> Then it would have been installed in the MBR of hdb
>
That's what I thought, but it didn't happen. It just said it can't.
> > primary master to not to screw it up. So, I choose hdb3. It said
something
>
> That wasn't required.
>
> > that I'm not booting from primary master drive and maybe my BIOS doesn't
> > support it. Well, it does! But when I try to boot to Linux, it says
"LILO"
> > and then keeps writing "01".
>
> And lilo is at hdb3?
> could you post `cat /etc/lilo.conf`
> and `fdisk -l /dev/hd[ab]`
>
It's pretty hard to run any commands if Linux can't load :) And I don't
think a floppy would help as LILO actually starts but then can't load
anything.
>
------------------------------
From: Steve Martin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Kernel compilation failed ? Redhat 7.0
Date: Wed, 04 Apr 2001 07:07:30 -0400
Kevin Hayes wrote:
> > All looks good, but at
> >
> > make bzImage
> > at some point, it says:
> >
>
> There were some problems with the gcc version shipped with RH70, its possible
> to recompile a kernel by renaming gcc and making a symlink kgcc -> gcc...
>
> There were a lot of posts on this subject in the recent past.
An alternative (and one that may be better) is to modify the top-level
Makefile in the kernel source. Find the lines that mention "gcc"
(there should be two of them) and replace "gcc" with "kgcc". The
kernel should compile without problem then.
This approach allows you to use the newer gcc to compile your own
programs, if you believe (as some posts have indicated here) that
the newer gcc is technically "better". (According to what I have
read, the kernel won't compile under the 2.96 gcc from RH7 due to
code in the kernel that strictly speaking isn't "standard" but that
the earlier versions of the compiler gracefully accepted anyway.)
------------------------------
From: "Tristan Day" <[EMAIL PROTECTED]>
Subject: Re: upgrading rpm
Date: Wed, 4 Apr 2001 13:01:27 +0100
"b" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> how does one upgrade from an older verserion of rpm to a newer one.
>
> rpm -Uvh rpm* doesnt work
>
> likewise trying to upgrade netscape has the same effect you cannot
> upgrade with out backing up, remove old package and install new version
> recover backup.
>
> as netscape is looking at different packages and a rpm -Uvh --force
> doesn't work
>
> barry
for netscape (for example).....
# rpm -Fvh netscape*.rpm
F = Freshen (will not install something not already installed)
v = verbose
h = hash
if u want to ADD these packages, change the F for U.
hth.
------------------------------
From: [EMAIL PROTECTED] (Mohd-Hanafiah Abdullah)
Crossposted-To: comp.os.linux.misc
Subject: rquota question
Date: 4 Apr 2001 20:26:29 +0800
Can anyone please point to me where I could get complete documentation
on setting up "remote quota" (rquota). I have diskless Linux workstations
sharing diskspace located on a server by means of NFS, and would like to set
user quota for each of the diskless workstations.
Thanks.
Napi
------------------------------
From: Stephen Rank <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: Memory pb : 384M real, 64M effective
Date: 04 Apr 2001 13:23:28 +0100
"Peter T. Breuer" <[EMAIL PROTECTED]> writes:
> In comp.os.linux.hardware Pantalacci Christophe
<[EMAIL PROTECTED]> wrote:
> > I've got this kind of problem : Linux only sees 64M of RAM. When I
> > modify lilo.conf by adding
>
> There's no problem with that.
>
> > append="mem=384M" and executing lilo, my whole system crashes at
> > reboot with segmentation faults everywhere, and i must reinstall.
[ replying to a reply, as the parent post has gone AWOL ]
You don't need to reinstall: just pass the `mem=xxxM' to the kernel
via the Lilo prompt when you boot (e.g., ``Linux mem=300M''). Once
your system boots, you can adjust your lilo.conf appropriately and
then run lilo.
HTH,
Stephen
------------------------------
From: [EMAIL PROTECTED] (H.Bruijn)
Subject: Re: Unkown users made by RedHat
Date: 4 Apr 2001 12:39:41 GMT
Reply-To: [EMAIL PROTECTED]
On Wed, 04 Apr 2001 13:36:36 +0200, Jimbo allegedly wrote:
>On 3 Apr 2001 20:32:41 GMT, [EMAIL PROTECTED] (H.Bruijn)
>wrote:
>
>>> [ snip: more system users that never log in]
>>>
>>>Which of these am I allowed to just issue "userdel xx" and "groupdel
>>>xx"?
>>
>>These are "system users" added to the system by certain pieces of
>>software. They have no active password, ie the password field in
>>etc/shadow only contains a * . You can't use them to log in.
>>The reason they exist that for some software it is advisabe
>>that it doesn't run as root, but as an user with limited permissions.
