Hello everyone, a few days ago I decided to reinstall Unraid from scratch to give the server a general clean-up and change some configurations.

I downloaded the USB creator , followed the entire startup procedure, logged into the server remotely, reloaded the license, and formatted the disks( even though the server retained 2-3 folders from the old installation,and I don't understand why, since I deleted all the data from the disks; maybe I accidentally put the same disks back into the same slots), I started the parity check.

And that's when the problems started.
Random reboots at different percentages of the parity check, sometimes after 30 minutes, sometimes after hours of work, and I can't figure it out.

The server has not undergone any hardware changes, and I checked everything I could find posted on the forum, such as c-state, BIOS upgrade, XMP (even after updating the BIOS, if I activate it, the server won't even start, yet it worked before formatting), but nothing seems to solve the problem; The only thing I can add is that it would seem, or at least that's how it appeared to me, that leaving the server on in Maintenance mode keeps it stable (I tried it for about 12 hours).

Is the USB stick dying? Has the installation somehow become corrupted? Or is there another problem that I'm not seeing?

I'm attaching the log files, I hope someone can make sense of them....

thevault-diagnostics-20260208-2241.zip

Edited February 9Feb 9 by forbi

51 minutes ago, forbi said:

Maintenance mode

Can't tell anything about your filesystems when Diagnostics are taken like that.

Setup Syslog Server.

Why have you assigned Parity2 instead of Parity? And skipped disk1?

I should have taken the logs while the server was doing the parity check, not in maintenance mode. Did I get confused and download the ones in maintenance mode?

I left slot 1 free for parity because my plan was, once I had rebuilt the parity and copied the 12 terabytes of files I had saved to another 18TB disk (which I connected as Unassigned Devices), to add that other disk to the parity and use the one I have now as storage, once the parity had been rebuilt again.

Unfortunately, this is the only way I can avoid losing all the data I had before.

Normal syslog starts over after every boot, like mentioned, enable the syslog server, but note that a server rebooting by itself is almost always a hardware problem.

I don't know why, but after letting the server do the parity check this morning, convinced that I would come home from work and find it restarted so I could send a new log, it is at 92% of the check and continues to run.

Maybe the last RAM reseat, done almost for good luck, really did something...

I'm marking the topic as resolved at this point, even though I feel almost stupid for wasting two weeks on it and then solving it with something so trivial :D