[VOIPSEC] IPv6 and the demise (or not) ofNAT(wasRe: Interactive Connectivity Establishment (ICE))

[Hadriel Kaplan]

> -----Original Message-----
> From: Simon Horne [mailto:s.horne at packetizer.com]
> 
> Yeah, calling out to a PSTN from behind a symmetric NAT is supported in a
> lot of protocols, H.323 and IAX too. The problem is calling in, if we just
> say heck let''s not worry about that then we are really putting our head
> in
> the sand. My point, AFAIK, there is no standard method to do this in SIP
> or
> MGCP, there needs to be one.

Who's not worried about that?  The way SBCs do NAT traversal today does
support inbound calling for SIP and MGCP.  I just meant there's no spec
needed for their method because they don't change any protocols to do it -
the rub is you can only call users through/via the SBCs (well, that's not
totally true, but close enough).  

What the ietf is looking for, I think, is a way for users to get inbound
calling from anyone/anywhere through NATs - without having to go through a
stateful relay device, for signaling or media (unless it's a symmetric NAT).
So those are your standards.  STUN would be useful for all of us I think.
They didn't have to make it a new protocol, but I understand why they did.
Is TURN (or ICE) necessary?  No, not for service provider SIP deployments.
In fact, it makes things worse in some ways.  But it will help you get
non-voip-related NAT-breaking traffic through symmetric NATs.  (so when we
find that problem, we'll have a solution I guess) 

-hadriel