Re: [Sip] problem with NAT

[Ryan Kereliuk <ryker@ryker.org>]

On Tue, May 27, 2003 at 10:13:25AM +0530, Sujith Kumar wrote:
>
>             The sip registration should have a time out and after the time 
> out, the sip client needs to register again. The SIP register timeout 
> should be less than (or equal) the Transition table entry time out.
> I hope, this will solve your problem,.

Your comment is generally correct assuming a UA that uses UDP and supports
both symmetric signalling and very short REGISTER expiry times (and once
you add media into the picture it gets more complex).  Interop experience
shows that quite a few UAs don't support one or the other.

The original poster asked about TCP, however.  Unlike UDP, the NAT/PAT
bindings for TCP can be (and often are) torn down when the TCP state 
machine reaches a final state.  Unless you're using persistent and   
reusable connections you're going to run into trouble.

Let me point out the ICE draft, draft-rosenberg-sipping-ice-00.html,
which AFAIK is the latest attempt to reconcile SIP with the realities
of NAT/PAT.  I'm not sure what kind of traction it's getting but I think
it will be quite a while before something like this is part of common
SIP implementations.

_______________________________________________
Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use sip-implementors@cs.columbia.edu for questions on current sip
Use sipping@ietf.org for new developments on the application of sip