Quantcast

Strange behaviour of udpsrc

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Strange behaviour of udpsrc

Myzhar
Hi all,

I have a simple pipeline that listens on a port and redirects an audio stream to the speakers.

The input stream is really variable since there are different "talkers" that connect to the UDP source, but never in the same time.

The problem is that sometimes, in a "what-it-seems" random way, the pipeline hangs out and stops sending out the incoming stream, the only way to let it start again to stream is to kill the process and restart it.

This is the pipeline:
gst-launch-1.0 udpsrc port=5555 ! "application/x-rtp,media=audio,clock-rate=44100,width=16,height=16,encoding-name=L16,encoding-params=1,channels-1,payload=11" ! queue ! rtpL16depay ! audioconvert ! autoaudiosink

Any idea about what can cause this issue? Is there a good way to redirect a debug stream to a text file and analize it after the freeze?

Thank you for help
Walter
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange behaviour of udpsrc

Myzhar
Hi all,
debugging the pipeline I noticed that when the stream hangs out I have a long sequence of warning related to Clock Skew:

audiobasesink gstaudiobasesink.c:1287:gst_audio_base_sink_skew_slaving:<autoaudiosink0-actual-sink-pulse> correct clock skew xxxxxxxxxxxxxxxxxxxx > 20000000

where xxxxxxxxxxxxxxxxxxxx is a huge number increasing continuosly.

How can correct this effect?

Thank you
Walter
Loading...