Data Recovery - Deleted Folders/Files from Volume Group

daNutz

Member
Mar 24, 2023
22
0
6
Hi,

Long story short, ive had a miserable couple of days where my primary data source (ZFS) failed and im trying to recover on it, in the meantime my data replication stupidly sync'd (rclone) i thought i had disabled it... and deleted the data on the replication node. I have no recovery points on either system and while i try to recover on the primary, i was wondering if any of you intelligent people have any idea's on how i could recover the data on the backup?

The data i need to recover is on a volume group disk (/dev/sdb1) that is attached to a Debian VM in PVE (6.8.12-8-pve). the Debian OS sits on another drive/vg and the drive (/dev/sdb1) i need to recover from simply holds data.

The rclone sync job has basically marked all files for deletion on the /dev/sdb1, thus unable to see them anymore. nothing has been written on the drive since that point and currently it sits unmounted.

I have tried with extundelete but it gives the following error when running either of the commands:

#extundelete --restore-all --after '2025-02-12' /dev/sdb1
Only show and process deleted entries if they are deleted on or after 2025 and before 9223372036854775807.
NOTICE: Extended attributes are not restored.
Loading filesystem metadata ... 450560 groups loaded.
extundelete: Numerical argument out of domain while reading internal journal superblock
extundelete: Numerical argument out of domain while reading journal superblock.
extundelete: Numerical argument out of domain when trying to examine filesystem

#extundelete --restore-all /dev/sdb1
NOTICE: Extended attributes are not restored.
Loading filesystem metadata ... 450560 groups loaded.
extundelete: Numerical argument out of domain while reading internal journal superblock
extundelete: Numerical argument out of domain while reading journal superblock.
extundelete: Numerical argument out of domain when trying to examine filesystem

Ive checked the disk with fsck (fsck -f -n /dev/sdb1)) and e2fsck (e2fsck -c -n /dev/sdb1) and both say its ok:

Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
datapartition: 50974/922746880 files (1.7% non-contiguous), 66931970/14763949568 blocks

TIA
 
Last edited: