Machine check errors

[Alban Hertroys]

On 24 Jan 2011, at 6:02, perryh at pluto.rain.com wrote:

> Alban Hertroys <dalroi at solfertje.student.utwente.nl> wrote:
> 
>> Ever since installing 7.4-PRERELEASE I'm seeing MCA machine check
>> errors on my home-server ...
>> 
>> MCA: Bank 0, Status 0xb622000000000135
>> MCA: Global Cap 0x0000000000000104, Status 0x0000000000000004
>> MCA: Vendor "AuthenticAMD". ID 0x662, APIC ID 1
>> MCA: CPU 0 UNCOR PCC DCACHE L1 DRD error
>                       ^^^^^^^^^
>> MCA: Address 0x162933f0
> 
> That's an error in the on-chip data cache.

It is, I saw that too.

> The first thing I would check is cooling.  Maybe one of the fans
> has quit working, or the air filter (if the box has one) has gotten
> clogged up with dust, so that the CPU is running hotter than it
> should.

It's amazing how your perception of acceptable temperature ranges rises with the actual average temperature. Turns out the last time I unclogged the dust filter on the case was longer ago than I thought it was. Good call!
Temperature went down from 55C to 46C over the last 10 minutes already...

I think I'm starting to appreciate the MCA feature already :)

Alban Hertroys

--
Screwing up is an excellent way to attach something to the ceiling.


!DSPAM:363,4d3d2c0011731311418875!