(SOLVED) self-detected stall on CPU; server freezes/crashes

[Telos]

In the past 2 weeks my UNRAID server has gone down twice in the same way. It does not respond remotely or locally if I connect a monitor and keyboard, though the system is still on. The second time around I was able to capture the syslog file and also noticed that the GUI login on the local display was completely frozen.

 

- UNRAID Version 6.8.0 2019-12-10

- mainly use the system for Plex (with Sonarr, Radarr, and sab)

- I run my system on a UPS and there was no reported events logged on the UPS

- looking through syslog I found an error that states "rcu: INFO: rcu_sched self-detected stall on CPU" and "CPU: 3 PID: 13243 Comm: Thread Pool Wor Tainted: G D W O 4.19.88-Unraid #1" at the time of the crash

- over the holidays my docker image had become corrupted and mounted as read-only so I deleted it and reinstalled from the templates

- FixCommonProblems was not able to find any issues even with the extended test

 

I'm a little worried that it's an issue with the CPU itself, since it's an older i7 2600. I searched for some other threads on the error but wasn't able to find much other than one that mentioned something about the filesystem type. My servers been running well over the last year or two and the only changes recently have been the OS upgrade and the docker image problem I mentioned earlier.

 

Any insights or assistance would be appreciated!

tower-diagnostics-20200110-1745.zip Unraid Syslog.txt

[JorgeB]

8 hours ago, Telos said:

over the holidays my docker image had become corrupted and mounted as read-only so I deleted it and reinstalled from the templates

There was also corruption on the cache xfs filesystem, did you fix that?

 

In any case probably a good idea to run memtest for a few hours.

[Telos]

14 hours ago, johnnie.black said:

There was also corruption on the cache xfs filesystem, did you fix that?

 

In any case probably a good idea to run memtest for a few hours.

Thanks for catching that, I've repaired the filesystem and am now running memtest.

 

Since I haven't been able to directly reproduce the issue I'll have to monitor and report back after a few days.

 

**EDIT** Memtest has finished 1 pass with no errors, and the server is back up for now.

Edited January 11, 2020 by Telos

[Telos]

Just following up to this, I haven't had any issues in the past week. I'll mark this as solved, thanks for the assistance!