Hi,
I have a massive CPU usage (99%) from “tracker-miner-fs-3” each time I open my file manager.
My OS Endeavouros is up-to-date.
I use Gnome as a desktop manager.
killall tracker
return > tracker: no process found
Any idea how to resolve this problem?
This is supposed to act on the command that you provide tracker, which does not exist.
Find the specific command:
pgrep -a tracker
and either kill 9 the PID, or killall <command_name>.
OTOH, I would check more about this, with htop, or other process manager, to see what it is doing.
Actually, using htop, you can also do the above (find and kill) 
Happy hunting! 
I don’t use Gnome and i don’t know much about this tracker miner. Seems like it is used for indexing among other things? 
https://man.archlinux.org/man/tracker-miner-fs-3.1.en
https://wiki.gnome.org/Projects/Tracker
Edit: I use Kde Btw! 
I experienced the same issue the last couple of days. I am using gnome since a long time now. But this problem with CPU load for tracker-miner-fs-3 just started some days ago. Must have something to do with a recent update.
I also saw multiple coredumps related to the miner and the extract tool:
╰─# coredumpctl --no-pager
TIME PID UID GID SIG COREFILE EXE SIZE
Sat 2023-03-18 23:21:14 CET 10041 1000 1000 SIGSEGV missing /usr/lib/tracker-miner-fs-3 -
Sat 2023-03-18 23:21:48 CET 10616 1000 1000 SIGSEGV missing /usr/lib/tracker-miner-fs-3 -
Sat 2023-03-18 23:21:58 CET 10791 1000 1000 SIGSEGV missing /usr/lib/tracker-miner-fs-3 -
Sat 2023-03-18 23:22:39 CET 11158 1000 1000 SIGSEGV missing /usr/lib/tracker-miner-fs-3 -
Wed 2023-03-22 10:50:09 CET 23349 1000 1000 SIGSEGV present /usr/lib/tracker-miner-fs-3 327.1K
Wed 2023-03-22 10:51:38 CET 24594 1000 1000 SIGSEGV present /usr/lib/tracker-miner-fs-3 326.5K
Wed 2023-03-22 10:53:46 CET 25964 1000 1000 SIGSEGV present /usr/lib/tracker-miner-fs-3 326.0K
Wed 2023-03-22 11:02:11 CET 15386 1000 1000 SIGSEGV present /usr/lib/tracker-extract-3 12.9M
Wed 2023-03-22 11:02:12 CET 34172 1000 1000 SIGSEGV present /usr/lib/tracker-extract-3 8.6M
Wed 2023-03-22 12:54:07 CET 9728 1000 1000 SIGSEGV present /usr/lib/tracker-extract-3 14.4M
Wed 2023-03-22 12:54:07 CET 209760 1000 1000 SIGSEGV present /usr/lib/tracker-extract-3 4.9M
Wed 2023-03-22 13:15:03 CET 209872 1000 1000 SIGSEGV present /usr/lib/tracker-extract-3 13.0M
Wed 2023-03-22 14:25:03 CET 373400 1000 1000 SIGSEGV present /usr/lib/tracker-extract-3 13.7M
Wed 2023-03-22 15:05:03 CET 933355 1000 1000 SIGSEGV present /usr/lib/tracker-extract-3 13.7M
Then I saw in the journal that the miner has some issue with specific AVI files in a specific folder. The error messages look like this
tracker-extract: Failed to update ignored file 'file:///myfolder/DSCFxxxx.AVI':
I found thousands of them in the log. I used dconf-editor to put *.AVI files on the exclusion list and I renamed that folder. Now the CPU load issue does not occur anymore.
I wonder why there is no such report on the Gnome Bug Tracker…
https://gitlab.gnome.org/groups/GNOME/-/issues
I mean it could be a Bug related to avi file format indeed.
I can try reproducing.
And as a hint if you do not use Gnome apps like music and documents and you arte not browsing/searching a huge bunch of files over the filebrowser itself you can sevely disable the tracker-miner.
here is a good summary of Gnome tracker.
