Hi all 

I am on 7.0.0 with critical updates patch installed. I have moved to new hardware (new Mobo, ram and CPU) a few weeks ago.

 

I have been suffering form almost daily random unclean shutdowns with parity checks initiated after. Only started happening recently, after updating to 7.0.0.  I don't think the PC actually shuts down. Just a notification that states unclean shutdown and then I notice a parity check running. 

 

Any help would be much appreciated. thank you 

 

beast-syslog-20250310-1414.zipbeast-diagnostics-20250310-1714.zip

 

 

 

 

 

 

Edited March 10, 20251 yr by homiek33

1 hour ago, homiek33 said:

Just a notification that states unclean shutdown and then I notice a parity check running. 

That suggests the server is rebooting by itself, check the uptime, but if that is the case, it almost always means a hardware issue, or bad power.

Yes indeed the uptime restarts after such unclean shutdowns. 
 

My UPS needed a battery change so just switched to a new one as per your response to my previous thread (although not sure that alone would cause this problem).

 

Is there best practice on what/how to test the relevant hardware? 

Edited March 10, 20251 yr by homiek33

PSU would be a good suspect, but could also be RAM, board, CPU, etc.

I have updated to 7.0.1, disabled NUT (checking the logs, there was some incompatibility issue with it) and replaced my UPS. So far, it is been almost two days without issue. Will report if any issues but so far, looks to be stable. thanks 

Just happened again. Sigh... My PSU may not be beefy enough? (running i5-9600K using a 450w PSU, 4 HDDs, one NVME a few fans and GTX1660TI Super which is always idle). Based on all calculations I need close to 420W or thereabouts so I was confident I would be okay. Maybe Will update the PSU and see what happens. 

Edited March 23, 20251 yr by homiek33

I have now run a CPU test on cinebench, kept an eye on temp (sub 53c CPU) and ran memtest for 17 passes with no errors. Motherboard seems fine, no error lights or visible damage anywhere.
 

Is there anything I can do to diagnose this further ? I’m not convinced power supply is an issue given system ran under stress with no issues.  
 

any tips would be helpful. Happy to repost for logs for this latest unclean shutdown if helpful. 
 

thanks! 

Basically, and without anything logged, you would need to start swapping around some parts.

okay. step 1 will remove my GPU and see if it stops. After that, will swap PSU. if that doesn't resolve, I am nuking the whole thing and getting  some new hardware  

 

Thanks again for your your help. 

@JorgeB Could the hardware problem be the flash drive running the OS? 

41 minutes ago, homiek33 said:

Could the hardware problem be the flash drive running the OS? 

The OS doesn't run from the flash drive.

 

The Unraid OS is contained in archives on flash. Those archives are unpacked fresh at each boot, into RAM, and the OS runs completely in RAM. Think of it as firmware.

sigh. Just had another unclean shutdown randomly (a day after upgrading PSU). I am posting my diagnostics and logs from syslog server to check one last time if there are any other leads I should follow to further resolve.

 

If still likely a hardware issue, then I will just have to move on and get a new mobo cpu ram setup. Can't spend more time diagnosing this further unfortunately - have a whole village depending on this server.  

 

Like I said, I had run memtest for 17 passes (more than 24 hrs) and Cinebench stress test (granted it was only 10sh minutes) with no issues and have now upgraded PSU to no avail. 

 

Many thanks 

 

syslog2.txt beast-diagnostics-20250407-0008.zip

Some Nvidia related errors logged, but they were a few days ago, still may be worth retesting without the GPU, or the driver.

Okay will try and revert. Thanks for the suggestion as always. 

Removing GPU did not work. Unclean shutdown same day. As a final ditch effort, will run in safe mode. 

Safe mode (with dockers enabled) hasn't worked. I will change my mobo CPU ram and chalk it up to hardware (likely mobo) issue. Thanks all for the input.

I reverted to my old mobo cpu ram (3rd gen intel) and PSU and haven't had an unclean shutdown. I am pretty sure it was the new mobo but can't be 100% sure. In any case, this was such a big time suck and I appreciate the help. 

I had a similar problem with my server and after much effort, I replace the MB, RAM and Processor.  In addition, I replaced the USB thumb drive after noticing multiple corrupt files.  Since reactivation, I have resolved the issue.