Bug 40643

Summary: Pulseaudio crashes at volume/mute control of a RAOP device
Product: PulseAudio Reporter: youagree <n3ocort3x>
Component: modulesAssignee: Colin Guthrie <colin>
Status: RESOLVED MOVED QA Contact: pulseaudio-bugs
Severity: critical    
Priority: medium CC: colin, felix, lennart, n3ocort3x
Version: unspecified   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments: pulseaudio verbose log from its start till crash by trying to set RAOP volume (pulseaudio -vvv --log-target=file:)

Description youagree 2011-09-05 10:48:59 UTC
Created attachment 50926 [details]
pulseaudio verbose log from its start till crash by trying to set RAOP volume (pulseaudio -vvv --log-target=file:)

Circumstances:

I'm trying to use PulseAudio's RAOP discover and audio sink functionality with an Apple AirPort Base Station (A1264, 802.11bgn version, 2008 edition, with the latest firmware).

PulseAudio:

PulsAaudio can discover my AirPort device fine which is set up to get an IP address via wireless from my router via my real wireless AP. PulseAuidoo can set up an output device for it with a name of the "speaker name" configured in AirPort Base Station Utility (Windows version working fine in Wine with a little workaround).


Problem #1:
As soon as I try controlling the RAOP device's volume/mute state, without any application using it yet, pavucontrol errors out with a message box:

 <2> / pa_context_set_sink_volume_by_index() failed: Bad state

then when I close this, I can see another box:
 Connection failed: Connection terminated

I've tried to first kill the automatically respawned pulseaudio -k, then run it again with pulseaudio -vvv as normal user.

See complete log output in the attachment.

When I try setting the volume (nothing is playing on it yet) or muting it the following output is generated in pulseaudio's verbose output:

D: [pulseaudio] protocol-native.c: Client pavucontrol changes volume of sink raop.a1264.local.
D: [pulseaudio] module-raop-sink.c: Requested volume: 0:  63% 1:  63%
D: [pulseaudio] module-raop-sink.c: Got hardware volume: 0:  63% 1:  63%
D: [pulseaudio] module-raop-sink.c: Calculated software volume: 0: 100% 1: 100%
E: [pulseaudio] rtsp_client.c: Assertion 'c->url' failed at modules/rtp/rtsp_client.c:395, function rtsp_exec(). Aborting.

[See attachment for the full log]

Problem 2:

When I try assigning an application's output to the RAOP device without altering the volume (I named it A1264 in its utility), sounds like streaming starts through it, but it does stream with 2 seconds interruptions/pauses.

Therefore I tried to rebuild it from the current git source, it still does the same thing (crashes at volume setting of RAOP). 

I've tried recompiling it with the patch applied from:
http://pulseaudio.org/attachment/ticket/571/ignore-rtsp-requests-until-setup.patch

With the patch the daemon does not crash, but there is no sound output on the RAOP device, and applications that I tried (Clementine, qmmp) to output through the raop sink just hang without being able to stream any sound (they have to be forcefully killed first to get them streaming again through the normal default alsa-sink.)
Comment 1 Colin Guthrie 2011-09-05 12:12:29 UTC
Just FYI, it's generally not polite to change priorities. Everyone's bug is their own most important problem so this is really something the project organisers should set rather than users reporting bugs.

I'm going to see if I can plug in my AppleTV that I borrowed shortly and see what happens!
Comment 2 youagree 2011-09-05 12:24:32 UTC
Sorry about that, I didn't know. 

I was thinking the bug can be considered a bug of high importance if it causes the product's main process, i.e. the PulseAudio server to crash completely.

But if there was a similar problem with the alsa-sink, I think due to its higher popularity "high" would be more appropriate. 

(medium OK because not as many people want to use raop devices than alsa devices)
Comment 3 GitLab Migration User 2018-07-30 10:18:26 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/333.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.