Red Hat Bugzilla – Bug 496608
pulseaudio stops playing sound and hangs all sound-playing applications
Last modified: 2014-10-06 07:10:16 EDT
Created attachment 340314 [details]
Log of pulseaudio -vvvv up to and just after the bug was triggered
Description of problem:
This is an off-shoot of bug 472339 - I reported this problem there but it was thought to be due to a problem with snd_pcm_avail() which has been fixed in the latest kernels.
What happens is that sound will suddenly stop playing, though pulseaudio is still responsive. New applications can start and initialise streams which pulseaudio can see, but as soon as the application tries to send some sound data, polls to the application's file descriptor time out (and no sound is heard). This problem can occur with any sound playing application, including audacious, totem, mplayer etc. Pulseaudio needs to be restarted to fix the problem.
My sound driver is snd-intel8x0
My sound card is: 00:1f.5 Multimedia audio controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) AC'97 Audio Controller (rev 02)
Version-Release number of selected component (if applicable):
Seems fairly random - it can happen 3 times in an hour, then not happen for a day.
Steps to Reproduce:
1. Play some sound. I have no idea what actually triggers it.
2. After a while, sound stops playing. Any application from this point that
tries to play something hangs or busy waits.
3. Pulseaudio usually needs to be restarted to make sound work again. Occasionally it works again after you stop the sound, leave it for a bit and try playing the sound a few more times.
See first few comments of bug 362339.
I will attach a log of when this last happened. Sorry that it's very long -- it took quite a while to trigger on this particular occasion. I have stripped out duplicate lines; they are replaced with the line and a notice about how many times they were repeated.
The bug occurred ~100 lines before the end (I have marked the place in the file), and after that I tried to stop and play the sound again, so you can see a reconnection and where it starts to complain about a massive underrun because the client times out when trying to send any audio data.
There are many messages about underruns because it seems pulseaudio performs very badly on my machine - regularly using 15% of my 2.8GHz CPU. Only some of these underruns are audible though so while annoying I can live with it and they don't seem to have any direct relation to the problem.
Smells like a driver bug.
Could you please get me a backtrace overall streams of PA when this happens?
You'd need to install debug symbols for that. Then just reproduce the issue and attach to the running process with gdb. Then do a "bt full" on all threads. Do you think you can do that?
Sure - I will do that next time it happens. I tried that once before but I hadn't thought to do a full backtrace, so I just got the active thread and found nothing useful.
Created attachment 341037 [details]
GDB backtrace of pulseaudio & audacious client
Created attachment 355152 [details]
Log "pulseaudio --log-level=4"
I have the same problem.
Hmm, the backtrace shows nothing. Could you please provide the output of "pacmd ls" when this happens?
Created attachment 356970 [details]
"pacmd ls" output
Is this F10 specific? Can you reproduce this on F11?
For refference see also bug 524007
Lennart, this is reproducible on F11, but happens also without PulseAudio. For example, see the thread
where I reproduced it _without_ PulseAudio and saw the same problems with poll(). For a couple of weeks I've thought it had been fixed by a kernel upgrade that touched ALSA (which turnt deadlocks into underruns for me), but it's still there according to other users and not specific to Audacious.
Ok, reassigning to the kernel then.
This message is a reminder that Fedora 10 is nearing its end of life.
Approximately 30 (thirty) days from now Fedora will stop maintaining
and issuing updates for Fedora 10. It is Fedora's policy to close all
bug reports from releases that are no longer maintained. At that time
this bug will be closed as WONTFIX if it remains open with a Fedora
'version' of '10'.
Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version'
to a later Fedora version prior to Fedora 10's end of life.
Bug Reporter: Thank you for reporting this issue and we are sorry that
we may not be able to fix it before Fedora 10 is end of life. If you
would still like to see this bug fixed and are able to reproduce it
against a later version of Fedora please change the 'version' of this
bug to the applicable version. If you are unable to change the version,
please add a comment here and someone will do it for you.
Although we aim to fix as many bugs as possible during every release's
lifetime, sometimes those efforts are overtaken by events. Often a
more recent Fedora release includes newer upstream software that fixes
bugs or makes them obsolete.
The process we are following is described here:
Fedora 10 changed to end-of-life (EOL) status on 2009-12-17. Fedora 10 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.
If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version.
Thank you for reporting this bug and we are sorry it could not be fixed.