server rebooting itself often

[Frank DiPrete]

I want to just bounce this off you guys.

I have a machine, intel MB p4ht with e100, that has been rebooting 
itself after the latest fedora 8 yum update. Nothing in the logs, 
nothing on the console, just *paf!* reboot. Also have rsync problems 
with it. (hangs when rsyncing to the machine) The usual culprit for this 
sort of thing is heat or bad components.

So I tried:

Check Fans - still working

Ran with cover off in cool room - no change

Took the thing completely apart and cleaned all the gunk out - no change

Remove the cpu cooler, cleaned fins and fan, re-applied new heat sink 
compound. - no change

Installed lm_sensors to monitor temps and voltage - no alarms

Set up cron job to dump lm_sensors info to catch an alarm before reboot 
- nothing

installed Nagios monitor to check for alarms on lm_sensors - nothing

Ran memtest to completion (several hours) - no errors.


The latest change: I disabled ipv6.
It's been up for 3 days.

wtf.

I stumbled upon a forum post via google that mentioned turning off ipv6 
on a machine used as a NAT firewall / router. So that's why I tried it.

Anybody else run into strange problems with ipv6 turned on ?
kernel = 2.6.26.6-49.fc8