Random lock-ups

[Ulf Thomas Johansen]

Hi.

 

Been running a Ryzen 5600G on ROG Strix 570-F for a while now and I'm experiencing random lock-ups where I have to hard kill the server to get it back up. Luckily - I've recovered every singel time speaking volumes of the robustness of Unraid.

 

I do however have to try to find the cause and I have now reduced the amount of dockers to a minimum (Plex, MariaDB and Mosquitto) and I am running just 1 vm (Home Assistant). Furthermore I have reset the bios to default and removed all changes to Unraid. I have also upgraded to 6.10-rc2.

 

Whilst waiting for a lock-up I peeked into the syslog and found some strange logging between Dec 18 00:53:02 and 01:44:40 (right before it locked-up last time) which I hope you guys could help me understand. (Diags attached)

 

Any other insight into why my rig is not stable would be welcomed. For the record I've run a full pass with the latest memtest and I have replaced the motherboard.

 

Hopefully something will pop up.

 

//UlfThomas

 

 

algarheim-diagnostics-20211218-2053.zip

Edited December 18, 2021 by Ulf Thomas Johansen

[Squid]

8 minutes ago, Ulf Thomas Johansen said:

Dec 18 00:53:02 and 01:44:40

The diagnostics are after you rebooted, so we can't see what you saw...

 

But to get the obvious out of the way, have you looked at https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/?tab=comments#comment-819173?

 

Have you done a memtest?  Also try disabling the XMP/AMP overclock on the memory...

 

 

 

[Ulf Thomas Johansen]

Ah... correct. I have attached the full syslog here containing the logging before the lock-up.

 

I have run a full pass with memtest once and I have removed 1 ram chip at a time when a lock-up happens. Basically rotating all ram. Same result.

 

I have also run with and without overclocking.

 

 

syslog.tar.gz

[Ulf Thomas Johansen]

Adding to my investigations: those loglines seem to be related to the corefreq addon and since my cpu is not on their list I have removed the plugin. Could this perhaps have contributed to my lock-ups?

 

[ChatNoir]

2 hours ago, Ulf Thomas Johansen said:

Adding to my investigations: those loglines seem to be related to the corefreq addon and since my cpu is not on their list I have removed the plugin. Could this perhaps have contributed to my lock-ups?

Possibly.

[Ulf Thomas Johansen]

Closing in on 48 hours stable since removing corefreq. Fingers crossed.