1

After upgrading to Debian 12 with KDE multiple issues appeared. The following was the most problematic:

A database-wal file with a size of multiple GB appeared and I had to delete it like so rm ~/.local/share/kactivitymanagerd/resources/database-wal and did the same for ..resources/working backup/database-wal which was almost as large as the former file.

I found this large file with gdmap after running out of disk space which didn't show the file anymore after deleting the file and clicking refresh. According to the tool, I should have several GBs of free disk space.

I also ran sudo ls -l /home/ to make sure there only is the folder "lost+found" and "{username}" in it. sudo ls -l /home/lost+found shows total 0 so all data should be in the folder of my username. I also tried showing the largest files with sudo find /home/ -mount -type f -size +50M -exec du -h {} \; | sort -n which didn't show the files anymore. Nevertheless, the Dolphin file manager showed that no data is left and this was not merely adisplay error of Dolphin. I managed to delete a few MBs from the trash bin but at some point things crashed so I had to reboot yet another time from the TTY. After the latest reboot, several GBs are suddenly free on the /home/ partition.

This is likely linked to my problem that already appeared under Debian11 where disk space is not freed up when deleting files on a hard drive (not the /home/ partition).

What could have caused this large database-wal file (or how to find out) and how to see all files that take up disk space or refresh the drive's free disk space (without having to manually go through deleted files that are being used since the rm command should have actually deleted them and if not displayed some error about it not being able to delete the file since it's being used)? This would also be useful with the other drive which still shows that no disk space is free even when deleting things and this problem could happen again on other drives as well.


These error messages show often so there's a small chance they have something to do with it: plasmashell: kf.config.core: couldn't lock local file and Failed to load module "atk-bridge". During the upgrade I reset limits.conf which previously had * hard core 0. I also moved a vorta desktop file out of /home/username/config/autostart/. The issue may have appeared while using Wayland, I currently use X11 due to freeze issues that don't appear on a nearly identical Debian12 machine.

I don't know what caused it and the GUI for logs KSystemLog doesn't make it very easy to find relevant logs and doesn't show many logentries, I only looked at journalctl entries. Since there repetitive messages are not yet prevented in a good way (like a different log type if the message was repeated often) there are thousands of errors kded5: HistoryFile::add.write about no storage space being available anymore on this device (can't find any search engine entry for this error) I had to use sudo journalctl | tail -n 500000 where so far I only found:

 ksmserver[2859]: .. - root - DEBUG - Expected error parsing file in .ssh: known_hosts (You can safely ignore this)
 ksmserver[2859]: Traceback (most recent call last):
 ksmserver[2859]:   File "/usr/lib/python3/dist-packages/vorta/utils.py", line 195, in get_private_keys
  ksmserver[2859]:     parsed_key = key_format.from_private_key_file(key_file)
 ksmserver[2859]:                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 ksmserver[2859]:   File "/usr/lib/python3/dist-packages/paramiko/pkey.py", line 249, in from_private_key_file
 [...]

1 Answer 1

1

The space is not freed because a process was opening that file, you can find it by:

lsof ~/.local/share/kactivitymanagerd/resources

Then you can kill them, or simply reboot your machine

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .