RE: IPv6 / NAT

["Tony Hain" <alh-ietf at tndh.net>]

See RFC 2373  2.5.8 Local-Use IPv6 Unicast Addresses


-----Original Message-----
From: Kyle Lussier [mailto:lussier at autonoc.com]
Sent: Thursday, February 15, 2001 8:57 PM
To: Michael W. Condry
Cc: ietf at ietf.org
Subject: IPv6 / NAT


> Well the message I got earlier was the IPv6 will not fix
> the NAT problem - true or not true?  I assume
> with IPv6 there is no need for NATs. Who thinks
> they will still be around - humm maybe if the ISP charge
> a fortune for 4 IP addresses vs 1 IP address (IPv6 or IPv4).

I think what we need is the ability to provide for NAT like
functionality in a logical / theoretical sense in the IPv6
namespace, but without the "physical action of translation".
I.e., we need a logical construct that resides on IPv6 global
space that is mobile.

Why would you want this?  What problem is there to solve?

It was raised by a very sharp person a little while back on
this list, specifically the ability to switch providers without
consequences.  We need a logical / functional mapping  or
construct on top of IPv6 that allows a company to "move
it's entire self around" in the IPv6 namespace.

What immediately comes to mind, is that IPv6 should have some
kind of "relative addressing" capability, where a company
can build a network on the relative space, but move it at
a whim if they switch providers, or for any other purpose.

My point / the difference in this suggestion from NATs is
that it should be logical and defined on IPv6 requiring
no actual translation.

In summary, IPv6 should support absolute addressing as well
as relative addressing, and even indexed addressing as
primitive IPv6 operations.

Kyle Lussier
www.AutoNOC.com