[asterisk-users] Delayed RTP start
jeff at stratustalk.com
Wed Jul 24 13:47:08 CDT 2019
I am debugging an issue that unfortunately involves two NAT instances -
the firewall at our customer site, and the firewall in front of their
I have an HTEK phone at the customer site registering to the public
address of the Amazon instance running asterisk (and FreePBX). This
seems to work fine, and it can call local services (like fpbx *65 to
read back the extension) with no problems.
If it tries to make an outbound outside call, the remote phone (my cell
for example) rings, I answer it, but there is no audio in either
direction for nearly exactly 16 seconds, every time. Then audio starts
in both directions without issue.
I did a packet trace on the phone itself and see 16 seconds of outbound
RTP with no inbound, then suddenly RTP in both directions until the call
I did a packet trace on the asterisk side and see the call setup, then
sixteen seconds of nothing (??), then RTP starts in both directions.
In the asterisk console I see this bit of interestingness:
[2019-07-24 13:21:02] DEBUG: chan_sip.c:29923
__start_session_timer: Session timer started: 78 -
261da at 10.0.116.239:60060 1768000ms
-- SIP/ast01-0000024b answered SIP/7222-0000024a
[2019-07-24 13:21:02] DEBUG[C-000001f1]: bridge_native_rtp.c:660
native_rtp_bridge_compatible_check: Bridge '3bfbf253-d34f-
45e2-abc3-75e590d81739' can not use native RTP bridge as channel
'SIP/ast01-0000024b' has DTMF hooks
[2019-07-24 13:21:18] DEBUG[C-000001f1]: res_rtp_asterisk.c:4179
ast_rtp_write: Ooh, format changed from none to ulaw
[2019-07-24 13:21:18] DEBUG[C-000001f1]: res_rtp_asterisk.c:4019
rtp_raw_write: Starting RTCP transmission on RTP instan
So my main question is, what would cause a sixteen second delay before
the codec could be decided?
This is Asterisk 13.25.0 on the customer Amazon instance... the "ast01"
peer is ours also - one of our external gateways, also running 13.25.0.
703 496 4990 x108
815 546 6599 cell
More information about the asterisk-users