BTRFS partclone dropped in current Clonezilla in favor for DD partclone

[Shawn-Hub]

BTRFS partclone missing in current Clonezilla. DD is used instead for cloning BTRFS partitions. Will BTRFS partclone better support BTRFS partitions so it can be in later versions of Clonezilla. The block by block copy of DD for BTRFS pattitions is too large and too slow.

[sdgathman]

I used partclone.btrfs in partclone-0.3.37 (EL9). It had no issues - but does not copy redundant extents. This means the restored filesystem should be mounted and scrubbed (with btrfs scrub) soon after restoring. This should be noted in the docs. I didn't see a separate man page for partclone.btrfs, so not sure where to add it for a PR.

In the use mentioned, 9996 extents were corrected. A subsequent scrub had 0 errors. So that was 9996 extents not taking up space in the backup. A good design decision overall.

[sdgathman]

Zygo on #btrfs IRC points out that while not copying dup pages might be good design, partclone should track where they go and restore them. He considers this a bug. You don't need a full btrfs scrub implementation to restore dup pages.

Should I open a new bug for this?

[Thomas-Tsai]

Thank you for the bug report and the accompanying details.
We were able to reproduce the issue and have identified the correct path forward to resolve it.

The Btrfs filesystem restored with Partclone reports errors during a scrub, for example:

Tue Oct 14 13:45:16 2025] BTRFS info (device vdd1): scrub: started on devid 1                 
[Tue Oct 14 13:45:16 2025] BTRFS warning (device vdd1): tree block 22183936 mirror 2 has bad bytenr, has 0 want 22183936                                                                      
[Tue Oct 14 13:45:16 2025] BTRFS error (device vdd1): fixed up error at logical 22151168 on dev /dev/vdd1 physical 30539776                                                                   
[Tue Oct 14 13:45:16 2025] BTRFS error (device vdd1): fixed up error at logical 22151168 on dev /dev/vdd1 physical 30539776                                                                   
[Tue Oct 14 13:45:16 2025] BTRFS error (device vdd1): fixed up error at logical 22151168 on dev /dev/vdd1 physical 30539776                                                                   
[Tue Oct 14 13:45:16 2025] BTRFS error (device vdd1): fixed up error at logical 22151168 on dev /dev/vdd1 physical 30539776                                                                   
[Tue Oct 14 13:45:28 2025] BTRFS info (device vdd1): scrub: finished on devid 1 with status: 0                                                                                                

After submitting the change set 8254065, we have applied an update to the DUP component.

[Tue Oct 14 15:58:03 2025] BTRFS info (device vdf1): first mount of filesystem 0905c49c-321a-4bf9-9c65-be5a1e95d88e
[Tue Oct 14 15:58:03 2025] BTRFS info (device vdf1): using crc32c (crc32c-intel) checksum algorithm
[Tue Oct 14 15:58:03 2025] BTRFS info (device vdf1): using free-space-tree
[Tue Oct 14 15:58:17 2025] BTRFS info (device vdf1): scrub: started on devid 1
[Tue Oct 14 16:00:46 2025] BTRFS info (device vdf1): scrub: finished on devid 1 with status: 0

The update eliminates the error messages shown above.

Please let us know if you encounter any further issues.