Looks like I forgot to call rtpproxy_manage() for BYE and CANCEL requests.

netstat –unlp showed a lot of port being used by rtpproxy even after the load test finished.

The ports weren’t being freed and that’s probably the reason it gave the error. It was running out of ports to use.

From: sr-users [mailto:sr-users-bounces@lists.sip-router.org] On Behalf Of Grant Bagdasarian
Sent: Wednesday, August 12, 2015 10:55 AM
To: Kamailio (SER) - Users Mailing List (sr-users@lists.sip-router.org) <sr-users@lists.sip-router.org>
Subject: [SR-Users] RTPProxy incorrect port 0 in reply from rtp proxy

Hello,

I’ve been load testing kamailio(4.2) with rtpproxy(2.0) using sipp, and when I hit 50+ CPS I start getting the following error: “incorrect port 0 in reply from rtp proxy”.

I’m running 4 instances of rtpproxy on the same instance as kamailio.

Could somebody tell me why I’m receiving this error? Do I need to run more instances of rtpproxy?

# <#----------RTPPROXY Parameters----------#>

modparam("rtpproxy", "rtpproxy_sock", "udp:127.0.0.1:7721 udp:127.0.0.1:7722 udp:127.0.0.1:7723 udp:127.0.0.1:7724")

In request_route:

                if (is_method("INVITE")) {

                               record_route();

                               if (has_body("application/sdp")) {

                                               rtpproxy_manage();

                               }

                }

In onreply_route:

                if (has_body("application/sdp")) {

                               rtpproxy_manage();

                }

rtpproxy -s udp:127.0.0.1:7721 -u rtpproxy rtpproxy -p /var/run/rtpproxy/rtpproxy.pid -l 10.0.0.1

rtpproxy -s udp:127.0.0.1:7722 -u rtpproxy rtpproxy -p /var/run/rtpproxy/rtpproxy1.pid -l 10.0.0.1

rtpproxy -s udp:127.0.0.1:7723 -u rtpproxy rtpproxy -p /var/run/rtpproxy/rtpproxy2.pid -l 10.0.0.1

rtpproxy -s udp:127.0.0.1:7724 -u rtpproxy rtpproxy -p /var/run/rtpproxy/rtpproxy3.pid -l 10.0.0.1

Regards,

Grant