Sound no longer works unless you kill it each time for each application

Bug #1129990 reported by Laura Czajkowski
46
This bug affects 11 people
Affects Status Importance Assigned to Milestone
pulseaudio (Ubuntu)
Fix Released
High
David Henningsson

Bug Description

this has only recently started.

If I start any application that needs sound - vlc, rhythmbox I hear nothing

I killall -9 pulseaudio

play vlc or rhythmbox and I then have sound

I lose my ability to use the sound bar from the indicator area and need to control volume directly from the application

At all times I hear the drum noise on start up, but that's the only sound I hear.

ProblemType: Bug
DistroRelease: Ubuntu 13.04
Package: pulseaudio 1:3.0-0ubuntu2
ProcVersionSignature: Ubuntu 3.8.0-6.13-generic 3.8.0-rc7
Uname: Linux 3.8.0-6-generic x86_64
ApportVersion: 2.8-0ubuntu4
Architecture: amd64
AudioDevicesInUse: Error: command ['fuser', '-v', '/dev/snd/by-path', '/dev/snd/controlC0', '/dev/snd/hwC0D0', '/dev/snd/hwC0D3', '/dev/snd/pcmC0D0c', '/dev/snd/pcmC0D0p', '/dev/snd/pcmC0D3p', '/dev/snd/seq', '/dev/snd/timer'] failed with exit code 1:
Date: Tue Feb 19 09:45:05 2013
InstallationDate: Installed on 2011-12-26 (420 days ago)
InstallationMedia: Ubuntu 11.10 "Oneiric Ocelot" - Release amd64 (20111012)
MarkForUpload: True
PulseList: Error: command ['pacmd', 'list'] failed with exit code 1: No PulseAudio daemon running, or not running as session daemon.
SourcePackage: pulseaudio
UpgradeStatus: Upgraded to raring on 2013-02-10 (8 days ago)
dmi.bios.date: 07/31/2012
dmi.bios.vendor: TOSHIBA
dmi.bios.version: Version 1.70
dmi.board.asset.tag: 0000000000
dmi.board.name: Portable PC
dmi.board.vendor: TOSHIBA
dmi.board.version: Version A0
dmi.chassis.asset.tag: 0000000000
dmi.chassis.type: 10
dmi.chassis.vendor: TOSHIBA
dmi.chassis.version: Version 1.0
dmi.modalias: dmi:bvnTOSHIBA:bvrVersion1.70:bd07/31/2012:svnTOSHIBA:pnSATELLITEZ830:pvrPT22LE-00F004EN:rvnTOSHIBA:rnPortablePC:rvrVersionA0:cvnTOSHIBA:ct10:cvrVersion1.0:
dmi.product.name: SATELLITE Z830
dmi.product.version: PT22LE-00F004EN
dmi.sys.vendor: TOSHIBA

Related branches

Revision history for this message
Laura Czajkowski (czajkowski) wrote :
Revision history for this message
Luke Yelavich (themuso) wrote : Re: [Bug 1129990] [NEW] Sound no longer works unless you kill it each time for each application
Download full text (7.8 KiB)

At first I thought this was due to the change in speex, and there being some fallout on other architectures for some reason, but it proved not to be the case. here is a log from an IRc session I had with Laura.

I'll have to defer to David on this one, given the attached log not really showing me anything obvious, although I am probably not looking hard enough.

19-02-2013 20:43:12 > TheMuso: Hrm interesting. No problem, https://wiki.ubuntu.com/PulseAudio/Log.
19-02-2013 20:45:04 > TheMuso: np.
19-02-2013 21:07:34 > TheMuso: I am actually in London this week for a sprint, and I'm in the middle of something atm, but I'll take a look in a bit.
19-02-2013 21:17:07 > TheMuso: So this has only benn going on since pulse 3 correct?
19-02-2013 21:17:18 > TheMuso: Because we certianly don't wnat pulseaudio to die.
19-02-2013 21:19:02 > TheMuso: So, were you able to reproduce this when you were dumping a pulseaudio log? What app did you try to play audio with?
19-02-2013 22:14:58 > TheMuso: Ok saw that...
19-02-2013 22:15:11 > TheMuso: So what happens, does vlc not make any sound, or is it suttery?
19-02-2013 22:16:33 > TheMuso: But killing pulse, and restarting it works fine...
19-02-2013 22:16:44 > TheMuso: Hmm.
19-02-2013 22:16:56 > TheMuso: Right.
19-02-2013 22:17:30 > TheMuso: So that problem reared its head even when you ran pulse manually to get the log?
19-02-2013 22:19:33 > TheMuso: Yep thats right, I'm just trying to work out in my head why things work after things have been killed and brought up again...
19-02-2013 22:22:11 > TheMuso: Right, I have a tiny suspician I know what it is. So, I need you to test something for me.
19-02-2013 22:22:24 > TheMuso: 1. Open /etc/pulse/daemon.conf in an editor as root.
19-02-2013 22:22:35 > TheMuso: 2. Change the resample-method setting from speex-float-1 to trivial.
19-02-2013 22:22:42 > TheMuso: 3. Restart pulse, and log it as per the wiki.
19-02-2013 22:22:55 > TheMuso: 4. Make sure you only run app to play audio, i.e vlc. Try playing something and tell me what happens.
19-02-2013 22:23:01 > TheMuso: 5. Then kill a restart things as you have been.
19-02-2013 22:26:32 > TheMuso: Well kill it and run it as per the log instructions on the wiki.
19-02-2013 22:26:47 > TheMuso: np
19-02-2013 22:26:53 > TheMuso: Still the same thing?
19-02-2013 22:27:11 > TheMuso: Oh ok.
19-02-2013 22:28:59 > TheMuso: Ok, so sotp logging pulse, get rid of ~/.pulse/client.conf etc and just use it for a while and see how things go. The sound may not be as clear as previously due to a different resampler, but I want to see if you get the same behavior.
19-02-2013 22:29:12 > TheMuso: No, I don't really need the log now, I think I know whats going on.
19-02-2013 22:29:33 > TheMuso: In the meantime, I am going to prepare you a package to install, to verify the second part of my theory. I'll be coming back to you.
19-02-2013 22:29:45 > TheMuso: No problem.
19-02-2013 22:29:53 > TheMuso: its quicker to debug like this actually.
19-02-2013 22:30:52 > TheMuso: There is a bug, but I don't think its in pulse, and I don't think its a code bug as such.
19-02-2013 22:31:11 > TheMuso: I'll get this package prepared, and get back to you and ...

