Hey folks, new Unraid user here. I've attached three diagnostics files:

• One upon discovering shares were missing.

• The second after trying to stop the array (which hung).

• The third after running powerdown -r, which also hung.

The system I built myself. I've run memtest as well as stress-ng when booted to a live USB, neither uncovered anything interesting. I'm open to suggestions on other ways to rule out hardware issues.

Running Unraid 7.2.0, I believe I saw the issue on the previous version I was running as well, but I don't remember what that version was.

Every so often (between a few days to in this case about two weeks of uptime) the shares will all disappear. Screenshot attached. At this point my frigate container crashes (because it's trying to write video to disk). dmesg shows an oops, general protection fault, stack trace, etc. I believe something similar is also printed to the console, but I don't have my monitor hooked up right now.

At this point I try to stop the array, which hangs forever (UI says "Array Stopping•Retry unmounting disk share(s)" at the bottom). Shutdown from the UI or via powerdown -r also hangs indefinitely. As far as I can tell the only way to reboot the system in this state is to hold the power button.

What are my next steps for debugging this? As I mentioned I'm new to Unraid, but fairly Linux savvy (though it hasn't been my day job for a decade or so). Happy to try something out.

Thanks!

nas-diagnostics-20260115-1825.zip nas-diagnostics-20260115-1831.zip nas-diagnostics-20260115-1850.zip

Edited January 16Jan 16 by jplona
Added version number

Jan 6 02:00:12 NAS kernel: XFS (md2p1): Metadata corruption detected at xfs_dinode_verify+0xa4/0x6f0, inode 0xa45bcc36 dinode

Jan 6 02:00:12 NAS kernel: XFS (md2p1): Unmount and run xfs_repair

Check filesystem on disk2

Followed the instructions for disk2 (and disk1 and disk3 as well for good measure). The check ends with "No file system corruption detected" (screenshot attached).

For what it's worth, I don't remember seeing that message in the logs when my issue has occurred in the past.

It does seem strange to see that message and not have anything show up in the check, or somewhere in the UI for that matter.

Reboot and post new diags after array start.

Two diagnostics attached:

• One taken before the requested reboot. This was the (hard reset) boot required after the failure in the original post. It should have logs of the check on disk1/2/3 I ran in the previous comment (assuming that operation writes to logs).

• One after rebooting normally and starting the array. I scanned dmesg and syslog and didn't see anything interesting, but I don't really know what I'm looking for.

nas-diagnostics-20260116-0706.zip nas-diagnostics-20260116-0816.zip

Looks good for now; if it happens again, post new diags before rebooting.

Thanks for the quick responses, I'll report back here next time it happens.