Hello

 

So, my unRAID server runs a Z9PE-D8 WS motherboard with 64gb ram(ECC). Recently started to get a weird " mce: [Hardware Error]: Machine check events logged" in my logs so i installed mcelog to find out more, here's the output:

Quote

Hardware event. This is not a software error.
MCE 0
CPU 8 BANK 5
MISC 21420a6a86 ADDR 109bcd5200
TIME 1501366001 Sun Jul 30 00:06:41 2017
MCG status:
MCi status:
Error overflow
Corrected error
MCi_MISC register valid
MCi_ADDR register valid
MCA: MEMORY CONTROLLER RD_CHANNEL0_ERR
Transaction: Memory read error
STATUS cc1774c000010090 MCGSTATUS 0
MCGCAP 1000c14 APICID 20 SOCKETID 1
CPUID Vendor Intel Family 6 Model 45
 

 

From doing a bit of googling it seems like it might be an issue with a memory stick, that's fine i have a spare one so i can just replace it if that's the case. But what do you guys say, does this look like a memory-stick issue?

 

If this is the case, how do i locate the correct memory stick on my board? After all, there's 8 of them. I couldn't find any good schematics to go after, this is the only one that i cound find but unfortunately it's not to helpful:

https://www.pugetsystems.com/zoom_pic.php?id=19560

Check your bios, there should be a system event log and there could be more details there.

Thanks, but that didn't give me anything. Neither from there nor through the system log from my IPMI card.

 

I also ran a memtest overnight now and that gave me no errors so i'm not quite sure what's up here or if i should take that message seriously.

Not sure how it works on those boards, but any ECC error should be logged there, that's how it works with Supermicro boards for example.