Reverting "zfs destroy" - accidentally created dataset with existing name

[jm_]

Hi, yesterday I installed a fresh copy of proxmox. I imported one of my zpools from my previous install just fine. I then added a new ZFS storage on top of it via the GUI - trying to recreate my directory structure as it was before, I used the same name as on the previous install. Then I restored a VM that used this ZFS storage.

I'm pretty sure this essentially destroyed the old dataset and created a new one of the same name, as all of the data on the ZFS storage is gone.

The output of zpool history -il shows that a zfs destroy was called on the volume and shows the appropriate txg. Is it possible to somehow revert to that txg? This volume was never snapshotted as it contained unimportant data, but I would like to try to get it back anyway. Any help is appreciated

 

[jm_]

I've found a txg from before the destroy took place and tried to import the pool again, but this is the result:

# zpool import -N -o readonly=on -F -T 8947632 tank cannot import 'tank': one or more devices is currently unavailable

The pool and all its drives are ONLINE when calling zpool import. Any advice?

 

[Dunuin]

To import a pool you need to export it first (zpool export YourPool).

 

[jm_]

I have done that (otherwise it wouldn't show up in zpool import I think).

 

[jm_]

any ideas?