We had a power outage today which resulted in my Unraid server losing power and shutting down unexpectedly. I have had trouble booting with my Intel ARC GPU for a while so I also upgraded the BIOS since the server was down anyway.

I don't think the Intel ARC GPU is causing these issues, but I will troubleshoot that when possible. In the meantime, the server appears to boot consistently now but I can't get the web server to start. There is nothing listening on port 443 but the network is up and I can ping things on the internet with both IP addresses and DNS hostnames.

Starting in safemode or GUI mode does not help. When starting GUI mode the local GUI just sits on a HTTP error page. I have a diagnostics log from the boot after an unclean shutdown and from a manual collection by typing diagnostics in the terminal.

Currently running 7.0.1.

tower-diagnostics-20260216-0054.zip tower-diagnostics-20260216-0029.zip

Edited February 16Feb 16 by soja

There's a problem with the pool cfg file, delete /config/pools/cache.cfg from the flash drive, then reboot, don't start the array yet since the pool devices will be unassigned, and post the output from

btrfs fi show

11 hours ago, JorgeB said:

There's a problem with the pool cfg file, delete /config/pools/cache.cfg from the flash drive, then reboot, don't start the array yet since the pool devices will be unassigned, and post the output from

btrfs fi show

Thank you! The web server is listening on 127.0.0.1 at least now so I can't access it remotely yet but that's progress.

Here is the output of the command:
root@CUDA:~# btrfs fi show

Label: none uuid: 06fb067d-9420-4958-b107-d95fa340d514

Total devices 5 FS bytes used 138.46GiB

devid 1 size 372.61GiB used 258.03GiB path /dev/sdd1

devid 2 size 894.25GiB used 738.03GiB path /dev/sdo1

devid 3 size 372.61GiB used 258.00GiB path /dev/sdk1

devid 4 size 372.61GiB used 258.00GiB path /dev/sdn1

*** Some devices missing

I was trying to remove a 120GB drive a while ago from the cache and was having trouble which I think is the missing device.

You will need GUI access, then you can try reimporting the pool degraded, but you will then need to use the CLI to correct it:

on main, create a new pool with the same name and 4 slots

assign the pool device(s), leave the filesystem set to auto

start the array to import the pool then post new diags