Bug#278183: Acknowledgement (kernel-image-2.6.8-1-686: "invalid operand" in syscall)

["Ph. Marek" <philipp.marek@bmlv.gv.at>]

Today I found the following in /var/log/messages:
 Oct 29 03:50:06 inet -- MARK --
 Oct 29 04:10:06 inet -- MARK --
 Oct 29 04:30:06 inet -- MARK --
 Oct 29 04:34:45 inet syslogd 1.4.1#10: restart.
 Oct 29 04:50:06 inet -- MARK --
 Oct 29 07:24:38 inet syslogd 1.4.1#10: restart.

Where 7:24 is the time I rebooted the machine.

What can be the cause that syslogd can be restartet?
I believe that it was just a signal - as the 20 minutes intervals are not 
disturbed.

I had a vmstat running overnight some days ago - but it seems not to be an 
OOM.
/proc/slabinfo showed some changes, mostly reiserfs inodes ... but extended 
memory usage should be shown in vmstat too?