drkduncan

Thought I would post an update, as I have officially fixed it. Turns out it was the power monitor I had the server plugged into. It would occasionally under-volt just enough to reboot the server. So it was as you suspected a hardware issue, I just didn't expect the hardware to be outside the server.

Thanks for the advice. I think I have a backup PSU I can try. If that doesn't fix it I try the Ram.

I have been dealing with a problematic server. I have 3 of the dreaded Samsung 990 pro's that were previously dropping out of the cache array, but those issues seem to be solved with the latest firmware (6B2QJXD7). Recently (for the last week or two) I have been dealing with nightly reboots, and I cant seem to find any trace of why. Nothing that I can find in the syslog-previous, and I even setup a remote syslog server and no errors are reported there either. Hopefully one of you fine folks can help provide some insight! demonic-diagnostics-20250802-1350.zip

Unfortunately I could not access /boot but I was able to recover it via plugging it into another pc and running chkdsk. I had to reboot twice to get everyhting back in working order, but I am back up and running (and backup has been taken - lesson learned). I think the source of the problem was probably upgrading, then forgetting to reboot, then upgrading again a couple weeks later. Anyways - all is well and @JorgeB - thanks for your reply!

I have a somewhat unusual situation in that my system is still currently up and running, but has a corupted flash drive. I have attached the diag here. Is there anything I can do to backup my config while it is currently still running so that I dont have to rebuild from scratch when I reboot? demonic-diagnostics-20221208-1007.zip