[Users] Long: Debian etch, linux-image-2.6.18-openvz-13-39.1d1-amd64 -> Kernel panic

[Martin Trtusek]

Hello all,

 I experienced "Kernel panic - not syncing: Aiee, killing interrupt
handler!" with linux-image-2.6.18-openvz-13-39.1d1-amd64 on Debian etch
on 3 servers from 4 where this kernel was installed. I am new in OpenVZ,
perhabs it is faulty on my side. But this info can be interesting for
other users with Debian etch.

 I installed linux-image-2.6.18-openvz-13-39.1d1-amd64 kernel on 4
servers without setting up any VPS (I would like to prepare servers for
later switch some daemons to VPS). 3 of them crashes soon or later
(particular below). Today I was able to do foto of crash screen
(previous two crash was on server outside, and I did not have
possibility to see screen), foto is enclosed.

 On all 4 servers i did installation on this way:
- wget
http://download.openvz.org/kernel/debian/etch/linux-image-2.6.18-openvz-13-39.1d1-amd64_028.39.1d1_amd64.deb
- dpkg -i linux-image-2.6.18-openvz-13-39.1d1-amd64_028.39.1d1_amd64.deb
- reboot # to kernel linux-image-2.6.18-openvz-13-39.1d1-amd64
- aptitude install vzctl vzquota # from etch -> vzctl version 3.0.11
- reboot

1. server:
- HP Proliant DL140 G3, CPU 2x Xeon 3GHz, 2048MB RAM
 * 2x 80GB SATA HDD in Linux Software RAID level 1
- linux-image-2.6.18-openvz-13-39.1d1-amd64 installed 11/9/07
- no crash until today

2. server
- HP Proliant ML110G4, Pentium(R) D CPU 2.80GHz, 1024MB RAM
 * 3x 250GB + 1x160GB SATA HDD in Linux Software RAID level 5
- linux-image-2.6.18-openvz-13-39.1d1-amd64 installed 13/9/07
- crash 5 minute after installation, then until today without crash
 * in logs nothing usefull

3. server
- HP Proliant ML350 G5, CPU DualCore Xeon 1,6GHz/1066MHz, 2048MB RAM
 * 4x160GB SATA HDD in Hardware RAID 10 (/dev/cciss/c0d0)
- linux-image-2.6.18-openvz-13-39.1d1-amd64 installed 13/9/07
- crash 18/9/07 about 22:00, then until today without crash
 * in logs was row about regular 5min cron jobs, whereas missing one
from series was /usr/share/vzctl/scripts/vpsnetclean

4. server
- LYNX server AspenHill, Intell Core Duo E6320, 4GB RAM
 * 4x 320GB SATA HDD in Linux Software RAID level 10
- linux-image-2.6.18-openvz-13-39.1d1-amd64 installed 21/9/07 at 18:17
- crash 21/9/07 about 21:30
 * in logs is full series of 5min cron jobs (last
one /usr/share/vzctl/scripts/vpsreboot)
 * foto of screen enclosed

After crash of server 4, I decided
linux-image-2.6.18-openvz-13-39.1d1-amd64 is unstable for me, and I
downgraded all 4 servers to linux-image-2.6.18-openvz-13-1etch4-amd64
(by downloading prepared deb from download.openvz.org). Hope this solve
my situation.

Maybe the problem is in version of vzctl (kernel patch for
13-39.1d1-amd64 is in lenny and there is vzctl in version 3.0.18-1
instead 3.0.11-13 from etch).


Martin Trtusek
-------------- next part --------------
A non-text attachment was scrubbed...
Name: KernelPanic_20070921.jpg
Type: image/jpeg
Size: 847489 bytes
Desc: not available
Url : http://openvz.org/pipermail/users/attachments/20070923/de554b04/KernelPanic_20070921-0001.jpg