Right after updating to 2023-10-05 [Stable update] 2023-10-04 the HDD Led started blinking like crazy. I checked the running processes and noticed “tracker-extract-3” and “tracker-miner-fs-3” in the process list.
I have to admit I panicked and halted the 'tracker-extract-3" process. Then I checked with the Gnome System Logs viewer and noticed over 15,000 “Failed to update ignored file ‘… FILENAME…’” messages.
Is this related to the update? What’s the next step to avoid “tracker-extract-3” going crazy?
The Gnome System Logs viewer is running slow now because of the 15,000 records.
EDIT — (2023-10-07):
correction to ‘[Stable update] 2023-10-04’ instead of the prior mentioned ‘[Testing update] 2023-10-05’
I’ve checked and user-1000.journal is 67.1 MB in size now. The error messages read:
Failed to update ignored file ‘file:///usr/share/applications/cups.desktop’: Parser error at byte 0, expected one of ‘base’, ‘prefix’, ‘constraint’, ‘load’, ‘clear’, ‘drop’, ‘add’, ‘move’, ‘copy’, ‘create’, ‘insert’, ‘delete’, ‘with’:
Failed to update ignored file ‘file:///usr/share/applications/cups.desktop’: Parser error at byte 0, expected one of ‘base’, ‘prefix’, ‘constraint’, ‘load’, ‘clear’, ‘drop’, ‘add’, ‘move’, ‘copy’, ‘create’, ‘insert’, ‘delete’, ‘with’:
…
The error messages also seem to be repeating. I’ve roughly ~6,000 records for every file in file:///usr/share/applications/. Don’t know if only this directory is affected or if it would have gone on hadn’t I stopped the process.
The files in /usr/share/applications seem to be fine. All in Kb, no 0 byte files.
I also tried a reboot and it didn’t help. HDD still blinking until I stop the process ‘tracker-extract-3’. Strangely no new “Failed to update ignored file …" messages in the system log but now I get coredumps for ‘tracker-extract’
systemd-coredump[2332]: Process 2295 (tracker-extract) of user 1000 dumped core.
systemd-coredump[2878]: Process 2343 (tracker-extract) of user 1000 dumped core.
I didn’t have 100% cpu usage. The information is pretty outdated over there (2017) and deals with openSUSE instead of Manjaro. The only solution they come up with is disabling the tracker.
I’ve checked pacman.log to see if the tracker components have been updated. They have:
That’s why I think it’s definitely upgrade-related. It literally started to get crazy right after the upgrade was done with ‘sudo pacman -Syu’
When I started the PC today, nothing happend though. No more coredumps, nothing in the system logs about “Failed to update ignored file …”. Haven’t done anything since then. No updates, no changes and suddenly all seems to work.
There is no “tracker-extract-3” process running. Maybe it’s just pausing and the error will come back once it starts indexing again?
I’m still looking for a solution myself but don’t find any related reports. I’ve found this dicussion and similar ones but they have all been fixed by an update to the tracker back then.
Just to clarify, it seems I’ve initially posted this on the wrong post. Sorry!
The error described above wasn’t because of the [Testing update] 2023-10-05 but the [Stable update] 2023-10-04 a day earlier.
‘tracker-extract-3’ still hasn’t behave badly since then, so I’m not sure what to do or look out for. Maybe it fixed itself somehow?
It made me realize though that I’m not very well prepared in case something goes wrong after an update. Is there information/a wiki page out there with tips what to prepare for the worst case? Sort of a first aid kit for Manjaro? Would be thankful for something I can read up to prepare myself instead of going ‘full panic mode’ everytime something unusual seems to happen.