Hi everyone,

today I moved my Unraid server to a new location.

Before the move, I shut the system down cleanly and without any errors.

About 6 hours later, I powered it back on, but now I can no longer start the array.

System setup:

• Array disks: 2× USB drives (used for years without issues)

• File system: EXT4 encrypted

• Cache: 2× NVMe in ZFS mirror, encrypted

• Everything was working perfectly before the shutdown and move

Current situation:

• Unraid refuses to start the array and shows the error:
  “Unmountable: wrong or no file system” (see screenshot)

• The ZFS cache pool works fine:

  • I can manually import the pool using zpool import cache

  • I can access all data without any issues

• The EXT4 encrypted array disks also seem intact:

  • I can manually mount them

  • The data is accessible

So it looks like the file systems themselves are not damaged, but Unraid no longer recognizes them correctly when starting the array.

What I’m trying to understand:

• Why Unraid suddenly thinks the array disks have no or the wrong file system

• Whether this could be related to the move, USB device enumeration, encryption, or disk assignment

• What the safest way is to get the array running again without risking data loss

Any help or guidance would be greatly appreciated.
Thanks in advance!

earth-diagnostics.zip

Dec 14 18:21:08 earth kernel: EXT4-fs (dm-2): bad geometry: block count 976750080 exceeds size of device (488373220 blocks)

Dec 14 18:21:08 earth kernel: EXT4-fs (dm-3): bad geometry: block count 244186293 exceeds size of device (122091222 blocks)

These suggest we are missing some info, this has nothing to do with a server move; it could be related to shutting down. Post the output from

fdisk -l /dev/md1p1

fdisk -l /dev/md2p1

Also post the output from zpool import

Thanks for your reply!

root@earth:~# fdisk -l /dev/md1p1

Disk /dev/md1p1: 1.82 TiB, 2000393486336 bytes, 3907018528 sectors

Units: sectors of 1 * 512 = 512 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 512 bytes

root@earth:~# fdisk -l /dev/md2p1

Disk /dev/md2p1: 465.76 GiB, 500102422528 bytes, 976762544 sectors

Units: sectors of 1 * 512 = 512 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 512 bytes

The partitions are approximately half the device capacity. Any idea how that happened? Also, how are you manually importing the array disks?

And please

4 hours ago, JorgeB said:

post the output from zpool import

3 hours ago, JorgeB said:

The partitions are approximately half the device capacity. Any idea how that happened? Also, how are you manually importing the array disks?

And please

I dont know how that happened. I just hit shutdown from the ui and waited until the power led turned off.

Thats how i can mount the zfs:

root@earth:~# zpool status

no pools available

root@earth:~# zpool import

pool: cache

id: 4300929232344889756

state: ONLINE

action: The pool can be imported using its name or numeric identifier.

config:

cache ONLINE

mirror-0 ONLINE

nvme1n1p1 ONLINE

nvme0n1p1 ONLINE

root@earth:~# zpool import cache

root@earth:~# zpool status

pool: cache

state: ONLINE

config:

NAME STATE READ WRITE CKSUM

cache ONLINE 0 0 0

mirror-0 ONLINE 0 0 0

nvme1n1p1 ONLINE 0 0 0

nvme0n1p1 ONLINE 0 0 0

errors: No known data errors

root@earth:~# zfs list

NAME USED AVAIL REFER MOUNTPOINT

cache 66.5G 384G 136K /mnt/cache

cache/appdata 4.31G 384G 4.31G /mnt/cache/appdata

cache/domains 6.16G 384G 136K /mnt/cache/domains

cache/domains/k3s-master 158M 384G 158M /mnt/cache/domains/k3s-master

cache/domains/k3s-worker-1 158M 384G 158M /mnt/cache/domains/k3s-worker-1

cache/domains/k3s-worker-2 158M 384G 158M /mnt/cache/domains/k3s-worker-2

cache/domains/talos-cp-1 2.03G 384G 2.03G /mnt/cache/domains/talos-cp-1

cache/domains/talos-w-1 1.78G 384G 1.78G /mnt/cache/domains/talos-w-1

cache/domains/talos-w-2 1.89G 384G 1.89G /mnt/cache/domains/talos-w-2

cache/isos 1.17G 384G 1.17G /mnt/cache/isos

cache/kubernetes 124K 384G 124K /mnt/cache/kubernetes

cache/media 44.7G 384G 44.7G /mnt/cache/media

cache/system 10.1G 384G 10.1G /mnt/cache/system

Edited December 15, 2025Dec 15 by StratoGh0st

5 minutes ago, StratoGh0st said:

Thats how i can mount the zfs:

I was referring to the array disks, you mentioned that you could manually mount them.

For the pool try this:

Make sure it's exported first zpool export cache

then on main click on the first device for that pool and then "remove pool"

back on main, create a new pool with the same name and number of slots

assign the pool devices, leave the filesystem set to auto

start the array to import the pool and post new diags (it's expected that the array disk will still fail to mount)

2 hours ago, JorgeB said:

I was referring to the array disks, you mentioned that you could manually mount them.

For the pool try this:

Make sure it's exported first zpool export cache

then on main click on the first device for that pool and then "remove pool"

back on main, create a new pool with the same name and number of slots

assign the pool devices, leave the filesystem set to auto

start the array to import the pool and post new diags (it's expected that the array disk will still fail to mount)

The workaround for the cache worked!

The behavior of the array disks is strange. Yesterday i could mount them in unraid (cryptsetup luksOpen, etc...), now (i just shut the pc down yesterday) the luksOpen works but the ext4 is not accessible.

I connected on of the drives to my mac and tried to mount it, but with the same result. I can understand, that one of these USB drives fails, but two at the same time?

I will try to repair one drive or at least get the data out of it. The two USB drives has only backup data on it, not critical but strange.

Big thanks for the help to get the cache back online!

It's strange how both partitions shrank. Also, Unraid should complain about disks "being wrong", when that happened, i.e., it would detect the size change.

You can try this; it may work, and it won't make things any worse.

Stop the array

type

sgdisk -o -a 8 -n 1:1M:0 /dev/sdX

Replace X with the correct identifiers for disks 1 and 2, they were /dev/sdb and /dev/sdc, but confirm, they can change with a reboot

After that reboot and the next boot, Unraid should complain that both disks are wrong. Go to Tools and to a New Config: Tools - New config - Keep all - apply

Back on main, start the array and post new diags.