Re: v6ops and v6 connectivity deployment model

[Pekka Savola <pekkas@netcore.fi>]

On Wed, 12 Nov 2003, Harald Tveit Alvestrand wrote:
> --On 12. november 2003 15:59 +0200 Pekka Savola <pekkas@netcore.fi> wrote:
> > The main point of this deployment model for something like Teredo is that
> > it is assumed that the hosts behind NATs *primarily* want to communicate
> > with other hosts behind NATs (e.g., peer-to-peer apps) -- but other *IPv6*
> > hosts are not so important.
> 
> I don't get this.... will a Teredo host have difficulty communicating with 
> a native IPv6 host? If so, why?
> 
> I thought that was the problem it was supposed to solve - but I may have 
> missed something....

You'd think it's the goal of the document.. :-)

Teredo hosts will not communicate with native v6 hosts (which don't
additionally implement (part of) Teredo that is), unless they're
configured with a relay.

Native v6 hosts will not communicate with Teredo hosts, unless they can 
find a relay for the address used by the Teredo relay.  The current 
Microsoft deployments work because every host (even if deployed natively) 
implements and enables Teredo relay for itself.

The problem should be obvious.  Every v6 host has to implement Teredo 
relay or be 100% sure that such a relay is provided by the site or 
somewhere else in the network if they wish to communicate with Teredo 
hosts.

Remember, the Microsoft's reason for deploying Teredo was not to make them 
communicate with v6 hosts.  They're concerned with deploying a globally 
usable NAT traversal mechanism -- their deployment assumption is that 
Teredo relays are deployed on every node, so there is no problem 
communicating with "native" and "Teredo" hosts.

Hope this clarifies..

-- 
Pekka Savola                 "You each name yourselves king, yet the
Netcore Oy                    kingdom bleeds."
Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings