Seg fault when creating snapshot

[lazicdanilo]

Describe the bug
Snapshot creation goes over 100% and fails with Segmentation fault. After this partial snapshot is created but not displayed in the GUI. I see it and can remove it manually from the drive

Screenshot

To Reproduce
Steps to reproduce the behavior:

1. Create a snapshot

System:

• uname -a: Linux archlinux 6.12.51-1-lts #1 SMP PREEMPT_DYNAMIC Mon, 06 Oct 2025 11:05:41 +0000 x86_64 GNU/Linux
• Desktop: i3
• Application Version: 25.07.7

[lazicdanilo]

When downgraded to 25.07.5 the snapshot was successful (no seg fault), but the percentage was around 140

[ygerlach]

@lazicdanilo Thanks for your report!
The percentage is raising over 100%, because the size it assumes to be 100% is an estimate or the size of the last snapshot. See here: #383

I had problems reproducing it on Linux mint using timeshift 25.07.7. Which distro are you using?

Could you capture a coredump of timeshift crashing?
Here are some steps for that:

# start timeshift inside gdb
sudo gdb --args timeshift-gtk --debug

# use the command "run" to start timeshift
# create a snapshot and let it crash
# use the following commands to gather informations

thread apply all bt full

generate-core-file timeshift-crash.core

quit

If you could upload the result of those commands, it would help us investigate the issue.
If you could upload timeshift-crash.core would also help alot (it contains all the memory of the process during the crash, so it might contain some private information like your username or comments of your snapshots).

[lazicdanilo]

I'm using Arch Linux.

I did the following:

1. Mounted the drive manually and went to the timeshift/snapshots/ dir and created dummy img.bin file with all 0s to simulate drive being almost full (Command used: sudo dd if=/dev/zero of=zeros.img bs=1G count=48 status=progress). The remaining space on drive is small enough that the next snapshot will definitely fail.
2. Unmounted the drive and made sure everything is synced.
3. Started timeshift-gtk using gdb.
4. Started creating the snapshot. The snapshot did start to sync files as the pre-check if there's enough space returned space needed as 20.6 MB (From logs: Space required for snapshot: 20613069 (20.6 MB). Space available: 12331614208 (12.3 GB)). This is quite optimistic as my last snapshot was on 2025-07-07 (more then 3 months ago).
5. As expected, snapshot failed to create. I got the error message and now my remaining space is 0 B. One of the issues for now is that the partially created snapshot (which I can see in /run/timeshift/149942/backup/timeshift/snapshots/) is not displayed in the GUI, and I don't think I can remove it from GUI.
6. Deleted one of my old snapshots to get some space on the drive. I got 13.3 GB.
7. Started creating new snapshot again. Again the pre-check was a success and file sync started.
8. Again, as expected the snapshot failed and did not exist in the GUI.

The steps above are the same steps I did when I encountered SegFault for the first time. Unfortunately I was unsuccessful in triggering the SegFault again. A few days ago (when I first reported the issue) I had only one SegFault, I did not try to reproduce it then, so I'm not sure how consistent it is.

But I did notice two more potential issues:

1. Pre-check estimates the needed size way too low.
2. Partial snapshots are not displayed in the GUI.

In the attachment I'll send the timeshift-gdb-out.txt which is the output of the gdb.

Hope this helps!

timeshift-gdb-out.txt

[ironhak]

Hello

Arch update of ~two days ago broke lots of apps and they throw error "segmentation fault". Now it should all be fixed

[ygerlach]

So it was an Arch issue and this can be closed here?

[lazicdanilo]

As SegFault issue is not reproducible I think we can close this. I would however still look into two potential issues mentioned above:

1. Pre-check estimates the needed size way too low.
2. Partial snapshots are not displayed in the GUI.