Hi, long awaited, it's there now: The kernel updates for 7.1-8.1 are on their way to the mirrors. 8.2 already has the fix when it hits the stores. You'll find directories named kernel/2.4.18-20030324/ in each of the update directories of the 7.1, 7.2, 7.3, 8.0, 8.1 distributions. The changelogs are not fully consistent, but the packages are ok and fix the known problems. 8.1 has a 2.4.19 kernel, 8.0 and below have 2.4.18. Basically, the 2.4.19 kernel from 8.1 should run with the older distributions as well, but your mileage may vary. We have had some problems when we tested the kernel packages, mostly because the first suggested fixes were incorrect. This is why it took us so long. The announcement will follow later tonight. It will be copied to the update/*/kernel/2.4.1?-20030324 directories as soon as it is published. It will contain detailed information about the commands necessary to install the updates. In short: rpm -qf /boot/vmlinuz download the package for your system. rpm -Uhv your_package mk_initrd . /etc/sysconfig/bootloader case "$LOADER_TYPE" in *lilo*) lilo ;; *) ;; esac Roman.
Hi,
long awaited, it's there now: The kernel updates for 7.1-8.1 are on their way to the mirrors. 8.2 already has the fix when it hits the stores.
You'll find directories named kernel/2.4.18-20030324/ in each of the update directories of the 7.1, 7.2, 7.3, 8.0, 8.1 distributions. The changelogs are not fully consistent, but the packages are ok and fix the known problems.
8.1 has a 2.4.19 kernel, 8.0 and below have 2.4.18. Basically, the 2.4.19 kernel from 8.1 should run with the older distributions as well, but your mileage may vary.
We have had some problems when we tested the kernel packages, mostly because the first suggested fixes were incorrect. This is why it took us so long.
The announcement will follow later tonight. It will be copied to the update/*/kernel/2.4.1?-20030324 directories as soon as it is published. It will contain detailed information about the commands necessary to install the updates.
Does this mean the 2.2.x kernels for the 7.x distributions are not being patched? I know they are kinda old but there are still boxes out there running them. Regards Hubba ----------------------------------------------------------------------- - Hubertus Alexander Haniel ,,, http://www.unixcook.com - - Upnor, Kent, UK (o o) ICQ UIN: 1470384 - ----------------------------oOOo--(_)--oOOo----------------------------
Hi, how did you solve the ps problem with Alan Cox' cset-1.1076.txt patch (child processes refuse to show full path)? Thanx! Leppo. Am Montag, 24. März 2003 18:38 schrieb Roman Drahtmueller:
Hi,
long awaited, it's there now: The kernel updates for 7.1-8.1 are on their way to the mirrors. 8.2 already has the fix when it hits the stores.
You'll find directories named kernel/2.4.18-20030324/ in each of the update directories of the 7.1, 7.2, 7.3, 8.0, 8.1 distributions. The changelogs are not fully consistent, but the packages are ok and fix the known problems.
8.1 has a 2.4.19 kernel, 8.0 and below have 2.4.18. Basically, the 2.4.19 kernel from 8.1 should run with the older distributions as well, but your mileage may vary.
We have had some problems when we tested the kernel packages, mostly because the first suggested fixes were incorrect. This is why it took us so long.
The announcement will follow later tonight. It will be copied to the update/*/kernel/2.4.1?-20030324 directories as soon as it is published. It will contain detailed information about the commands necessary to install the updates.
In short:
rpm -qf /boot/vmlinuz download the package for your system. rpm -Uhv your_package mk_initrd . /etc/sysconfig/bootloader case "$LOADER_TYPE" in *lilo*) lilo ;; *) ;; esac
Roman.
-- "Es gibt keinen guten Krieg und keinen schlechten Frieden." (Benjamin Franklin)
The announcement will follow later tonight. It will be copied to the update/*/kernel/2.4.1?-20030324 directories as soon as it is published. It will contain detailed information about the commands necessary to install the updates.
Does this mean the 2.2.x kernels for the 7.x distributions are not being patched? I know they are kinda old but there are still boxes out there running them.
Our kernel packager/maintainer is also working on 2.2.19 for 7.1, yes. But this will probably last another few days. 7.0 is in discontinued state. Roman.