It appears that Gnome Tracker is a core part of Gnome and is hard to eliminate,
but you might be interested in the How can I disable Tracker in GNOME? section.
Pudge
I’ll never use Gnome again even though it has been many years since, there getting as bad as TikToc.
Thanks for Openbox.
I actually do not want to disable it. I like the functionality it offers. I am an in the file manager nautilus and do a search, it not only searches for filenames but also for file content.
In the meantime I have reverted the changes I mentioned earlier. I removed *.AVI from the file exclusion list (in fact I am using default exclusions again) and I renamed the folder with the AVI files back to its original name. I rebooted and everything is normal. No CPU load.
But I do see a new core dump:
Thu 2023-03-23 16:27:24 CET 13707 1000 1000 SIGSEGV present /usr/lib/tracker-miner-fs-3 326.7K
The timestamp indicates that this happened directly after I reverted the changes. I wil monitor this now and see if it is getting worse again.
I get a core dump every time I do a “tracker3 info ...” on a movie file. If it is mkv, mp4 or avi. Is that the same for you?
╰─# tracker3 info Steve-Jobs.mkv
Querying information for entity: 'Steve-Jobs.mkv'
No metadata available for that URI
Data object ?/data/media/Filme/MKV/Steve-Jobs.mkv? currently exists
results in:
Mär 23 16:58:14 rakete kernel: tracker-miner-f[33848]: segfault at 5562a7de8b3c ip 00007f141fbb2c3e sp 00007ffd28a04ac0 error 4 in libtracker-sparql-3.0.so.0.500.0[7f141fbad000+73000] likely on CPU 22 (core 12, socket 0)
Mär 23 16:58:14 rakete kernel: Code: 74 11 48 89 df ff 15 49 0a 0b 00 85 c0 0f 84 19 01 00 00 48 85 ed 74 54 ff 15 96 0d 0b 00 48 89 c6 48 8b 45 00 48 85 c0 74 05 <48> 39 30 74 3d 48 89 ef ff 15 1c 0a 0b 00 85 c0 75 30 48 8d 15 81
Mär 23 16:58:15 rakete systemd[1]: Started Process Core Dump (PID 33853/UID 0).
Mär 23 16:58:15 rakete systemd-coredump[33854]: [🡕] Process 33848 (tracker-miner-f) of user 1000 dumped core.
Stack trace of thread 33848:
#0 0x00007f141fbb2c3e tracker_sparql_connection_new (libtracker-sparql-3.0.so.0 + 0x1ec3e)
#1 0x00005567f109ac38 main (tracker-miner-fs-3 + 0xec38)
#2 0x00007f141f955790 n/a (libc.so.6 + 0x23790)
#3 0x00007f141f95584a __libc_start_main (libc.so.6 + 0x2384a)
#4 0x00005567f109bf45 _start (tracker-miner-fs-3 + 0xff45)
Stack trace of thread 33851:
#0 0x00007f141fe452d4 g_socket_condition_check (libgio-2.0.so.0 + 0x972d4)
#1 0x00007f141fec4a8a n/a (libgio-2.0.so.0 + 0x116a8a)
#2 0x00007f141fec4ba9 n/a (libgio-2.0.so.0 + 0x116ba9)
#3 0x00007f141fec617d n/a (libgio-2.0.so.0 + 0x11817d)
#4 0x00007f141fe567f4 n/a (libgio-2.0.so.0 + 0xa87f4)
#5 0x00007f141fe5a5ed n/a (libgio-2.0.so.0 + 0xac5ed)
#6 0x00007f141fec4931 n/a (libgio-2.0.so.0 + 0x116931)
#7 0x00007f141fe41ebe n/a (libgio-2.0.so.0 + 0x93ebe)
