00:53 <riteo> no there's something fishier going on... I'm absolutely baffled on why it stopped working all of a sudden, I've been debugging this shit for the whole day

06:34 <sewn> ehawkvu: what is the c in echawk in your github name mean?

07:09 <riteo> aight folks, I cracked the secret: the latest pipewire package at version 1.0.3+0.4.82 is broken. 1.0.0+0.4.16-1 works fine

07:09 <riteo> I had a lot of trouble debugging the thing because, very embarassingly, the daemon would persist between sessions (I assumed it wouldn't) hehe

07:52 <riteo> testuser[m]: you're the maintainer of the pipewire package, right? Does pipewire work for you? I can consistently use older 1.0.0+0.4.16-1 version but not the new 1.0.3+0.4.82 one, which is weird, as all the alsa-udev stuff got unchanged between those versions, so I'm not sure what changed to break device probing (I only have "Dummy Output")

08:25 <riteo> it broke again wtf

08:31 <sewn> i love pipewire

08:38 <riteo> newsflash: I ran it at root and it worked

08:38 <riteo> my existence is becoming more and more miserable at every attempt

08:39 <riteo> sewn: the funny thing is that it just worked the first time I setup it and was like "uh, that's it?" and worked with anything I threw at it

08:39 <riteo> then one day I reboot and here we are

08:41 <riteo> ok while testing I had set up a broken user config, 0.3.85+0.4.15-1 definitely works

08:44 <riteo> 1.0.3 does _not_ work when ran from the user, but it does work when run from root

08:46 <riteo> 1.0.0 works fine

08:50 <riteo> the issue might be wireplumber, as mixing 1.0.3 and a new wireplumber version works

09:13 <riteo> yup, got it finally. You can use _any_ pipewire version (1.0.3 works fine), but the only wireplumber setup I could find was 0.4.16 WITH no-dbus.patch applied

09:14 <riteo> if you take that out it just stops working, migth be related to the device reservation module, not sure. One thing though is that 0.4.82 completely revamped the scripts so it might "just" be a matter of again disabling the right things.

09:14 <riteo> I've been debugging this stuff for more than one day, thus, my adventure ends here for now, happy to have pipewire 1.0.3+0.4.16 (with no-dbus.patch).

09:15 <riteo> Sorry for the wall of text

09:16 <riteo> testuser[m]: hope this ends up useful to troubleshoot any problem that new wireplumber versions might give