Hello Roman, hello List, I installed this kernel-update this night and after that I have a problem with my network-card. I'm running SuSE 8.1 on a Puretec-Root-Server with a network-card that uses the 8139too module. I can boot into a rescue-system and in the /var/log/boot.msg I see the following error: ##################### <notice>/etc/init.d/rc3.d/S05network start Setting up network interfaces: lo done eth0 ifup: Could not get a valid interface name: -> skipped failed <notice>'/etc/init.d/rc3.d/S05network start' exits with status 0 ##################### I need your help very quickly.... What should I do? Thanks in advance Mario
-----Original Message----- From: Roman Drahtmueller [mailto:draht@suse.de] Sent: Monday, March 24, 2003 6:38 PM To: suse-security@suse.de Subject: [suse-security] kernel updates
Hi,
long awaited, it's there now: The kernel updates for 7.1-8.1 are on their way to the mirrors. 8.2 already has the fix when it hits the stores.
You'll find directories named kernel/2.4.18-20030324/ in each of the update directories of the 7.1, 7.2, 7.3, 8.0, 8.1 distributions. The changelogs are not fully consistent, but the packages are ok and fix the known problems.
8.1 has a 2.4.19 kernel, 8.0 and below have 2.4.18. Basically, the 2.4.19 kernel from 8.1 should run with the older distributions as well, but your mileage may vary.
We have had some problems when we tested the kernel packages, mostly because the first suggested fixes were incorrect. This is why it took us so long.
The announcement will follow later tonight. It will be copied to the update/*/kernel/2.4.1?-20030324 directories as soon as it is published. It will contain detailed information about the commands necessary to install the updates.
In short:
rpm -qf /boot/vmlinuz download the package for your system. rpm -Uhv your_package mk_initrd . /etc/sysconfig/bootloader case "$LOADER_TYPE" in *lilo*) lilo ;; *) ;; esac
Roman.
-- Check the headers for your unsubscription address For additional commands, e-mail: suse-security-help@suse.com Security-related bug reports go to security@suse.de, not here
Hello Roman, hello List,
I installed this kernel-update this night and after that I have a problem with my network-card. I'm running SuSE 8.1 on a Puretec-Root-Server with a network-card that uses the 8139too module. I can boot into a rescue-system and in the /var/log/boot.msg I see the following error:
Ok, it somehow can't find the nic driver module.
See if
depmod -a
rcnetwork start
solves the problem.
I have the same card in my machine at home, but I'm using the rtl8139
driver (not 8139too).
vi /etc/modules.conf and search the file for all occurrences of eth0. The
(only) line should be close to the top. Remove the "too" from the module
name, run depmod -a;rcnetwork start and see if it works.
Right now, I can't understand why the driver should not work after the
update.
Roman.
--
- -
| Roman Drahtmüller
Hello again, I followed your advice but it does not work. How can I revert this update... Thanks Mario
-----Original Message----- From: Roman Drahtmueller [mailto:draht@suse.de] Sent: Tuesday, March 25, 2003 4:04 AM To: Mario Neubert Cc: suse-security@suse.de Subject: RE: [suse-security] kernel updates
Hello Roman, hello List,
I installed this kernel-update this night and after that I have a problem with my network-card. I'm running SuSE 8.1 on a Puretec-Root-Server with a network-card that uses the 8139too module. I can boot into a
rescue-system and in the
/var/log/boot.msg I see the following error:
Ok, it somehow can't find the nic driver module. See if
depmod -a rcnetwork start
solves the problem.
I have the same card in my machine at home, but I'm using the rtl8139 driver (not 8139too).
vi /etc/modules.conf and search the file for all occurrences of eth0. The (only) line should be close to the top. Remove the "too" from the module name, run depmod -a;rcnetwork start and see if it works.
Right now, I can't understand why the driver should not work after the update.
Roman. -- - - | Roman Drahtmüller
// "You don't need eyes to see, | SuSE Linux AG - Security Phone: // you need vision!" | Nürnberg, Germany +49-911-740530 // Maxi Jazz, Faithless | - -
Hello Roman, hello List, I installed this kernel-update this night and after that I have a problem with my network-card. I'm running SuSE 8.1 on a Puretec-Root-Server with a network-card that uses the 8139too module. I can boot into a rescue-system and in the /var/log/boot.msg I see the following error:
Ok, it somehow can't find the nic driver module. See if
depmod -a rcnetwork start
solves the problem.
