2.6.30.5-43.fc11.x86_64 hang or crash.

[Paul]

gary artim wrote:
> Hi --
>
> Anyone seeing machine lockup with this kernel? profile:
>
> supermicro AS1041-T2 (1u box)
> 12 AMD 2.4 opterons
> 48GB memory
> BIOS is most current
>
> I get nothing on the console, last output is:
> Sep 21 13:51:23 n0 ntpd[2036]: synchronized to 72.52.190.26, stratum 2
> Sep 21 14:58:54 n0 ntpd[2036]: synchronized to 140.99.51.114, stratum 2
>
> no oops.
>
> I added to ksyslog.conf option for *.kernel, waiting for it to happen.
> Happened 3x today.
>
> attached is lshw output.
>
> The box itself is acting flaky, need to turn it on and off sometimes
> to get a clean
> start. Ran memcheck -- no errors.
>
> stressed system - ping of death, cp /dev/zero /tmp/z , this ran fine
> once, hung the system/hardware
> the second time. One note is the system was bought in 10/2007 -- i
> checked the logs in the
> bios and didn't see any errors, but did note that of the 3 chips (quad
> cpus) one had a higher temp
> then the others 78c vs 22c -- maybe new thermal grease? the system is
> used round the clock on
> cpu bound processes that sometime run for weeks.
>
> Any advice is greatly appreciated?
>
> -- gary
>   
78 vs 22? Yeah, that's where I would start. Thermal paste is cheap.

Cheers,


-- 


Paul