MCA error, but what to do with that?

A

akil

Hi,

I've experienced an MCA error, however, I don't know how to interpret them and it is hard to find out any useful documents about error reports.

Code: 
MCA: Bank 0, Status 0x9000004000010005
MCA: Global Cap 0x0000000000000c09, Status 0x0000000000000000
MCA: Vendor "GenuineIntel", ID 0x306a9, APIC ID 3
MCA: CPU 3 COR (1) internal parity error
MCA: Bank 0, Status 0x9000004000010005
MCA: Global Cap 0x0000000000000c09, Status 0x0000000000000000
MCA: Vendor "GenuineIntel", ID 0x306a9, APIC ID 2
MCA: CPU 2 COR (1) internal parity error

That refers to an error with parity on a CPU or on a memory bank, or maybe something else. If that occurs again should I send that hardware to my guarantee vendor or is this only an information that it was corrected or not?

Really I don't know what that would mean and how important that message really is, also I don't know if my (and which) hardware is broken?

Can somebody help me with it?
 
Oh, finally I found information about that. If you have overclocked your CPU, on high load you may notice above errors, bumping the Vcore may help you with it. To reproduce that I used burnCPU, after increasing Vcore those messsages don't appear again.

Anyway, if someone has any knowledge about that, maybe they can type some replies here because I may be wrong with that.
 
That is what I've got:
Code: 
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 3 BANK 0 
MCG status:
MCi status:
Error enabled
MCA: Unknown Error 5
STATUS 9000004000010005 MCGSTATUS 0
MCGCAP c09 APICID 3 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 58
mcelog: Unsupported new Family 6 Model 3a CPU: only decoding architectural errors
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 2 BANK 0 
MCG status:
MCi status:
Error enabled
MCA: Unknown Error 5
STATUS 9000004000010005 MCGSTATUS 0
MCGCAP c09 APICID 2 SOCKETID 0 
CPUID Vendor Intel Family 6 Model 58


I have changed my FSB from 103 Mhz to 100 Mhz (default). I will see if MCA appear again, if yes then the hardware will be sent to the vendor.
 
What kind of system is this? I hope you're not overclocking a server system in hopes of some tiny improvements in performance and throwing away the data integrity protection you get from ECC memory and cache checksumming.
 
FreeBSD 9.x, i7-3770, 16 GB DDR3, 6x HDD RAID. That is not my server but my private machine. I've got that message error when I recompile Perl dependencies after last update, however, a few weeks ago I run for 10 hours cpuburn and I didn't observe that kind of message. So I'm curious how compilation processes would influence that error, it doesn't overload the CPU like cpuburn does.
 
You must log in or register to reply here.
Back
Top