RE: automatic tunneling and v6 interoperation

["Pascal Thubert (pthubert)" <pthubert@cisco.com>]

On the side of this discussion (: sorry for the bother :) Here's (yet
another) alternate possibility to enable IPv6 to traverse IPv4 PAT/NAT.
It applies to IPv6 Mobile Nodes and Mobile Routers only, but if you have
MIPv6 available, the rest of the way is pretty simple. I proposed it to
Nemo and the suggestion I received was actually to post it to v6Ops. So
here it is... Can you please consider:

http://www.ietf.org/internet-drafts/draft-thubert-nemo-ipv4-traversal-01
.txt

Thanks :)

Pascal

> -----Original Message-----
> From: owner-v6ops@ops.ietf.org [mailto:owner-v6ops@ops.ietf.org] On
Behalf Of Pekka Savola
> Sent: jeudi 13 novembre 2003 20:32
> To: Christian Huitema
> Cc: v6ops@ops.ietf.org
> Subject: RE: automatic tunneling and v6 interoperation
> 
> On Thu, 13 Nov 2003, Christian Huitema wrote:
> > > Could you describe how the Teredo host finds this relay (close to
the
> > > native host)?  Note that I was not describing the traffic from the
> > native
> > > host to a Teredo host.
> >
> > This is actually documented in section 4.1.5 of the Teredo draft.
The
> > Teredo client sends an ICMP request to the IPv6 node, through the
Teredo
> > server. The native host receives the request over the native
interface,
> > and sends an ICMP reply. The ICMP reply is routed to the nearest
Teredo
> > relay over native IPv6. (Not routed very far if the gateway is on
the
> > host, but that is not necessary.) The Teredo client receives the
reply
> > from the relay, and associates the native IPv6 address to the IPv4 &
UDP
> > port of the relay. Further packets to that address are sent directly
to
> > this relay.
> 
> Ok, thanks -- missed that.
> 
> Who has the incentive of deploying Teredo servers then?  Assume there
will
> be 10,000,000 hosts implementing Teredo, and assuming that the users
surf
> the web, use the services, etc. and e.g. 1/10 of the traffic is going
> towards native destinations (could be a lot more as well).  Even if
this
> would result in one new site in a minute, wouldn't there still be
dozens
> or even hundreds of thousands ICMP packets passing through Teredo
servers,
> if they would not be distributed as well?  (The number of packets
would be
> quite high, and the associated state as well?)
> 
> --
> Pekka Savola                 "You each name yourselves king, yet the
> Netcore Oy                    kingdom bleeds."
> Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings
>