Thanks Alex.
Early tests after your change seem to indicate that fixed the problem. I'm looking at the docs with this new perspective, and when we read the description of the i flag and the e flag it seems to suggest that they are used independently. Nothing jumped out at us to use them together in the order that the packet was flowing from the outside to the inside. Thanks for your quick insight on this, i'm not sure I would have come up with it.
Stagg
On 10/6/10 6:45 PM, Alex Balashov wrote:
On 10/06/2010 06:41 PM, Stagg Shelton wrote:
I'm working on a setup where we have rtpproxy on a machine with eth0 IP 10.10.5.141/19 and eth1 IP 10.10.10.78/24. When using force_rtp_proxy("","10.10.5.141"); the connection information (c) field in the SDP is not being set to the ip 10.10.5.141 It is instead being set to 10.10.10.78
We are starting to run out of ideas on getting the proper contact address into the sdp based on call source signaling ip. Does anyone have any ideas what the problem could be?
Are you invoking rtpproxy in bridging mode?
If so, try one of:
force_rtp_proxy("ocfaie");
or the opposite direction:
force_rtp_proxy("ocfaei");