mcelog reports AMD DRAM Parity Error?

[jarod-ajLrJawYSntWk0Htik3J/w at public.gmane.org (Jarod Wilson)]

On Dec 1, 2010, at 10:28 AM, Derek Atkins wrote:

> Jarod,
> 
> Just noticed a few more entries in my MCElog.  Unfortunately the log
> entries aren't time stamped, but the file is dated yesterday, so
> obviously the last one was on the 30th around 13:01.
> 
> Argh..  This is my production system; I don't like seeing hardware
> issues, and I don't really have the time to go an explore issues.  I
> could move the services (read: VMs) off onto another piece of
> hardware...  if I had one to move to.

Trust me, I feel your pain.


> I've installed edac-utils, but not sure what to do with it.

To be honest, I'm not sure either. :) Its been probably 3 years since
I last poked at it any. I have vague recollections of some way to
load a memory slot name mapping file provided with edac-utils into
the kernel, so that error logs would say "its slot CPU0_DIMM1 where
the error happened".


> This machine is under 3 years old (possibly under 2 -- I'd have to go
> check); I shouldn't be having these issues.  :-/

As they say, stuff happens... :(


-- 
Jarod Wilson
jarod-ajLrJawYSntWk0Htik3J/w at public.gmane.org