As the title mentions I'm looking for advice in re-creating a zfs poo called data the current pool is corrupt and needs to be rebuilt, but I'm hoping that I don't need to re-build all shares/vm/docker containers in Unraid.

I've also created another local pool to backup all of the data. This pool is called databack and so once I re-create the data pool I'll just need to move it over. Currently the Unraid UI sees these drives as unassigned since I did it all locally though the terminal. As I mentioned I'm hoping that I can re-create the data zfs pool without impacting unraid shares/vms/docker containers.

I've read that you can create the pool with the CLI and then export it before adding it with the GUI, but again I'm not actually sure if this is correct or the right way to go about it.

Just some additional context did run a memtest for 3+ passes and it didn't return any errors so I do no think it was a memory issue with the ZFS pool.

Thanks

tarantula-diagnostics-20250523-1748.zip

1 hour ago, cparman said:

but I'm hoping that I don't need to re-build all shares/vm/docker containers in Unraid.

Assuming those are on the current bad pool and no backups, it will depend on what you can recover, disable the services and see what you can copy, you can also check the output from zpool status -v, it will show the corrupts files, those will fail to copy.

I currently have the pool mounted as read-only and copied all of the data off already.

What i did was change the setting to auto-mount

Imported the pool as read-only zpool import -o readonly=on data

Created a new pool for backups zpool create -n -m /mnt/databackup databackup raidz sdk sdl sdm sdp sdq

and then copied everything over to databackup and oh man those small files for PC backups took forever.

NAME         SIZE  ALLOC   FREE  CKPOINT  EXPANDSZ   FRAG    CAP  DEDUP    HEALTH  ALTROOT
data         131T      0   131T        -         -     0%     0%  1.00x    ONLINE  -
databackup  90.9T  71.2T  19.7T        -         -     0%    78%  1.00x    ONLINE  -

There's a few things from Backups that are still getting copied, but overall its looking good.

The next step will be deleting data zfs pool and re-creating it, but my confidence in this step so that way unraid will be back to normal is at an all time low.

Within the setting i've also disabled docker and VMS for now since the pool isn't mounted anyways.

Apricate your advice.

Edited May 27, 20251 yr by cparman

One other thing that i want to make sure this doesn't impact the databackup pool that i created since unraid doesn't even see it as existing, just shows those drives as unassigned. Which is fine if they're still unassigned when data is re-created.