I built my first unRAID system about six weeks ago, and I've been having issues with it becoming unresponsive, sometimes multiple times per day.  I know when it happens because I get the notification that my Nextcloud can't connect.  I try to go to the web GUI and there's no response.  I've tried pinging it from my computer and get no response.  When I go to the machine, it is still powered on but there's just a blank screen on the video output.  I have to do a hard shutdown and reboot, which I know is not ideal but it never responds otherwise.

 

I've not found any obvious or consistent cause, but I'm new at this so I'm hoping someone else may be able to see a problem.   I turned on the system log output to the cache drive last month, so I'm attaching the entire cumulative log along with the diagnostics file.  

 

This is frustrating because I actually built two systems, and this is the one I bought "better" parts for but it's the only one that I'm having trouble with.   System info:  i7 14700k, 128GB DDR5 RAM, two 1TB NVME cache drives.  I'm running roughly 15 docker containers (Plex, Nextcloud, the Radarr/Sonarr type apps, etc.), but I haven't set any VMs up yet because the stability is not great.  At idle, the CPU is running 2-4% and the RAM is only at 4% so it's not being stressed at all.  (The other machine I built is using an Aliexpress motherboard/CPU combo and it is rock solid, staying on for days and days without lock ups.  Go figure.)

 

I ran the memtest86 overnight on my main machine and it passed.  I'm reading that maybe I should run that longer to catch any RAM issues.  I've installed Fix Common Problems plugin and it didn't find any errors,  but I wanted to see if there's anything obvious in the log/diagnostic file.  I'd appreciate if anyone could take a glance at these files to see what's going on.

tower-diagnostics-20240330-1828.zip unraid_log.txt

There are some call traces but difficult to say if they are hardware or software related, one thing you can try is to boot the server in safe mode with all docker containers/VMs disabled, let it run as a basic NAS for a few days, if it still crashes it's likely a hardware problem, if it doesn't start turning on the other services one by one. 

I did what you recommended and disabled all docker containers, and so far I'm nearing two days of uptime.  This is definitely an improvement, but I will let it run it a little longer to test it.  I'm happy this is starting to suggest it is a docker container and not hardware problem, but I know that will take more testing to confirm.  I will update this post when I discover the culprit causing my problems.  Thanks for the reply.

I think I figured my problem, finally.  I will post my solution in case someone else has similar issues. 

 

I KNOW I setup the Plex container to use the Intel iGPU when I installed it (100% did it because it intially worked without issues for weeks), but somehow the settings were cleared in the advanced docker settings and it was doing full CPU transcoding, which was throwing errors in the log with the x264 encoder (new errors that weren't in the original log I posted). I was also getting Plex Transcoder segfault log messages.  Thinking that it might've been bogging down the CPU, I began by limiting the CPU cores in the docker setting to four cores and it stopped locking up.  But then I discovered the iGPU wasn't being used at all for transcoding!!   I followed this guide and re-enabled the iGPU and it seems to have fixed my problem.

 

Thanks for looking at my log/diagnostic and helping me.  I'm glad it ended up being something that I inadvertently caused and my hardware is okay.

 

Edited April 4, 20242 yr by sahuff2