Add non empty storage pool

LXD
,
1

Hi,

After a crash of my machine, I had to reinstall the OS (Ubuntu 24.04 LTS Server). The storage pool for my LXD node was/is on a separate partition on the disk and is not affected.

I have no reinstalled everything, and try to add my existing pool.

The pool is a zfs pool named lxd. It is mounted (online) and I can see the contents with zfs list.

Trying to create the pool gives me:

lxc storage create zfs_lxd zfs source=lxd
Error: Provided ZFS pool (or dataset) isn't empty, run "sudo zfs list -r lxd" to see existing entries

I also tried lxd recover:

This LXD server currently has the following storage pools:
Would you like to recover another storage pool? (yes/no) [default=no]: yes
Name of the storage pool: lxd
Name of the storage backend (lvm, powerflex, zfs, btrfs, ceph, cephfs, pure, cephobject, dir): zfs
Source of the storage pool (block device, volume group, dataset, path, ... as applicable): /dev/sda3
Additional storage pool configuration property (KEY=VALUE, empty when done): 
Would you like to recover another storage pool? (yes/no) [default=no]: 
The recovery process will be scanning the following storage pools:
 - NEW: "lxd" (backend="zfs", source="/dev/sda3")
Would you like to continue with scanning for lost volumes? (yes/no) [default=yes]: 
Scanning for unknown volumes...
Error: Failed validation request: Failed checking volumes on pool "lxd": Instance "nerves-hub-device" in project "default" has pool name mismatch in its backup file ("default" doesn't match's pool's "lxd")

So recover did do something but it is not able to add the container instances due to a mismatch in the pool name it seems.

Is there a way to fix this? I really would hate to loose all these containers.

Thank you very much for your help!

2

Hi, this forum no longer supports LXD, as it is no longer a Linux Containers project. You should ask your question on LXD - Ubuntu Community Hub.

1 Like
3

I guess it also applys to lxd.

4

Thank you for the replies. I finally managed to get it working again. I reinstalled LXD, created a new storage pool with a zfs dataset that I created under the zpool root and with a different name. And once I had done that, the lxd recover started behaving differently and no longer complained about the name mismatch since I could reuse the old name (default) for the recovery.

I had to remove a couple of VMs with zfs destroy but at the end all containers were imported and to my surprise they all started and worked without a problem.

To finish I removed the new storage pool and edited the default profile to use the old pool, and now everything works as if nothing has happened.

Thanks again for the hints!