I have been using the “Teams For Linux” flathub app for a few days and at first I celebrated, because I could finally use teams without major workarounds and so that it just works. However today it decided to crash my audio device after about 20 minutes and it did so consistently… It crashes it so hard, that I need to restart my system every time it occurrs…
Things that do not work to get my audio back up and running:
- restarting pipewire service with
systemctl --user restart pipewire.service
- restarting pipewire-pulse with
systemctl --user restart pipewire-pulse.service
- restarting alsa with
sudo alsactl -F restore
- relogging
- unplugging USB
I am using Fedora 41 and my audio interface is the Motu M2
As far as I can tell the audio is just broken beyond repair and I have no idea why. I’ve always had audio problems on linux, but never to this degree. Maybe I have to try another distro, but I actually don’t really want to. It is just super annoying and if anyone has an idea how to fix this mess I will try it and hope that it works :D
I’ve never used the --user flag to reset/restart a service.
Have you tried
systemctl reset-failed
?And then
systemctl daemon-reload
ordaemon-reexec
No I have not tried that, though pw-top still worked, in general pipewire still worked, I just didn’t get any sound anymore…
The
--user
flag commands come from this reddit thread: https://www.reddit.com/r/pop_os/comments/v3g2w9/is_there_a_cli_command_to_restart_pipewire/Maybe try without the flag.
You can see what errors may have caused the issue by checking
systemctl status NAMEOFSERVICE.service
or
journalctl -xem -o with-unit -p1 -p2 -p3 | grep -i pipewire
That may additionally help you diagnose the cause.
Since they are started with the user session these services do only exist there. If you try to restart them without the flag they just don’t exist.
The journal would be a nice way to look for sure, but the command you posted throws an error for me:
weird. Just change -xem to -xeb
You can also do
-xe --since yesterday
The
reset-failed
should also work to reset any/all failed units/services so you don’t have to go through one by one.Fedora may be a bit different in the need for the --user flag, so try with/without.
But I’ve had an entire user scope crash and dump and recovered using it.