Hi we had one of our mail servers , going on kernel panic mode ( not syncing fatal exception) ... cause /var/queue/postfix was on reiserfs part .. it has not been giving us any isssue quite some time , but yesterday and today it went on Kernel panic mode , when we hashed out the reiserfs part it booted properly .. we formated the part^ on ext3 later and things working fine now . Could any one have a idea why this could have happened . The server hung at 19:52:56 hours Here is the logs ------------------------------------------------------------------ ug 2 19:50:01 mmail2 snmpd[4460]: Connection from UDP: []:53980 Aug 2 19:50:55 mmail kernel: ReiserFS: sda6: warning: vs-13060: reiserfs_update_sd: stat data of object [2 203562 0x0 SD] (nlink == 2) not found (pos 1) Aug 2 19:51:33 mmail last message repeated 2 times Aug 2 19:52:34 mmail last message repeated 12 times Aug 2 19:52:52 mmail kernel: ReiserFS: sda6: warning: vs-13060: reiserfs_update_sd: stat data of object [2 203562 0x0 SD] (nlink == 2) not found (pos 1) Aug 2 19:52:56 mmail kernel: ReiserFS: warning: is_leaf: free space seems wrong: level=1, nr_items=2, free_space=65512 rdkey Aug 2 19:52:56 mmail kernel: ReiserFS: sda6: warning: vs-5150: search_by_key: invalid format found in block 37030. Fsck? Aug 2 19:52:56 mmail kernel: ReiserFS: sda6: warning: vs-13050: reiserfs_update_sd: i/o failure occurred trying to update [2 203562 0x0 SD] stat data Aug 2 20:13:27 mmail syslogd 1.4.1: restart. Aug 2 20:13:27 mmail kernel: klogd 1.4.1, log source = /proc/kmsg started. Aug 2 20:13:27 mmail kernel: Linux version 2.6.18-128.el5 ( mockbuild@builder10.centos.org) (gcc version 4.1.2 20080704 (Red Hat 4.1.2-44)) #1 SMP Wed Jan 21 10:41:14 EST 2009 Aug 2 20:13:27 mmail kernel: Command line: ro root=LABEL=/1 rhgb quiet Aug 2 20:13:27 mmail kernel: BIOS-provided physical RAM map: Aug 2 20:13:27 mmail kernel: BIOS-e820: 0000000000000000 - 00000000000a0000 (usable) Aug 2 20:13:27 mmail kernel: BIOS-e820: 0000000000100000 - 00000000cfb50000 (usable) Aug 2 20:13:27 mmail kernel: BIOS-e820: 00000000cfb50000 - 00000000cfb66000 (reserved)
On Mon, Aug 2, 2010 at 9:13 AM, Agnello George agnello.dsouza@gmail.com wrote:
(snip)
(snip)
This is a known problem and there is a patch for that:
http://bugs.centos.org/view.php?id=3156
Your kernel is so old (meaning many security holes). Please update the system and while doing so, grab the centosplus kernel which has the fix for this issue.
Akemi
On Tue, Aug 3, 2010 at 2:01 AM, Akemi Yagi amyagi@gmail.com wrote:
Hi Thanks for you reply , currently we have the default kernel 2.6.18-128.el5 ( centos 5.3 ) , can i install the kernel-2.6.18-8.1.8.el5.src.rpm and add the patch to this or do i have to download the kernel 2.6.21 then compile it and then install it . I had been to the kernel.org and cant find the kernel 2.6.21 , the kernels available are 2.6.27.49 and above .
On 08/03/10 12:44 AM, Agnello George wrote:
# yum update
should install the latest centos5 kernel (centosplus if you installed your reiser supporting kernel from there) and that should have the fixes installed.
On Tue, Aug 3, 2010 at 1:25 PM, John R Pierce pierce@hogranch.com wrote:
Hi I have install the kernel using the Centosplus repo .. currenlly the kernel is 2.6.18-194.8.1.el5.centos.plus but how would i know if that patch is fixed in this version of the kernel as mentioned in http://bugs.centos.org/view.php?id=3156
How about reading the bug you linked where it states quite explicitly that it was fixed in the 5.3 centos plus kernel (2.6.18-128.el5.centos.plus) and recognising the version number as previous to the one you just installed...
James
On Tue, Aug 3, 2010 at 6:13 AM, James Hogarth james.hogarth@gmail.com wrote:
And if you wish to confirm that the patch is in the current centosplus kernel you are running, issue a command:
rpm -q --changelog kernel-`uname -r` | more
You will find this line:
- added reiserfs patch (bug3156)
somewhere near the top of the output.
Akemi
On Tue, Aug 3, 2010 at 9:42 PM, Akemi Yagi amyagi@gmail.com wrote:
Thanks so much i was able to confirm it !! ------------------- here what i did
refer http://wiki.centos.org/AdditionalResources/Repositories/CentOSPlus
edit /etc/yum.repos.d/CentOS-Base.repo search [centosplus] enabled=1 includepkgs=kernel* jfsutils reiserfs-utils
yum --enablerepo=centosplus install kernel
cat /proc/filesystems |grep reiser
possible that kernel module and not yet loaded.
modprobe reiserfs
-
Agnello George -
Akemi Yagi -
James Hogarth -
John R Pierce