Regression: -gmf (Generate checksum for files in device after saving) option hangs with 0% ETA progress

[Ken Pergola]

Hello Steven,

I noticed what appears to be a regression in the -gmf ("Generate checksum for files in device after saving") option.

Using the same machine and imaging the same destination drive:

alternative stable 20260913-resolute ✅ Works fine
alternative testing 20261025-tenacious ❌ Hangs
alternative testing 20261030-tenacious ❌ Hangs

In the attached screenshot, approximately 15 minutes had elapsed with the progress still showing 0% ETA, which is not the behavior I observed previously.

If I recall correctly, on this attempt I waited roughly 25 minutes before pressing Ctrl+C in an attempt to exit gracefully and obtain the clonezilla-img file.

Since the hardware, source, destination, and workflow were unchanged, this appears to be a regression introduced between the stable and recent testing releases.

Hopefully this is something that is easy to reproduce.

Best regards,

Ken

P.S. If it would be useful, I can also provide the output of dmesg, or any other logs from a failed run.

[Ken Pergola]

Late-breaking clues:
Yikes, this is not a systemic regression so it may be more difficult to reproduce.

So far, the issue outlined in this forum post only happens when imaging my Samsung 990 Pro 1TB NVMe SSD (when I use a build later than alternative stable 20260913-resolute).

nvme0n1 931.5G
nvme0n1p1 100M|vfat
nvme0n1p2 16M|No_file_system
nvme0n1p3 834.2G|ntfs
nvme0n1p4 4G|ntfs

With all else being equal, the issue cannot be reproduced when imaging using alternative testing 20261030-tenacious on my:
1) Transcend M.2 SSD 430S (MTS430S)

sda 476.9G
sda1 127.9G|ntfs
sda2 32M|vfat
sda3 1G|vfat
sda4 977M|ext4

2) SanDisk Extreme Pro 128 GB USB flash drive:
/dev/sdc1 filesystem: ntfs

The regression when imaging the Samsung 990 Pro 1TB NVMe SSD is not making sense yet...

[Steven Shiau]

Please give testing Clonezilla live >= 3.3.4-22 or 20261102-* a try:
https://clonezilla.org/downloads.php
We have improve the mechanism about the option "-gmf". It might work better.
Let us know the results if you give that a try.
Thanks.

[Ken Pergola]

Hello Steven and good day to you,

I just tried20261102-tenacious and I can happily confirm that the "-gmf" option regression I experienced is now fixed.

Thank you very much for your attention to this -- it is greatly appreciated!

Best regards to you and your team,

Ken