[OpenSIPS-Users] OpenSIPS / RTPProxy integration. Opensips in local networks.

Bogdan-Andrei Iancu bogdan at voice-system.ro
Tue Jun 15 12:11:59 CEST 2010


Hi José,

José María Jiménez wrote:
> Hi all,
>
> I'm testing the integration of OpenSIPS + RTPproxy and I have some doubts:
>
> 1) Is it possible to test a VoIP infrastructure (with OpenSIPS, 
> RTPProxy and IP-phones) in a local network? I'm working on the local 
> network of an university and it's difficult to me to work with an 
> public IP.
Sure it it. Typically you need to use rtpproxy for relaying media in 
case of NAT presence, but you can use it in any case.
>
> 2) Is it possible to chain directly 2 OpenSIPS+RTPProxy solutions (in 
> a local network)?
>
> IP-Phones <----- sip/rtp -----> OpenSIPS/RTPProxy <----- sip/rtp 
> -----> OpenSIPS/RTPProxy <----- sip/rtp -----> Asterisk
It is possible - but see the "f" and "r" flags for "force_rtp_proxy" :
       
http://www.opensips.org/html/docs/modules/1.6.x/nathelper.html#id271384

Those flags are required if you want to chain multiple rtpproxies.
>
> I'm getting some problems in the SDP body of the second 
> OpenSIPS/RTPProxy. I don't know if this wrong behavior is due to the 
> fact I'm working in a local network and functions liks fix_nated_sdp 
> doesn't work in this context. Can I modify the SDP body directly with 
> functions like subst_body() (from nathelper module)?
why using fix_nated_sdp() ? if you use rtpproxy, the force_rtp_proxy() 
will do all the required changes over the SDP (requests and replies).

To debug the problem check first if the SDP is properly changed:  if the 
requests / replies leaving your proxy do have in SDP the C= and M= line 
updated with the address of your rtpproxy.

Regards,
Bogdan
>
> Thanks in advance!
>
> José M.


-- 
Bogdan-Andrei Iancu
www.voice-system.ro




More information about the Users mailing list