Lost microphone, some news

[Joakim Tjernlund]

On Wed, 2018-06-27 at 09:31 +0100, David Woodhouse wrote:
> On Wed, 2018-06-27 at 11:33 +1200, Eion Robb wrote:
> > Hi Jocke,
> > 
> > The devel@ mailing list might be more appropriate to discuss code and
> > technical changes.  You might also want to get in touch with the sipe
> > plugin developers on their forum, as I don't believe the monitor the
> > Pidgin mailing lists.
> 
> I suspect it isn't SIPE-specific; I'm seeing something fairly similar
> with Chime. Sometimes, the pulsesrc element just stops feeding me data.

Does it happen in mid call or just during call connect?

I get in mid (Test) call, I also see the PA dead(not just pause):
0:01:00.724872738 26402 0x55b6da430cf0 WARN                 srtpdec gstsrtpdec.c:1254:gst_srtp_dec_chain:<srtpdec_1> Invalid buffer, dropping
0:01:00.898523845 26402 0x55b6da3ed190 FIXME               basesink gstbasesink.c:3144:gst_base_sink_default_event:<autoaudiosink1-actual-sink-pulse> stream-start event without group-id. Consider implementing group-id handling in the upstream elements
0:01:01.178840759 26402 0x7f7c1c020790 WARN                   pulse pulsesink.c:702:gst_pulsering_stream_underflow_cb:<autoaudiosink1-actual-sink-pulse> Got underflow
0:01:09.311587636 26402 0x7f7c1c020790 WARN                   pulse pulsesink.c:702:gst_pulsering_stream_underflow_cb:<autoaudiosink1-actual-sink-pulse> Got underflow

These underflow always seem to be present but underflow does not always cause the problem.

0:01:28.330415659 26402 0x55b6da423ad0 WARN                   pulse pulsesrc.c:1117:gst_pulsesrc_read:<autoaudiosrc1-actual-src-puls> error: pa_dead: OK
0:01:28.330577425 26402 0x55b6da423ad0 WARN                   pulse pulsesrc.c:1204:gst_pulsesrc_read:<autoaudiosrc1-actual-src-puls> error: pa_unlock_and_fail: OK
0:01:28.330653202 26402 0x55b6da423ad0 WARN                audiosrc gstaudiosrc.c:221:audioringbuffer_thread_func:<autoaudiosrc1-actual-src-puls> error xxxreading data -1 (reason: Resource temporarily unavailable), skipping segment
0:01:28.343417227 26402 0x7f7be40296d0 WARN         audio-resampler audio-resampler.c:274:convert_taps_gint16_c: can't find exact taps
0:01:28.344178723 26402 0x7f7be40296d0 WARN                 basesrc gstbasesrc.c:2939:gst_base_src_loop:<autoaudiosrc1-actual-src-puls> error: Internal data stream error.
0:01:28.344219932 26402 0x7f7be40296d0 WARN                 basesrc gstbasesrc.c:2939:gst_base_src_loop:<autoaudiosrc1-actual-src-puls> error: streaming stopped, reason error (-5)
0:01:28.383420661 26402 0x55b6d9a03ad0 WARN                   pulse pulsesrc.c:1785:gst_pulsesrc_pause:<autoaudiosrc1-actual-src-puls> error: pause_pause: Bad state

On my laptop I seldom see this problem though(wonder if the presence of a camera make is go away?)

 Jocke
> 
> https://bugzilla.gnome.org/show_bug.cgi?id=795374
> 
> I tried a hack to set the pipeline into PLAYING later, in an attempt to
> make it go away. It made it happen 100% of the time instead. Which I
> suppose is a step towards diagnosing it...
> 
> I think this is the hack I had at the time; I need to get back to it
> now I've fixed the more important problem of the incoming audio going
> away (which required a Pidgin restart to fix it; this one just needs me
> to close the call and rejoin it).
> 
> Added Jan to Cc since he seemed to have coherent thoughts about this
> when I first reported it.