Read more...

Revision history for this message
Raymond (superquad-vortex2) wrote :

[ 4333.091729] CPU3: Package power limit notification (total events = 590)
[ 4333.091734] CPU0: Package power limit notification (total events = 590)
[ 4333.091736] CPU1: Package power limit notification (total events = 590)
[ 4333.091738] CPU2: Package power limit notification (total events = 590)
[ 4333.102722] CPU3: Package power limit normal
[ 4333.102725] CPU1: Package power limit normal
[ 4333.102726] CPU0: Package power limit normal
[ 4333.102728] CPU2: Package power limit normal
[ 4676.302174] CPU2: Package power limit notification (total events = 689)
[ 4676.302177] CPU3: Package power limit notification (total events = 689)
[ 4676.302181] CPU0: Package power limit notification (total events = 689)
[ 4676.302184] CPU1: Package power limit notification (total events = 689)
[ 4676.313204] CPU0: Package power limit normal
[ 4676.313210] CPU3: Package power limit normal
[ 4676.313213] CPU2: Package power limit normal
[ 4676.313216] CPU1: Package power limit normal

Revision history for this message
David Henningsson (diwic) wrote :

Hi Laura,

Can you please do the following:

cp /etc/pulse/default.pa ~/.pulse/default.pa

then edit ~/.pulse/default.pa and add these lines (just above the first .nofail line)

set-log-target file:/tmp/pulseverbose.log
set-log-level 4
set-log-time 1

save and reboot. Log in and try to playback some sound, and notice that it is not working.
Now attach the file /tmp/pulseverbose.log to this bug. Thanks!

Changed in pulseaudio (Ubuntu):
status: New → Incomplete
Revision history for this message
Evgeny Brazgin (xapienz) wrote :

The bug affects me too after updating to pulseaudio 1:3.0-0ubuntu2.
pulseaudio-1:3.0-0ubuntu1 worked well.

Revision history for this message
Evgeny Brazgin (xapienz) wrote :

I attached the log file according to the request.

Revision history for this message
Laura Czajkowski (czajkowski) wrote :

attaching as requested.

Changed in pulseaudio (Ubuntu):
status: Incomplete → New
Revision history for this message
David Henningsson (diwic) wrote :

D ( 43.377| 0.000) [pulseaudio] reserve-wrap.c: Device unlock of reserve-wrapper@Audio0 has been requested and succeeded.

This line is the likely problem; it should never happen unless you're running JACK, and it should definitely never happen as a result of starting playback.

Also, I don't think it's a coincidence that both Laura, xapienz, and the reporter of bug 1127872 all seem to run VLC...

Changed in pulseaudio (Ubuntu):
importance: Undecided → High
assignee: nobody → David Henningsson (diwic)
Revision history for this message
David Henningsson (diwic) wrote :

Uploaded 3.0-0ubuntu3 of PulseAudio with a fix for this issue.

Changed in pulseaudio (Ubuntu):
status: New → Fix Committed
Revision history for this message
Evgeny Brazgin (xapienz) wrote :

David,
as far as I remember, the bug appears not with VLC only (now I'm unable to check it, but later I'll make similar log without vlc).

Amarok also stops sounding after pausing it and trying to play music some time later.

Revision history for this message
David Henningsson (diwic) wrote :

@xapienz: If you run amarok (before vlc), will you then see "Device unlock of reserve-wrapper@Audio0 has been requested and succeeded." in PulseAudio's log when starting amarok? If so, it's the same bug. It just seems some applications are more likely to trigger them than other ones. E g, I tried with "paplay" and that didn't trigger the bug.

Revision history for this message
David Henningsson (diwic) wrote :

Also, a symptom of this bug is that if you run "pacmd list" while the sound is non-working, you will find a line somewhere saying

   suspend cause: APPLICATION

...under the sink you're currently trying to play back through.

Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package pulseaudio - 1:3.0-0ubuntu3

---------------
pulseaudio (1:3.0-0ubuntu3) raring; urgency=low

  * 0106-reserve-Move-get_name_owner-to-the-public-rd_device-.patch,
    0107-reserve-Fix-leaking-NameLost-signals-after-release-a.patch:
    Fix sound suddenly stops working, seems particularly frequent with VLC
    (LP: #1129990, LP: #1127872)
 -- David Henningsson <email address hidden> Thu, 21 Feb 2013 10:55:16 +0100

Changed in pulseaudio (Ubuntu):
status: Fix Committed → Fix Released
Revision history for this message
Evgeny Brazgin (xapienz) wrote :

Seems to work like a charm.
Thank you =)

Revision history for this message
David Henningsson (diwic) wrote :

Thanks for confirming!

Revision history for this message
Simplehuman (simplehuman) wrote :

Someone has problems with Skype segmetation fault after this update?

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.