Last week I migrated some of our servers to Asterisk 1.4.18 and we started seeing audio drops of several seconds during SIP calls. After investigating it we noticed that Asterisk was increasing the RTP timestamps abnormally during a conversation.
I'm including a text file with a subset of the data collected by Wireshark that shows the problem (I have the complete packet capture if anybody needs it to analyze it). The Asterisk server is the one whose IP address ends in .38. If you look at the packet with the number 14910 (seq 23369) you'll see that the next packet from Asterisk (14919, seq 23370) increases the RTP timestamp from 77120 to 2280582632. We've tried enabling and disabling internal timing and the jitter buffer, but it made no difference whatsoever. I also added the patch present in ticket #10355 to Asterisk 1.4.18, but it didn't help. Has anybody else experienced a problem like this one?
14898 52.678422 63.215.27.55 66.150.122.38 RTP PT=ITU-T G.711 PCMU, SSRC=0x352D1216, Seq=22216, Time=1043051951 14899 52.678576 66.150.122.38 63.215.27.55 RTP PT=ITU-T G.711 PCMU, SSRC=0x404D77E4, Seq=23368, Time=76960 14909 52.698326 63.215.27.55 66.150.122.38 RTP PT=ITU-T G.711 PCMU, SSRC=0x352D1216, Seq=22217, Time=1043052111 14910 52.699321 66.150.122.38 63.215.27.55 RTP PT=ITU-T G.711 PCMU, SSRC=0x404D77E4, Seq=23369, Time=77120 14917 52.718417 63.215.27.55 66.150.122.38 RTP PT=ITU-T G.711 PCMU, SSRC=0x352D1216, Seq=22218, Time=1043052271 14919 52.720938 66.150.122.38 63.215.27.55 RTP PT=ITU-T G.711 PCMU, SSRC=0x404D77E4, Seq=23370, Time=2280582632 14921 52.721029 66.150.122.38 63.215.27.55 RTP PT=ITU-T G.711 PCMU, SSRC=0x404D77E4, Seq=23371, Time=2280582792 14922 52.721052 66.150.122.38 63.215.27.55 RTP PT=ITU-T G.711 PCMU, SSRC=0x404D77E4, Seq=23372, Time=2280582952
_______________________________________________ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
