Hi Unraid-ers,

 

i have used unraid since last June, recently upgraded my rig (Switch CPU from 5820K to 6950X, add a GeForce GTX 1060 3GB and add 2x8GB RAM), then it starts randomly lockup, all docker containers frozen but VMs are working, can access GUI but dashboard is not accessible.

 

After dig up to syslog and found PCIe error in there:

Quote

Apr  1 20:20:39 UnRAID-Tower kernel: pcieport 0000:00:01.1: AER: Uncorrected (Fatal) error received: 0000:00:00.0

Apr  1 20:20:39 UnRAID-Tower kernel: nvme 0000:04:00.0: AER: PCIe Bus Error: severity=Uncorrected (Fatal), type=Inaccessible, (Unregistered Agent ID)

Apr  1 20:20:39 UnRAID-Tower kernel: nvme nvme0: frozen state error detected, reset controller

Apr  1 20:20:40 UnRAID-Tower kernel: pcieport 0000:00:01.1: AER: Root Port link has been reset

Apr  1 20:20:40 UnRAID-Tower kernel: pcieport 0000:00:01.1: AER: device recovery successful

Apr  1 20:21:06 UnRAID-Tower kernel: pcieport 0000:00:01.1: AER: Corrected error received: 0000:00:01.1

Apr  1 20:21:06 UnRAID-Tower kernel: pcieport 0000:00:01.1: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Receiver ID)

Apr  1 20:21:06 UnRAID-Tower kernel: pcieport 0000:00:01.1:   device [8086:6f03] error status/mask=00000080/00002000

Apr  1 20:21:06 UnRAID-Tower kernel: pcieport 0000:00:01.1:    [ 7] BadDLLP              

Apr  1 20:27:47 UnRAID-Tower kernel: pcieport 0000:00:01.1: AER: Corrected error received: 0000:00:01.1

Apr  1 20:27:47 UnRAID-Tower kernel: pcieport 0000:00:01.1: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Receiver ID)

Apr  1 20:27:47 UnRAID-Tower kernel: pcieport 0000:00:01.1:   device [8086:6f03] error status/mask=00000080/00002000

Apr  1 20:27:47 UnRAID-Tower kernel: pcieport 0000:00:01.1:    [ 7] BadDLLP    

 

From System Device, i found this nvme0 is my SSD:

Quote

[N:0:1:1]disk LENSE20256GMSP34MEAT2TA__1 /dev/nvme0n1 256GB

 

I have researched in forum, people usually have PCIe error but with "corrected" and "non-fatal" message showing, and can be resolved by updating BIOS or add script to Sysconfig, my case is much different.

 

I am wondering if PCIe Bus Error: severity=Uncorrected (Fatal), type=Inaccessible, (Unregistered Agent ID) means NVMe SSD is faulty?

 

Here are specs of my rig:

MOBO: ASUS x99-deluxe

CPU: Intel i7-6950X

RAM: Corsair Vengeance 2666MHz 6x8GB

NVMe SSD: a Lenovo laptop M.2  

GPU: GeForce GTX 1060 3GB

 

Syslog is also attached, recently lockup happens around "Apr  1st 20:00-20:30" in line 61751

 

Thanks in advance.

Wesley

syslog

Edited April 28, 20224 yr by Wesley_Sun

Solved after moving to a new NVMe SSD and upgrad to 6.10.0 rc5