0

I want to use 2 interfaces one eth0 one tun0(ovpn).

client <-LAN-> NAT(router) <-Internet-> opensips <-OVPN-> asterisk <-LAN-> client <---eth0-----> <-----tun0----->

The opensips is running with mhomed=1 and a rtpproxy is running on the opensips host. I want that all media traffic runs thought the rtpproxy. Also, I'm using NATHelper to resolve NAT traversal issues. The config was already working, but only for one interface. Could anyone help me to resolv my problem?

you can see my code here: http://opensips-open-sip-server.1449251.n2.nabble.com/Opensips-Nat-RTPProxy-in-bridged-mode-td7588578.html

Thanks in advance.

CR7
  • 3
  • 1
  • 3

1 Answers1

1

I have implemented a couple of opensips servers + rtpproxy using the configuration given by this book: "Building Telephony Systems with OpenSIPS 1.6 by Flavio E.Goncalves" Take a look of it, you will find all info you need for your deployment.

Regardless of that, I noticed a couple of issues in your config: 1) -u opensips:opensips -l 10.10.0.1/[publicIP] << you are configuring your RTPProxy in Bridge mode since you are setting two IP addresses.

2) rtpproxy_answer("o"); Bridge mode, requires for you to specify the i/e parameter as described below:

i/e - when RTPProxy is used in bridge mode, these flags are used to indicate the direction of the media flow for the current request/reply. 'i' refers to the LAN (internal network) and corresponds to the first interface of RTPProxy (as specified by the -l parameter). 'e' refers to the WAN (external network) and corresponds to the second interface of RTPProxy. These flags should always be used together. For example, an INVITE (offer) that comes from the Internet (WAN) to goes to a local media server (LAN) should use the 'ei' flags. The answer should use the 'ie' flags. Depending on the scenario, the 'ii' and 'ee' combination are also supported. Only makes sense when RTPProxy is running in the bridge mode.

Hope it will help you

JesusT
  • 11
  • 1