Il martedì 20/04/10 00.31, Laszlo ha scritto:
Hi,
2010/4/19 Roberto Ovani <[email protected]
<mailto:[email protected]>>
Hello,
i have :
1) one host with opensips installed (opensips.roberto.com
<http://opensips.roberto.com>) on one public ip
2) one host with rtpproxy installed (rtpproxy.roberto.com
<http://rtpproxy.roberto.com>) on another public ip
3) two clients into another LANs...
I forwarded the ports for making all work :
1) opensips : 5060 udp (for the registering in sip)
2) rtpproxy : 7890 udp (the socket), and the range 10,000 : 20,000
(i compiled with this range)
when on sip client wants to call another sip clients (they're both
behind different nats and into different lans), then opensips must
rewrite the sdp so that the rtp flow goes through rtpproxy: in
this moment, opensips logs tell me :
/nathelper:force_rtp_proxy_body: incorrect port 0 in reply from
rtp proxy
/and the communication cannot happen correctly...... any advice ?
what shoul I have to check ?
Roberto
That row means no such session.
Do you see other rows like
- message body has zero length
- can't extract body from the message
?
Do you have force_rtp_proxy on INVITE?
-Laszlo
_______________________________________________
Users mailing list
[email protected]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
From /var/log/syslog I can't read neither "message body has zero
length" nor "can't extract body from the message".
In my opensips.cfg there's a route, route6 :
route[6] {
#---- RTP Proxy handling ---#
if (is_method("BYE|CANCEL")) {
unforce_rtp_proxy();
}
else if (is_method("INVITE")){
#---- Activates the RTP Proxy for the CALLEE ---#
force_rtp_proxy();
t_on_failure("1");
};
}
so, when an UAC is checked as "natted" (via flags 5 and 6), then the
flow enters into route6.
Roberto
_______________________________________________
Users mailing list
[email protected]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