>>
>>The second are some groups which exist on a system. They are used,
>>obviously, to give some groups of users some extra permissions and such.
>
>Thx for the reply.
>
>I've put * in the /etc/shadow for the users that don't need to log-in
>(ie..every user that's not mapped to a 'human' person has a *).
>
>I've figured it out for httpd, dnscache and all the others. But user
>"sync", "adm", "shutdown" doesn't seem to be connected to any service.
sys and adm are connected to the entries in /dev/ and the entries in
/var/log on my debian system.
>What runs under "shutdown" & "sync"? Obvious shutdown has to deal with
>shutting down the system but why should I use a specific user-account?
>I can issue "halt" from the command-line as root!
sync has as shell /bin/sync so I would imagine that it is used to
synchronize memory and the filysystem from a cron job. As it will be
done by sync it doesn't have the protentially destructive powers it
would have when root would do it. "man sync"
shutdown will have a shell like /sbin/halt. Now if you want other people
to be able to shut the system down, or reboot it, you can set a password
for the "halt" account, and give them that password. That way they can
do a proper shutdown, but you don't need to hand over the root password,
and give them complete control over the system. This is quite usefull
with a dual boot system, where the others don't feel confortable using
linux. Just tell them to log on as halt, with a trivial password and
they can shut the system down and reboot to whatever other OS they
prefer. As you don't want them to log in over telnet to the halt account
and shut the system down when you're working (it's a good practical joke
though) add a line to /etc/security/access.conf
-:shutdown:ALL EXCEPT LOCAL
--
If a trainstation is the place where trains stop, what is a workstation?
========================================================================
Herman Bruijn website: http://hermanbruijn.com
The Netherlands
------------------------------
From: "Ron Nicholls" <[EMAIL PROTECTED]>
Subject: PTBL
Date: Wed, 4 Apr 2001 22:50:41 +1000
After repartitioning a secondary HD
which somehow was trashed by a windows reinstall
it is now reported during boot as [ PTBL ] [531/255/63]
obviously chs geometry.
Does it mean I am in trouble - can I fix the drive
It does work so far.
RH 7.0
--
-
-
Regards
RonN
------------------------------
From: ImaLuzer <[EMAIL PROTECTED]>
Subject: Re: Logrotate...or not?
Date: Wed, 04 Apr 2001 09:17:08 -0400
Okay, thanks to the help you guys gave me, I was able to get logrotate to
execute. (feel kinda dumb).
My conf file for logrotate>apache tries to do an olddir (old logs are sent
to the directory specified in olddir). I get an error saying:
error: /etc/logrotate.d/apache:13 olddir /u/homes/httpd/html/admin/old_logs
and log file /var/log/httpd/arenet_access_log are on different devices
This isn't a big deal, but I was under the assumption that Linux doesn't
care about what physical device a directory/file is located on. This error
apparently means that it does. Or is there some other type of violation that
can cause this error (permission/security..etc)?
Thanks for your help - mp @ 48040
------------------------------
From: Rolf Magnus <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux,comp.os.linux.admin,comp.os.linux.questions
Subject: Re: ReiserFS for /
Date: Wed, 4 Apr 2001 15:24:06 +0200
Rod Smith wrote:
> As to the root filesystem proper, that need not be a problem. I believe
> some older versions of ReiserFS had issues with the kernel under some
> circumstances. IIRC, it wasn't safe to boot the kernel with LILO from a
> ReiserFS partition unless it was formatted with certain options. I'm not
> sure if that's still an issue. In case it is, you might want to create a
> separate ext2fs /boot partition and put the kernel and LILO on it. That
> should work around kernel-related problems. I'm not currently running
> any systems with the root partition on ReiserFS, but I've done so in the
> past with no problems.
If lilo makes problems with a boot partition in reiserfs, you can try grub.
It's another boot loader which has direct support for reiserfs.
------------------------------
From: "Rex Dieter" <[EMAIL PROTECTED]>
Subject: Re: update to glibc 2.2.2
Date: Wed, 4 Apr 2001 08:53:28 -0500
There is no official/released glibc 2.2.2 redhat update (that I'm aware of).
I belive you're trying to install the rpm from rawhide, which is usually not
recommended unless you're running at least RedHat 7 (and you know what
you're doing).
Otherwise, for RedHat 6, stick with glibc-2.1.3-22, which is the latest
update available.
In no circumstances should you --nodeps or --force on a glibc install...
you'll most likely get an unusable system.
--
Rex Dieter
Computer System Administrator
Mathematics and Statistics
University of Nebraska Lincoln
"chris" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hello all -
>
> I have attempted to update glibc to v.2.2.2 with rpms from the
> updates.redhat.com site. When I try i get errors (dependencies) [rpm
> -Uvh] and I don't want to screw up the system by doing a --nodeps.
