Okay, I'm back. and GOOD NEWS EVERYONE! I fixed the random hard crashes that I've been having! So it seems that unraid didn't like the "mismatched" RAM in my system. (it was the same model number and everything, same timings, the works, but 2 of my 4 sticks had different pcb layouts.) That being said, I'm not out of the woods yet. I had a good run of like 10 days in a row with no crashes, and I'm back to square one before all this started.
So now i am back to the issue where rarely i get home, hit refresh on my webui and get met with a 500 internal server error. But, now I have proper logging setup, and was able to capture the error in a syslog. I think this shows what's going on. (Hopefully)
Would love if you guys could take a look. It does look to me like an nvidia issue, hence why I posted here.
I'm no pro on this, but that kernel dump to me looks like a thermal failure. Your card (or motherboard?) overheated and got a a bad memory call. The nvidia-smi called by your dashboard is what crashed, and then nginx died trying to query it.
[Plugin] Nvidia-Driver
in Plugin Support
Posted
I'm no pro on this, but that kernel dump to me looks like a thermal failure. Your card (or motherboard?) overheated and got a a bad memory call. The nvidia-smi called by your dashboard is what crashed, and then nginx died trying to query it.