Rescue data from un-importable pool

[derekdai]

I have a single-disk raidz pool. When I just learned to use zfs many years ago, I mistakenly added a file to it as stripe vdev, then I realize that I couldn't remove it from the pool, then I panicked, delete it before data writing into the file, and then reboot system, as a result, the pool can no longer be imported.

This pool contains personal data that I don't want to lose, and finally saved as a disk image with dd. I would like to ask if there is any way to get the data out of it?

zpool can recognize the image as a zpool

# zpool import -F -X -d .
   pool: data-pool
     id: 3292954182846563831
  state: UNAVAIL
status: The pool was last accessed by another system.
 action: The pool cannot be imported due to damaged devices or data.
   see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-EY
 config:

        data-pool                      UNAVAIL  insufficient replicas
          personal-data.zfs  UNAVAIL

[rincebrain]

Try setting zfs_max_missing_tvds to 1 and doing zpool import -o readonly=on?

[derekdai]

Hi, I also googled your suggestion to learn about zfs_max_missing_tvds parameter, thanks for the information, but in my case, zpool still reports

cannot import 'data-pool': one or more devices is currently unavailable

My commands

# echo 3 >> /sys/module/zfs/parameters/zfs_max_missing_tvd
# cat /sys/module/zfs/parameters/zfs_max_missing_tvds
3
# zpool import -d . -o readonly=on -m -N -f -F -X data-pool
cannot import 'data-pool': one or more devices is currently unavailable

[derekdai]

# zdb -l personal-data.zfs 
------------------------------------
LABEL 0
------------------------------------
    version: 5000
    name: 'data-pool'
    state: 0
    txg: 84811
    pool_guid: 3292954182846563831
    errata: 0
    hostid: 16845321
    hostname: 'xxxx'
    top_guid: 9059809511536026621
    guid: 9059809511536026621
    vdev_children: 3
    vdev_tree:
        type: 'disk'
        id: 0
        guid: 9059809511536026621
        path: '/dev/sdb3'
        whole_disk: 0
        metaslab_array: 33
        metaslab_shift: 35
        ashift: 12
        asize: 3998781800448
        is_log: 0
        DTL: 93
        create_txg: 4
    features_for_read:
    labels = 0 1 
failed to unpack label 2
failed to unpack label 3

About these two lines

failed to unpack label 2
failed to unpack label 3

Does it means the pool actually lost two top level vdevs?

[rincebrain]

No.

vdevs have four labels, two near the front of the disk, two near the rear.

If it can't find labels 2 and 3, that means either the device was truncated, or otherwise is not the size ZFS expects it to be.

[derekdai]

Got it.

Is there any other tool I can try to recover the data?