How to decode MCE errors?

[idata]

Hello,

Intel Compute Module HNS2600KPF

S2600KPF mobo

2x Xeon E5-2670

BIOS 01.01.0016, BMC 1.43.9685, ME 03.01.03.021, SDR 1.14

CentOS 6.8, kernel 2.6.32-642.3.1.el6.x86_64

After reboot I have the following entry in mcelog:

Hardware event. This is not a software error.

MCE 0

CPU 12 BANK 20

MISC 100000000000000

TIME 1470234024 Wed Aug 3 16:20:24 2016

MCG status:

MCi status:

Error overflow

Uncorrected error

MCi_MISC register valid

Processor context corrupt

MCA: BUS error: 1 12 Level-3 Generic Generic Other-transaction Request-did-not-timeout

Running trigger `bus-error-trigger'

QPI: Intel QPI link layer detected control error from R3QPI

STATUS ea00000000100e0f MCGSTATUS 0

MCGCAP 7000c16 APICID 20 SOCKETID 1

CPUID Vendor Intel Family 6 Model 63

Can anybody decode this error ?

thanks in advance, Pawel

[idata · Accepted solution]

Hello,

We would like to inform you that decoding MCE errors is out of the scope of support for us. As a general recommendation based on the wording for the log, please check the memory modules installed and confirm if one of them has failed.