Not sure how to proceed. I've tried upgrades, downgrades, latest firmware, BIOS at default. Would really appreciate any insight that could be offered. Keeps happening with way too much frequency.

 

TAI,

 

-Rob

 

-diagnostics-20240704-1631.zipsyslog-192.168.50.100.log

Is the problem because a normal shutdown is not clean or is the server rebooting by itself?

It's rebooting itself. I built it last November and it would typically be stable for about a week. Now it's rare that it stays online for a day. Normally after a crash I can reach it, verify unclean shutdown, reboot, then bring the array back online. Now after a reboot I still have an unclean shutdown message. One item of note, during a parity check it seems to start at regular speed, ~200 Mbs but after an hour or two it will drop to ~100-200 kbps. Once it's dropped to that speed it will eventually crash.

Server rebooting by itself is almost always a hardware problem, if you have multiple sticks try using the server with just one, if the same try with a different one, that will basically rule out bad RAM.

JorgeB, Thanks so much for the suggestions. I think I have it so much in my head that I got great parts so hardware couldn't be the problem. I've taken a stick out and will monitor. Time will tell. 

 

Thanks for the focus,

 

-Rob

Ok, over the past few days I've removed one stick of RAM, let it run until it crashes (still appears random, most uptime so far has been 22 hours) swapped the other stick in until it crashes, and done the same procedure with the second slot.

 

It's still crashing. I don't have a cooling issue. Currently I'm running memtest. It's passed the first round of checks. I intend to let it run 24 hours.

 

What are my next steps if both sticks of RAM are good?

If it still crashes with either stick of RAM it basically rules it out, next suspects for me would be PSU, board, CPU.

Was afraid you'd say that. Not really sure how to troubleshoot those.

Was afraid you'd say that. Not really sure how to troubleshoot those.

 

In addition I've noticed all shutdowns are now unclean even if I stop the array and initiate a reboot.

2 hours ago, RobFlix said:

In addition I've noticed all shutdowns are now unclean even if I stop the array and initiate a reboot

This should not be the case if you have successfully stopped the array unless for some reason Unraid is unable to update the status on flash drive.

Ok, this is intriguing. Are you thinking it might be worthwhile to buy a new USB drive and transfer my license?

I'm starting to think this is actually a software issue. I've had 48 hours of straight up time which is a bit of a record. After the last crash I started only the Plex docker. It appears Some other container is causing the crash. I'm running the Arr suite, SAB, Flaresolver, and Cloudflare tunnel. I'm thinking I start one additional docker per day to see what causes the crash. Any thoughts on this process or what order to start them?

Container causing a crash is very possible, reboot would be unusual, but won't say it's impossible.

 

8 minutes ago, RobFlix said:

I'm thinking I start one additional docker per day to see what causes the crash.

I would start with the containers you miss the most.

Well, the mystery deepens. I started a container a day and waited for a crash. I'm now up to almost 8 days of uptime with all containers running. When its crashed in the past I usually do a manual reboot, start the array, then fire up all the containers, usually all or most at once. Could this be causing some sort of hiccup that results in the crashes I've been experiencing? 

8 hours ago, RobFlix said:

Could this be causing some sort of hiccup that results in the crashes I've been experiencing? 

Doesn't seem very likely, but it's a possibility