I have the same card in my machine at home, but I'm using the rtl8139 driver (not 8139too).
Fine thing if you have direct access to your server. Not fine if it is somewhere in a serverfarm and can only be remoted via ssh or confix-lite. I think SuSE updates should be more reliable while the update runs. Had a similar problem with intel pro/100 nic on SuSE 7.3. While yast ran my server got out of control. For good sake a friend of me rebooted the server some hours later. The logs in /var/log/allmessages did not show the problem and nothing in /var/log/y2*. Debugging is not enabled in my kernel, so I cannot find the reason for this crash. The y2* logs even did not show me updates to the kernel. Now I'm scared to update any other servers running SuSE. Remark: I run Red Hat and Debian as well. With both distributions I never had problems with any updates and followed reboots of each system. Debian wins any uptimes and is totally free so I decide to change.
vi /etc/modules.conf and search the file for all occurrences of eth0. The (only) line should be close to the top. Remove the "too" from the module name, run depmod -a;rcnetwork start and see if it works.
This should be the part of the update and not of the user itself! Philippe
Am Dienstag, 25. März 2003 12:25 schrieb Philippe Vogel:
Hello Roman, hello List, I installed this kernel-update this night and after that I have a problem with my network-card. I'm running SuSE 8.1 on a Puretec-Root-Server with a network-card that uses the 8139too module. I can boot into a rescue-system and in the /var/log/boot.msg I see the following error:
Ok, it somehow can't find the nic driver module. See if
depmod -a rcnetwork start
solves the problem.
I have the same card in my machine at home, but I'm using the rtl8139 driver (not 8139too).
Fine thing if you have direct access to your server. Not fine if it is somewhere in a serverfarm and can only be remoted via ssh or confix-lite. I think SuSE updates should be more reliable while the update runs.
Same here (different hardware, though). Just installed k_deflt-2.4.19-274 on my 8.1, now X11 wouldnt start, no matter wether I installed the nvidia drivers that came with YOU or the latest from www.nvidia.com. Had to go back to the 'default' versions on both kernel and drivers. not funny. bye, [MH]
Mathias Homann wrote:
Just installed k_deflt-2.4.19-274 on my 8.1, now X11 wouldnt start, no matter wether I installed the nvidia drivers that came with YOU or the latest from www.nvidia.com. Had to go back to the 'default' versions on both kernel and drivers.
not funny.
depends. You chose to use the binary-only NVidia supplied drivers for X. Those are compiled for an specific kernel-release. When you updated your kernel you broke the driver. Go ask NVidia when they will supply compiled drivers for the bug-fixed kernel. And remind them that ".rpm" could contain proper dependencies, so that you will not break your system inadvertently. Or simply use your graphic card without 3D-acceleration and use the free "nv" driver, which came with X. Binary driver are bad for stable systems. Peter
Am Dienstag, 25. März 2003 13:54 schrieb Peter Wiersig:
Mathias Homann wrote:
Just installed k_deflt-2.4.19-274 on my 8.1, now X11 wouldnt start, no matter wether I installed the nvidia drivers that came with YOU or the latest from www.nvidia.com. Had to go back to the 'default' versions on both kernel and drivers.
not funny.
depends.
You chose to use the binary-only NVidia supplied drivers for X. Those are compiled for an specific kernel-release. When you updated your kernel you broke the driver.
But it wasn't a _version_ upgrade... just 2.4.29-74 -> 2.4.19-274? Never seen such a behaviour before...
Go ask NVidia when they will supply compiled drivers for the bug-fixed kernel. And remind them that ".rpm" could contain proper dependencies, so that you will not break your system inadvertently.
Then they will tell me 'but we _have_ SuSE rpm packages for kernel version 2.4.19...'
Or simply use your graphic card without 3D-acceleration and use the free "nv" driver, which came with X.
yea. on a machine dedicated to gaming that wouldn't be the best idea... bye, [MH] -- Die unaufgeforderte Zusendung einer Werbemail an Privatleute verstößt gegen §1 UWG und 823 I BGB (Beschluß des LG Berlin vom 2.8.1998 Az: 16 O 201/98). Jede kommerzielle Nutzung der übermittelten persönlichen Daten sowie deren Weitergabe an Dritte ist ausdrücklich untersagt! gpg key fingerprint: 5F64 4C92 9B77 DE37 D184 C5F9 B013 44E7 27BD 763C
Mathias Homann wrote:
Am Dienstag, 25. März 2003 13:54 schrieb Peter Wiersig:
But it wasn't a _version_ upgrade... just 2.4.29-74 -> 2.4.19-274? Never seen such a behaviour before...
