I have two 5TB USB drives attached to my unRAID server, formatted in BTRFS and lashed together as a single 10TB drive, used as a tertiary backup to my media library. Space was starting to get a little tight so I connected an unused 2TB drive, formatted it, and added it to the pool. Well, I screwed something up and ended up removing the partition from the 2TB drive. Now the entire pool won't mount. I realized my mistake and know what to do now, but I can't get back to the starting point (unless, I guess, I scrap the entire pool and start over from scratch again). I'm pretty sure the problem is that there's now a big 2TB unformatted chunk of space in the pool.

Below is the error in the log. The mount point for the pool is /mnt/disks/PlexBackup3. It currently doesn't exist and any time I manually create it, it gets deleted when I attempt to mount the pool. The id of the missing disk is devid 3.

May 7 21:36:20 unRAID unassigned.devices: Mounting partition 'sdd1' at mountpoint '/mnt/disks/PlexBackup3'...

May 7 21:36:20 unRAID unassigned.devices: Mount cmd: /sbin/mount -t 'btrfs' -o rw,relatime,space_cache=v2 '/dev/sdd1' '/mnt/disks/PlexBackup3'

May 7 21:36:20 unRAID kernel: BTRFS info (device sdd1): first mount of filesystem c545cf29-eba2-4f8b-a185-63522a08304d

May 7 21:36:20 unRAID kernel: BTRFS info (device sdd1): using crc32c (crc32c-intel) checksum algorithm

May 7 21:36:20 unRAID kernel: BTRFS info (device sdd1): using free space tree

May 7 21:36:20 unRAID kernel: BTRFS error (device sdd1): devid 3 uuid 45e0b911-317c-4954-9de4-fd44d79c8e5f is missing

May 7 21:36:20 unRAID kernel: BTRFS error (device sdd1): failed to read chunk tree: -2

May 7 21:36:20 unRAID kernel: BTRFS warning (device sdd1): folio private not zero on folio 1107125698560

May 7 21:36:20 unRAID kernel: BTRFS warning (device sdd1): folio private not zero on folio 1107125702656

May 7 21:36:20 unRAID kernel: BTRFS warning (device sdd1): folio private not zero on folio 1107125706752

May 7 21:36:20 unRAID kernel: BTRFS warning (device sdd1): folio private not zero on folio 1107125710848

May 7 21:36:20 unRAID kernel: BTRFS warning (device sdd1): folio private not zero on folio 1107139280896

May 7 21:36:20 unRAID kernel: BTRFS warning (device sdd1): folio private not zero on folio 1107139284992

May 7 21:36:20 unRAID kernel: BTRFS warning (device sdd1): folio private not zero on folio 1107139289088

May 7 21:36:20 unRAID kernel: BTRFS warning (device sdd1): folio private not zero on folio 1107139293184

May 7 21:36:20 unRAID kernel: BTRFS error (device sdd1): open_ctree failed

May 7 21:36:21 unRAID unassigned.devices: Mount of 'sdd1' failed: 'mount: /mnt/disks/PlexBackup3: mount system call failed: No such file or directory. dmesg(1) may have more information after failed mount system call.'

Any ideas on how to pry this missing disk out of the BTRFS config? I spend some time googling for help but almost everything I found assumed that the filesystem can be mounted.

Post the output from btrfs fi show

root@unRAID:~# btrfs fi show

Label: none uuid: 6effd5dd-ac37-4c79-8b7e-1b092477d839

Total devices 4 FS bytes used 4.56GiB

devid 1 size 465.76GiB used 64.00GiB path /dev/nvme0n1p1

devid 2 size 1.86TiB used 177.00GiB path /dev/sdj1

devid 3 size 465.76GiB used 54.01GiB path /dev/sde1

devid 4 size 465.76GiB used 63.01GiB path /dev/sdf1

Label: none uuid: 3bb64198-8832-41c3-aa32-210bca89b80b

Total devices 1 FS bytes used 8.81GiB

devid 1 size 200.00GiB used 17.02GiB path /dev/loop2

Label: none uuid: 36899d09-a8d8-43c1-be01-7f91d98ebd4b

Total devices 1 FS bytes used 144.00KiB

devid 1 size 1.82TiB used 2.07GiB path /dev/sdk1

Label: 'PlexBackup3' uuid: c545cf29-eba2-4f8b-a185-63522a08304d

Total devices 3 FS bytes used 8.31TiB

devid 1 size 4.55TiB used 4.20TiB path /dev/sdd1

devid 2 size 4.55TiB used 4.20TiB path /dev/sdb1

*** Some devices missing

The /dev/sdk1 is the device that is missing. Thanks for taking a look at this, JorgeB

UD will not mount a btrsf pool with a missing device; Unraid could, but only if the pool were redundant, which appears to not be the case.

I see that the 2Tb disk is showing as mounted with a different filesystem, did you reformat it after deleting the partition?

9 hours ago, JorgeB said:

UD will not mount a btrsf pool with a missing device; Unraid could, but only if the pool were redundant, which appears to not be the case.

I see that the 2Tb disk is showing as mounted with a different filesystem, did you reformat it after deleting the partition?

Yes, deleting the partition was part of the screw-up I made. I thought reformatting might fix it.

<morganfreeman>It didn't </morganfreeman>

If you hadn't reformatted the partition, it could likely be recovered; after reformatting it's not possible.

You can try to see if the pool mounts read-only with the remaining devices if there's any data you want to try and recover, but then the pool will need to be recreated.

If you wish to try

mkdir /temp
mount -o degraded,rescue=all,ro /dev/sdd1 /temp

Kinda figured. Loosing the pool isn't a big deal; I already have a backup copy of my library stored elsewhere, so the only thing I'd be losing is the 36-48 hours it'll take to copy the library back over.

Thanks JorgeB 👍