Ever so often I get one of these processes utilising 100% of one CPU core:
baloo_file_extractor
baloo_file
There was a baloo_runner as well, but I might have been on openSUSE 42.3 machine and not the TW.
Kinfo File Indexer Monitor says that indexer is idle, but balooctl says the indexer is running.
Executing
sudo balooctl stop
does not kill either process.
I do want file to be indexed for search purposes, but is there a way to make the indexing less aggressive? Having one core at 100% utilisation heats up the CPU and fan noise (fan doesn’t even kick in under normal usage, when indexer is not running) is getting at me.
Baloo runs as the logged in user, not root.
So do not use sudo!
I do want file to be indexed for search purposes, but is there a way to make the indexing less aggressive?
It shouldn’t be that aggressive normally.
Maybe it has problem with a certain file. You can exclude specific directories from indexing in its settings.
OTOH, a corrupt database can cause something like this, a seemingly similar problem has been fixed recently:
https://cgit.kde.org/baloo.git/commit/?id=f57c4bc27b81f2b40d910d376a0bb531cd8472c6
So try to delete ~/.local/share/baloo/ to recreate the index from scratch.
Hi,
running
balooctl stop
without root would kill the processes either.
Anyway, I’ve deleted ~/.local/share/baloo/, manualy killed baloo processes, run
balooctl start
followed by
balooctl monitor
baloo_file_extractor does jump to 100% utilisation of one CPU core, but it is not constantly at 100%. Also, “balooctl monitor” now show what files it has indexed, whereas before deleting ~/.local/share/baloo/ monitoring would not show anything, just say “press ctrl+c to stop monitoring” or something along the lines.
Will see what happen after it re-indexes my SSD.
Deleting database and re-indexing everything seems to have worked. I’ve even saved about 3GB of space. Hopefully Baloo DB corruption will not occur every other week as reindexing does take some time to complete.