I expect every binary-only module to break when I upgrade my kernel. Less nasty surprises on my side, I guess.
Go ask NVidia when they will supply compiled drivers for the bug-fixed kernel. And remind them that ".rpm" could contain proper dependencies, so that you will not break your system inadvertently.
Then they will tell me 'but we _have_ SuSE rpm packages for kernel version 2.4.19...'
Because of that .rpm have that "release:" number, which is now "274", as you posted. If they've tested only with "-74" they need to retest with "-274". Have you tried to reinstall the nividia-kernel package and ran "depmod" afterwards. If the interface and compile options were the same even the nvidia module should load. Another alternative is the use of the nvidia-kernel.src.rpm to build the kernel module on your machine. But you need additional software for that, which I would not install on a gaming machine. Peter
Am Dienstag, 25. März 2003 15:18 schrieb Peter Wiersig:
Have you tried to reinstall the nividia-kernel package and ran "depmod" afterwards. If the interface and compile options were the same even the nvidia module should load.
Did that. it _did_ load, just that my machine crashed when starting X11. Without any hints in ~/.X.err or /var/log/XFree86.0.log at all.
Another alternative is the use of the nvidia-kernel.src.rpm to build the kernel module on your machine. But you need additional software for that, which I would not install on a gaming machine.
well its not only a gaming box but I _do_ have a few games on it ;) The 'additional software' would be no problem. bye, [MH] -- Die unaufgeforderte Zusendung einer Werbemail an Privatleute verstößt gegen §1 UWG und 823 I BGB (Beschluß des LG Berlin vom 2.8.1998 Az: 16 O 201/98). Jede kommerzielle Nutzung der übermittelten persönlichen Daten sowie deren Weitergabe an Dritte ist ausdrücklich untersagt! gpg key fingerprint: 5F64 4C92 9B77 DE37 D184 C5F9 B013 44E7 27BD 763C
Mathias Homann wrote:
Am Dienstag, 25. März 2003 15:18 schrieb Peter Wiersig:
Have you tried to reinstall the nividia-kernel package and ran "depmod" afterwards. If the interface and compile options were the same even the nvidia module should load.
Did that. it _did_ load, just that my machine crashed when starting X11. Without any hints in ~/.X.err or /var/log/XFree86.0.log at all.
SuSE 8.1 - new kernel and nvidia gforce4: no problems at all. X starts fine, network etc. is working without problems (and the ptrace exploit doesn't work anymore ;) so i cannot confirm such problems here. Regards, Sven
Mathias Homann wrote:
Am Dienstag, 25. März 2003 15:18 schrieb Peter Wiersig:
Have you tried to reinstall the nividia-kernel package and ran "depmod" afterwards. If the interface and compile options were the same even the nvidia module should load.
Did that. it _did_ load, just that my machine crashed when starting X11. Without any hints in ~/.X.err or /var/log/XFree86.0.log at all.
Hm, that's strange. I was always unable to laod the kernel-module, the glx-package was never a problem. And when my machine crashed it always left some trace I was able to look up afterwards. Let's discuss that topic in a more appropriate ML: suse-linux@suse.com Peter
Hi folks, Seems this discussion slowly shifts towards the multimedia list :)
Another alternative is the use of the nvidia-kernel.src.rpm to build the kernel module on your machine. But you need additional software for that, which I would not install on a gaming machine.
Probably should be mentioned that nvidia-kernel.src.rpm does not compile by just rpm --rebuild, and the tar.gz does not compile with "make" (at least on my SuSE 7.3) Eduard __________________________________________________ Do you Yahoo!? Yahoo! Platinum - Watch CBS' NCAA March Madness, live on your desktop! http://platinum.yahoo.com
On Tuesday 25 March 2003 04:26 pm, Mathias Homann wrote:
Am Dienstag, 25. März 2003 15:18 schrieb Peter Wiersig:
Have you tried to reinstall the nividia-kernel package and ran "depmod" afterwards. If the interface and compile options were the same even the nvidia module should load.
