Hi there,
i did some tests with two Cisco 7912G phones (SIP stack) yesterday. With
both ethereal and tcpdump listening on the Asterisk-Server's NIC, it
came up that all RTP packets were doubled, with some small but almost
constant delay (~460 us).
The setup is
7912G <--> ASTERISK <--> 7912G
The tcpdump output shows RTP traffic ASTERISK --> 7912G:
000000 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4851 0
000279 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4851 0
006736 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4852 160
000460 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4852 160
015967 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4853 320
000460 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4853 320
019229 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4854 480
000458 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4854 480
019679 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4855 640
000459 IP $ASTERISK.17944 > $DEST.16384: udp/rtp 160 c0 4855 640
^^^^^^ time delta in micro-seconds
^^^^ ^^^
Sequence No |||
Timestamp
Since sequence number and timestamp are equal for every two consecutive
packets, and the payload is completely the same as i discovered using
ethereal, it's obvious to me that this is one RTP stream, but two times
sent.
This doesn't occur using X-Lite.
It's the same issue with SIP requests, the Cisco phones send two equal
INVITES when you make a call (not so with X-Lite).
My questions: Is this normal behaviour (i guess not)? What is the
problem, the Cisco phones, Asterisk, tcpdump/ethereal...?
Why are the calls not "bridged" between the two phones (RTP traffic just
between the end-users) as they are when i use two X-Lite clients?
I hope you have some answers to my problem :-)
Thanks in advance,
Franz
_______________________________________________
--Bandwidth and Colocation provided by Easynews.com --
Asterisk-Users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