> Any ideas.
>
> TIA,
>
> chris
------------------------------
From: "Rex Dieter" <[EMAIL PROTECTED]>
Subject: Re: Kernel compilation failed ? Redhat 7.0
Date: Wed, 4 Apr 2001 08:54:54 -0500
"WONG SAI-KEE" <[EMAIL PROTECTED]> wrote in message
news:9ae02k$16v$[EMAIL PROTECTED]...
> I need to config the Redhat 7.0 to have mars_nwe. So I followed the
FAQ: http://www.redhat.com/support/docs/gotchas/7.0/gotchas-7-7.html
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: hdb boot
Date: Wed, 4 Apr 2001 16:04:29 +0200
> > And lilo is at hdb3?
> > could you post `cat /etc/lilo.conf`
> > and `fdisk -l /dev/hd[ab]`
> >
>
> It's pretty hard to run any commands if Linux can't load :) And I don't
> think a floppy would help as LILO actually starts but then can't load
> anything.
It can't find the files required to boot.
If these would be on the floppy, you'd have no problem.
If you have a floppy, put it in, and boot from that.
If not, put the CD in again, and at the lilo prompt enter
"linux root=/dev/hdcXXX" (or "linux rescue root=/dev/hdcXXX")
The XXX should be replaced by the partition number of your root partition.
Eric
------------------------------
From: "Marc Ulrich" <[EMAIL PROTECTED]>
Subject: eth0 initialization fail
Date: Wed, 4 Apr 2001 09:51:20 -0400
I am using a 3com 3C905c and an intel EtherExpress Pro adapter. When the
system boots up, eth0 takes about 5 minutes before it fails and gives me the
red [FAILED] signal. I've checked and I know that the drivers that I am
using are the correct drivers. I also know that both interfaces are good
because I am in Win2000 now and sending this email using one of them. I've
also tried the other. Any ideas what causes this and what I can do to fix
it? I am using kernel 2.4.2, but I also noticed that if I boot up using the
standard RH7's 2.2.16 kernel, the same thing happens.
Thanks,
Marc
------------------------------
From: [EMAIL PROTECTED] (H.Bruijn)
Subject: Re: Logrotate...or not?
Date: 4 Apr 2001 14:29:22 GMT
Reply-To: [EMAIL PROTECTED]
On Wed, 04 Apr 2001 09:17:08 -0400, ImaLuzer allegedly wrote:
>Okay, thanks to the help you guys gave me, I was able to get logrotate to
>execute. (feel kinda dumb).
>
>My conf file for logrotate>apache tries to do an olddir (old logs are sent
>to the directory specified in olddir). I get an error saying:
>
>error: /etc/logrotate.d/apache:13 olddir /u/homes/httpd/html/admin/old_logs
>and log file /var/log/httpd/arenet_access_log are on different devices
>
>This isn't a big deal, but I was under the assumption that Linux doesn't
>care about what physical device a directory/file is located on. This error
>apparently means that it does. Or is there some other type of violation that
>can cause this error (permission/security..etc)?
Linux doesn't care, but logrotate does. It explicitely states in the
manual page "The "olddir directory" must be on the same physical device
as the log file being rotated." I would imagine that it has to
do with the fact that you would create a lot of additional pitfalls
when you move to a different device, fi a NFS mounted drive across the
network, or a tape device, which are all a lot slower then moving files
on the same physical device/partition. Take a large file and check for
yourself with
time mv /tmp/largefile /tmp/large_file
time mv /tmp/large_file /different/device/large_file
--
If a trainstation is the place where trains stop, what is a workstation?
========================================================================
Herman Bruijn website: http://hermanbruijn.com
The Netherlands
------------------------------
From: Dave Bossie <[EMAIL PROTECTED]>
Subject: Re: problems w/ Mandrake install
Date: Wed, 04 Apr 2001 14:30:07 -0000
Thomas McNamara wrote:
>
>
> I've got WindowsME installed on a 20.4 Gig Maxtor, and want
> to dual boot Mandrake 7.2. Upon installation, when I get to the point
where
> "The DrakX Partitioning wizard found the following solutions:"
> I click on "Use Existing Partition" and get this error:
>
> ---The FAT resizer is unable to handle your partition, the following
> error occured: Can't locate object method "new" via package
> "resize_fat::main" at /usr/bin/perl-install_interactive.pm
> line 108, <> line 5---
>
> Now, I'm a newbie, so I'm stumped. I haven't had any problems with
> lnx4win, but that method makes for a sluggish OS. TIA
>
> Note: I can bypass this error by going into Expert Mode, but the I get
> another
> error when Mandrake attempts to install/configure the bootloader.
