Help with MCELOG hardware issue found by Fix Common Problems

[TwoFive5]

Hi all,

 

So Fix common problems found a hardware error and suggested using mcelog to find out what the issue is. I have attached my log files below.

 

I am not sure if this is the error it is detecting, but in the log this is the error I keep seeing over and over:

 

Feb 24 09:37:00 Unraid kernel: usb 1-3.2: new low-speed USB device number 116 using ehci-pci
Feb 24 09:37:00 Unraid kernel: usb 1-3.2: device descriptor read/64, error -32
Feb 24 09:37:00 Unraid kernel: usb 1-3.2: device descriptor read/64, error -32

I am guessing it might be my USB stick going bad, but that seems strange as it is only 6 months old. All of this is running on a Dell R710 if that matters.

 

Hopefully someone can give me some advice on this.

 

Thanks!

 

 

unraid-syslog-20200224-1705.zip

[trurl]

We seldom want just syslog.

 

Go to Tools - Diagnostics and attach the complete diagnostics zip file to your NEXT post.

 

Diagnostics includes syslog and many other useful things.

[TwoFive5]

5 hours ago, trurl said:

We seldom want just syslog.

 

Go to Tools - Diagnostics and attach the complete diagnostics zip file to your NEXT post.

 

Diagnostics includes syslog and many other useful things.

 

Ah ok,

 

I thought I had included that. Here is the additional information.

 

 

unraid-diagnostics-20200224-1617.zip

[trurl]

Do you have any actual symptoms?

 

Have you done memtest?

 

 

[TwoFive5]

No, there doesn't seem to be any symptoms yet. But this just started a few days ago from what I can tell.

 

Have not done a memtest. I am assuming you mean for the RAM?

 

I have seen an error before on some of the ram sticks but it seemed to go away.

 

I was planning on pulling the RAM and CPU's this weekend to upgrade them.

[trurl]

3 minutes ago, TwoFive5 said:

I have seen an error before on some of the ram sticks but it seemed to go away.

Where did you see these errors if you didn't do a memtest? There should never be any memory errors. Everything goes through memory. Your data, the executable code, everything.

[TwoFive5]

It was an error reported on the server's idrac. I moved some of the sticks around thinking the order I had them in was wrong.

 

Sat Dec 21 2019 06:44:11	Persistent correctable memory error rate has increased for a memory device at location DIMM_A2.	
Thu Oct 24 2019 17:30:28	Persistent correctable memory error rate has increased for a memory device at location DIMM_A2.

 

I am running a memtest now.

Edited February 25, 2020 by TwoFive5
Added Information

[TwoFive5]

It has completed 1 pass with no errors so far

[trurl]

2 hours ago, TwoFive5 said:

It was an error reported on the server's idrac. I moved some of the sticks around thinking the order I had them in was wrong.

 

Sat Dec 21 2019 06:44:11	Persistent correctable memory error rate has increased for a memory device at location DIMM_A2.	
Thu Oct 24 2019 17:30:28	Persistent correctable memory error rate has increased for a memory device at location DIMM_A2.

 

I am running a memtest now.

Is this ECC memory? I didn't see that mentioned when I googled that Dell model.

[TwoFive5]

Yes it is ECC 12x2GB DDR3

[trurl]

1 hour ago, TwoFive5 said:

Yes it is ECC 12x2GB DDR3

Memtest won't get any errors since the memory is self-correcting.

[TwoFive5]

12 hours ago, trurl said:

Memtest won't get any errors since the memory is self-correcting.

That makes sense, I never knew that was the case with ECC ram.