TLDR; I have loads of SQUASHFS error and I can't figure out what's wrong. Changed USB port, changed USB drive, updated unraid, updated bios, ran memtest for 13hours with no errors...

Potentially unrelated context?:

About a year ago I got myself an I9-14900k for my gaming PC, things didn't turn out quite as planned and I ended up with a 14900k collecting dust, so figured I could just get a motherboard for it and put in it my unraid server. Now I've had my unraid server for probably 10years or so running AMD with no issues apart from user errors. But as soon as I swapped to the I9 I started having issues with my VMs being very unstable, they'd freeze between an hour to a week so it was hard to figure out what was wrong, but isolating a p-core to each VM made it much better, like they'd freeze up maybe every two weeks or whatever, and I've had my unraid like this since then.

But now (past month or two) things gotten worse. Unraid webui is unresponsive, but the dockers and VM running fine, dumb as I am I didn't figure out it was just the webui so I had to reboot the server which always made me have to restart the docker container else no dockers were available. So every other day I restarted unraid and disabled/enabled dockers, and instead of taking my time figuring out what's wrong I obviously neglected it so it got worse. Now when I restart the unraid I need to restart the webui and even run '/usr/local/sbin/emhttp start' to get the webui online.

 

And now I've had it, time to figure out wtf is going on. After some searching it seems as the SQUASHFS errors are the most critical ones and could probably be the reason for emhttp failing to start? Either way, new flash drive was the answer I found. Got a new samsung flash drive, migrated and same errors. I swapped to a different USB port and same errors. Ran memtest for 13h which returned with 6passes and no errors. Updated unraid from 7.0.0 to 7.0.1, did not fix anything.

I have now disabled VMs and dockers and it looks like I don't have issues with emhttp not starting any longer, coincidence or not I can't tell.
Which other logs would you gurus need in order to help me figuring out what's wrong?

Edit:
When I get the server running, it'll run for a few days without issues. And when the webui dies I can restart it through terminal. But I'd prefer to have it like it used to be with my AMD system. Turn it on, and it can stay on for probably a year without any hickups.

syslog.txt

tower-diagnostics-20250412-1017.zip

Edited April 12, 20251 yr by Mokkisjeva

SQUASHFS errors as typically a flash drive problem, but can also be bad RAM, or since you have a 14900K, the Intel 13/14 gen issue, if you have already changed the flash drive, unlikely that is the problem, but since it's the easiest thing to replace, it may be worth to do it again, if issues persist, and since you have multiple RAM sticks, try using the server with just one pair, if the same try with the other one, that will basically rule out bad RAM, leaving the CPU and the main suspect if the problem continues.

I've cycled all ram sticks, same errors. Could the actual OS (files) on the flash drive be bad? When I migrated USB drive I didn't do a clean install, just install from the backup I did. So the problematic files would still be present? I'm clueless so just thinking out loud here...

8 minutes ago, Mokkisjeva said:

Could the actual OS (files) on the flash drive be bad?

They are checksummed on every boot, so it should not be possible, if it's not the flash drive or the RAM, it's possibly the Intel issue.

Oh my... now my cache pool reports: 'Unmountable: wrong or no file system'

Guess 40$ USB flash drive was not the cheap fix I was hoping for. Back to AMD, fingers crossed.