Panic: NMI indicates hardware failure

[F1aX]

Hello.

I've been running Freenas for 2.5years now without any issues.
This morning my nas was not responding, so I connected to the IPMI and opened up the remote console.

I found the following:
panic: NMI indicates hardware failure

I restarted the server and it has been running for about an hour now.
I'd like to find the reason for the panic, but don't know where to start looking.

I've got the Asrock C2750D5I with 4x8Gb ECC ram.
I've been running 9.10.2-u4 until a couple of days ago, when I upgraded to 11.0-u2. (2017-08-04)

Any ideas?

 

[Ericloewe]

What does the IPMI log say?

 

[F1aX]

What does the IPMI log say?

Nothing really.. :/

 

[Ericloewe]

Nothing at all? The system threw a non-maskable interrupt. That's the kind of thing that is almost certainly caused by a problem logged by IPMI.

 

[F1aX]

This is the eventlog.

 

[farmerpling2]

In debugging NMI problems, over the years, I am not surprised that IPMI contains nothing.

You need to look at logs looking for some spurious event that seems out of place or is a one off. Hopefully logs will contain something. Look for retrys, offline occurring on controller cards/MB embedded parts.

 

[Ericloewe]

This is the eventlog.

*sigh*, they couldn't even be bothered to decode those...

 

[rs225]

Isn't this the dead Atom CPU problem?

 

[Ericloewe]

Hmm, the board is probably affected, but this doesn't match that issue.

 

[F1aX]

The only thing i found in the logs from that time, was the network going down/up for some reason.

 

[farmerpling2]

Is it onboard NIC or NIC card?