xfs problems (possibly after upgrading from linux kernel 2.6.27.10 to .14)

[Carsten Aulbert]

Hi Michael,

Michael Monnerie schrieb:
> "Carsten Aulbert" <carsten.aulbert at aei.mpg.de> schrieb:
>> Feb 16 20:34:49 n0035 kernel: [275873.343508] Filesystem "sda6":
> Corruption of
>> in-memory data detected.  Shutting down filesystem: sda6
> 
> Are you absolutely sure your RAM doesn't have problems? Run memtest86.
> At least I would from that error description.

Pretty sure, all boxes have ECC memory and mcelog has not reported
anything on the nodes in question right now - we usually have only a few
spurious MCE errors a week. We have encountered 17 of these incidences
within the past week and basically nothing before - I don't think that
memory in 17 nodes goes bad just by upgrading the kernel or out of
coincidence.

I'll pick 4-5 nodes for memtests just to be sure.

Cheers

Carsten