#8 0x00007f141ffddafb g_main_context_dispatch (libglib-2.0.so.0 + 0x5aafb)
#9 0x00007f142003a5d9 n/a (libglib-2.0.so.0 + 0xb75d9)
#10 0x00007f141ffdd0cf g_main_loop_run (libglib-2.0.so.0 + 0x5a0cf)
#11 0x00007f141febde4c n/a (libgio-2.0.so.0 + 0x10fe4c)
#12 0x00007f142000b7c5 n/a (libglib-2.0.so.0 + 0x887c5)
#13 0x00007f141f9b7bb5 n/a (libc.so.6 + 0x85bb5)
#14 0x00007f141fa39d90 n/a (libc.so.6 + 0x107d90)
Stack trace of thread 33850:
#0 0x00007f141fa2c9df __poll (libc.so.6 + 0xfa9df)
#1 0x00007f142003a53f n/a (libglib-2.0.so.0 + 0xb753f)
#2 0x00007f141ffdb382 g_main_context_iteration (libglib-2.0.so.0 + 0x58382)
#3 0x00007f141ffdb3d2 n/a (libglib-2.0.so.0 + 0x583d2)
#4 0x00007f142000b7c5 n/a (libglib-2.0.so.0 + 0x887c5)
#5 0x00007f141f9b7bb5 n/a (libc.so.6 + 0x85bb5)
#6 0x00007f141fa39d90 n/a (libc.so.6 + 0x107d90)
Stack trace of thread 33849:
#0 0x00007f141fa320dd syscall (libc.so.6 + 0x1000dd)
#1 0x00007f142002ea35 g_cond_wait (libglib-2.0.so.0 + 0xaba35)
#2 0x00007f141ffa7fc4 n/a (libglib-2.0.so.0 + 0x24fc4)
#3 0x00007f142000d43e n/a (libglib-2.0.so.0 + 0x8a43e)
#4 0x00007f142000b7c5 n/a (libglib-2.0.so.0 + 0x887c5)
#5 0x00007f141f9b7bb5 n/a (libc.so.6 + 0x85bb5)
#6 0x00007f141fa39d90 n/a (libc.so.6 + 0x107d90)
Stack trace of thread 33852:
#0 0x00007f141fa2c9df __poll (libc.so.6 + 0xfa9df)
#1 0x00007f142003a53f n/a (libglib-2.0.so.0 + 0xb753f)
#2 0x00007f141ffdb382 g_main_context_iteration (libglib-2.0.so.0 + 0x58382)
#3 0x00007f141eca2fde n/a (libdconfsettings.so + 0x5fde)
#4 0x00007f142000b7c5 n/a (libglib-2.0.so.0 + 0x887c5)
#5 0x00007f141f9b7bb5 n/a (libc.so.6 + 0x85bb5)
#6 0x00007f141fa39d90 n/a (libc.so.6 + 0x107d90)
ELF object binary architecture: AMD x86-64
EDIT:
The core dump also happens with jpg and png files. It does not happen for txt, ods, odt, pdf.
its quite easy to disable tracker, its the first thing i always do, done it for years, no problems
sudo systemctl mask tracker-miner-fs-3.service tracker-store.service tracker-miner-fs.service tracker-miner-rss.service tracker-extract.service tracker-miner-apps.service tracker-writeback.service
might need a reboot
to revert i guess you do unmask
I want to make it work. I do not want to disable it.
let me check…
i do not have this here it just works fine and shows file infos…
But … i do use gnome unstable (44) currently…
same on current…
what does:
LANG=C tracker3 status show on yours?
Easy to disable it …just install KDE! 
╰─# LANG=C tracker3 status
Currently indexed: 11009 files, 892 folders
Remaining space on database partition: 317,8?GB (78,52%)
All data miners are idle, indexing complete
2 recorded failures
Path Message
?path/f0598c6dfaee4909b1169b822035e537.gif Could not get any metadata for uri:'file:///home/matthias/Dokumente/?
/path-in-home//xyzxyz.pdf ?ms.tex? is not an absolute IRI
But then you have to disable baloo 
Pudge
Not always…