>
> I want to run Linux--I'm actually really excited about it--but I still
need
> Windows for some things, and don't want to settle for lnx4win.
>
>
> --Tom McNamara
>
> PS:
>
> FWIW, here are the rest of my system specs:
>
> Duron 700 Mhz
> Abit KT7 (KT133 chipset)
> 128 megs generic SDRAM
> Visiontek GeForce MX
> Soundblaster Live! Value
> Pioneer 16x DVD
> Realtek PCI 10/100 NIC
> Creative Labs PCI Modem Blaster
>
>
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: Dave Bossie <[EMAIL PROTECTED]>
Subject: Re: problems w/ Mandrake install
Date: Wed, 04 Apr 2001 14:30:07 -0000
Thomas McNamara wrote:
>
>
> I've got WindowsME installed on a 20.4 Gig Maxtor, and want
> to dual boot Mandrake 7.2. Upon installation, when I get to the point
where
> "The DrakX Partitioning wizard found the following solutions:"
> I click on "Use Existing Partition" and get this error:
>
> ---The FAT resizer is unable to handle your partition, the following
> error occured: Can't locate object method "new" via package
> "resize_fat::main" at /usr/bin/perl-install_interactive.pm
> line 108, <> line 5---
>
> Now, I'm a newbie, so I'm stumped. I haven't had any problems with
> lnx4win, but that method makes for a sluggish OS. TIA
>
> Note: I can bypass this error by going into Expert Mode, but the I get
> another
> error when Mandrake attempts to install/configure the bootloader.
>
> I want to run Linux--I'm actually really excited about it--but I still
need
> Windows for some things, and don't want to settle for lnx4win.
>
>
> --Tom McNamara
>
> PS:
>
> FWIW, here are the rest of my system specs:
>
> Duron 700 Mhz
> Abit KT7 (KT133 chipset)
> 128 megs generic SDRAM
> Visiontek GeForce MX
> Soundblaster Live! Value
> Pioneer 16x DVD
> Realtek PCI 10/100 NIC
> Creative Labs PCI Modem Blaster
>
>
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: Dave Bossie <[EMAIL PROTECTED]>
Subject: Re: problems w/ Mandrake install
Date: Wed, 04 Apr 2001 14:30:09 -0000
Have you created separate partitions for your Linux and it's swap? I
believe that Lnx4win runs from a folder on your windows drive, but
Mandrake and other distros need their own partitions. Remember that you
must defrag your drive(s) before creating new partitions. I've set up my
20 gig as 2 fat32 partitions for Windows (1 gig & 9 gigs) and 2 Ext2
partitions for Linux (128 mb & 9+ gigs). Hope this helps!
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: "Marc Ulrich" <[EMAIL PROTECTED]>
Subject: more info: eth0 initialization fail
Date: Wed, 4 Apr 2001 10:03:22 -0400
I returned to linux to get the dmesg & see what info I could get. Here it
is:
...
eepro100.c:v1.09j-t 9/29/99 Donald Becker
http://cesdis.gsfc.nasa.gov/linux/drivers/eepro100.html
eepro100.c: $Revision: 1.36 $ 2000/11/17 Modified by Andrey V. Savochkin
<[EMAIL PROTECTED]> and others
eth0: Intel Corporation 82557 [Ethernet Pro 100], 00:D0:B7:E2:C6:CF, IRQ 19.
Receiver lock-up bug exists -- enabling work-around.
Board assembly 540775-040, Physical connectors present: RJ45
Primary interface chip i82555 PHY #1.
General self-test: passed.
Serial sub-system self-test: passed.
Internal registers self-test: passed.
ROM checksum self-test: passed (0x04f4518b).
Receiver lock-up workaround activated.
...
Perhaps that will shed light on something, but I'm rather clueless about
this problem.
eth1, which should be the 3com card, is not detected at all. It is a PCI
card, so that is bad -- right?
Marc
> I am using a 3com 3C905c and an intel EtherExpress Pro adapter. When the
> system boots up, eth0 takes about 5 minutes before it fails and gives me
the
> red [FAILED] signal. I've checked and I know that the drivers that I am
> using are the correct drivers. I also know that both interfaces are good
> because I am in Win2000 now and sending this email using one of them. I've
> also tried the other. Any ideas what causes this and what I can do to fix
> it? I am using kernel 2.4.2, but I also noticed that if I boot up using
the
> standard RH7's 2.2.16 kernel, the same thing happens.
>
> Thanks,
> Marc
>
>
------------------------------
** 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.setup.
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-Setup Digest
******************************
