Bug#455723: Linux kernel 2.6.18-5 locks up (null pointer dereference) reproducibly when using serial console (via IPMI)

[dann frazier <dannf@debian.org>]

On Tue, Dec 11, 2007 at 01:33:10PM +0100, Adorj??ni G??bor wrote:
> Package: linux-image-2.6.18-5-amd64
> Version: 2.6.18.dfsg.1-13etch5
> Severity: critical
> 
> After today's kernel upgrade from the Etch security repository, I rebooted our
> server and watched the boot process via serial console tunneled through the
> IPMI BMC's Serial Over Lan feature. To my surprise, the kernel died with a
> null pointer dereference!

Thanks for the detailed report. Unfortunately, I can't reproduce it on
my system, nor do I see anything in the changes that might contribute
to the problem. Since you say its 100% reproducible, my only guess is
a change in timing.

> If you need any further information, please contact me on my mail address.
> This is a production machine, so I am not going to reboot it for further
> experiments.

It would be interesting to see if this fixes the issue for you:
  http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=fdc30b3d448bf86dd45f9df3e8ac0d36a3bdd9b2

Also, fwiw, it looks like the sysstat package is triggering the
problem. A possible workaround is to remove or disable this package.
If you've made any changes to the default sysstat configs, I'd also
be interesting to try them locally to see if I can reproduce.

-- 
dann frazier