Hi, >>>But I suppose we are talking about networks where some B2BUAs will get >>>confused if you try the media from a different IP address from where >>>you do SIP (although if you have something like a >>>carrier grade NAT between the UA and the B2BUA there is no guarantee >>>that SIP and RTP would come from the same IP). Perhaps it would be good >>>enough to use two different instance id's? >> >>It doesn't have to be a B2BUA. It can be any kind of "gate control >>node" which will not allow media to pass unless there is an associated SIP >>session. > > >Yes, but I think such a node will need some information about SIP >sessions and that is given by a B2BUA-ish element (since it needs to dig >into SDP).
That is what I meant: the node controlling the gate need the SIP/SDP information in order to open the gates. But, if the signalling traverses some other node, in another network, which has no control over the gate where the media is sent, the gates won't be opened. So, you need to make sure that the signalling traverses the node which controls the gates where the media is to be sent. >But OK, the bottom line is that some folks claim they want to deploy SIP in such a way that media and signaling are strongly >associated like that. Do you know if a lot of them are *really* doing it too? I don't know. I am just saying that the signalling needs to traverse the entities which control the gates where the media is to be sent trough. Regards, Christer _______________________________________________ Sip mailing list https://www.ietf.org/mailman/listinfo/sip This list is for NEW development of the core SIP Protocol Use [EMAIL PROTECTED] for questions on current sip Use [EMAIL PROTECTED] for new developments on the application of sip