Did that. it _did_ load, just that my machine crashed when starting X11. Without any hints in ~/.X.err or /var/log/XFree86.0.log at all.
Another alternative is the use of the nvidia-kernel.src.rpm to build the kernel module on your machine. But you need additional software for that, which I would not install on a gaming machine.
well its not only a gaming box but I _do_ have a few games on it ;) The 'additional software' would be no problem.
bye, [MH] au contraire ...
I have recompiled the P_SMP kernel ... and the binnary from nvidia is still working ! But for a much stable system I needed to recompile also alsa9 and cipe.
After kernel update for 7.3, the alsa sound modules don't work. It seems like changing the snd_id=card1 to snd_id=1 and changing the module name from snd-card-intel8x0 to snd-intel8x0 lets the modules to be loaded but the alsactl can't set the mixer values and fails with protocal mismatch. I believe we need alsa module updates to go along with this kernel upgrade and possible yast2 updates too, to properly configure the sound cards. Selcuk
-----BEGIN PGP SIGNED MESSAGE----- Am Dienstag, 25. März 2003 16:22 schrieben Sie:
After kernel update for 7.3, the alsa sound modules don't work. It seems like changing the snd_id=card1 to snd_id=1 and changing the module name from snd-card-intel8x0 to snd-intel8x0 lets the modules to be loaded but the alsactl can't set the mixer values and fails with protocal mismatch.
I'm experiencing similiar problems, but with module snd-card-ens1371. /var/log/messages is complaining about a unlocateble module "sound-slot-0". Any hints or workarounds? I don't know how to handle this problem.
I believe we need alsa module updates to go along with this kernel upgrade and possible yast2 updates too, to properly configure the sound cards.
Not so funny. But fortunately I can live with a non-working soundcard for a couple of days. Tobias - -- - ------------------------------------- Tobias Cremer Hohenzollernstr. 20 80801 München to.c@gmx.net http://www.tobiwan-kenobi.de -----BEGIN PGP SIGNATURE----- Version: 2.6.3in Charset: noconv iQCVAwUBPoC/Drd9JiPf/I1hAQFdSQQAkiuSbAo01Lc38B+vRxVHHiCx3KKyMDaL EwivYrgrsbK1x2XJi2b83W/hWzLzuCakQL4DLrTwnOhVzyxH3OMJROcPNAN6XnDD SsDkFpYh6Bwr6n2VZZrFD7ZZtSZqlrWmLYyQ0oQIYr/M1rvO2gZw+gkE35r/Dcj7 uEzI3tZxcdE= =r+D8 -----END PGP SIGNATURE-----
Hi Tobias,
Am Dienstag, 25. März 2003 16:22 schrieben Sie:
After kernel update for 7.3, the alsa sound modules don't work. It seems like changing the snd_id=card1 to snd_id=1 and changing the module name from snd-card-intel8x0 to snd-intel8x0 lets the modules to be loaded but the alsactl can't set the mixer values and fails with protocal mismatch.
I'm experiencing similiar problems, but with module snd-card-ens1371. /var/log/messages is complaining about a unlocateble module "sound-slot-0". Any hints or workarounds? I don't know how to handle this problem.
I believe we need alsa module updates to go along with this kernel upgrade and possible yast2 updates too, to properly configure the sound cards.
--> I saw similar messages when compiling my own kernel from the kernel sources on the SuSE CDs. What helped me was to download the newest ALSA sources from http://www.alsa-project.org I have to re-compile my sound card module every time I compile a new kernel. Maybe this helps to solve your problem ? Good luck ! Armin -- Am Hasenberg 26 office: Institut für Atmosphärenphysik D-18209 Bad Doberan Schloss-Straße 6 Tel. ++49-(0)38203/42137 D-18225 Kühlungsborn / GERMANY Email: schoech@iap-kborn.de Tel. +49-(0)38293-68-102 WWW: http://armins.cjb.net/ Fax. +49-(0)38293-68-50
participants (13)
-
Armin Schoech
-
Eduard Avetisyan
-
Hubertus A. Haniel
-
Leppo
-
Mario Neubert
-
Mathias Homann
-
Peter Wiersig
-
Philippe Vogel
-
Roman Drahtmueller
-
Selcuk Ozturk
-
SuSE ASPEC GroundZero
-
Sven 'Darkman' Michels
-
tobiwan