Re: [BEHAVE] NAT66 - lets bring back FEC0:: ( was Lack of need for 66nat)

I agree with Keith here.  In my earlier comments on the NAT66 draft,
I mentioned that in my opinion, it MUST NOT be used with scoped
addresses (link-locals, site-locals).  Communication to such
addresses should never be routed outside their scope (even
if translated).

And use with ULAs is questionable (might affect some applications
if they assume[*] that ULAs aren't used outside their site).

-Dave

[*] I use assume in the same sense as other assumptions/myths in
draft-iab-ip-model-evolution.

> -----Original Message-----
> From: behave-bounces@ietf.org [mailto:behave-bounces@ietf.org] On
> Behalf Of Keith Moore
> Sent: Thursday, November 27, 2008 8:21 AM
> To: Eric Klein
> Cc: Behave Mailing List
> Subject: Re: [BEHAVE] NAT66 - lets bring back FEC0:: ( was Lack of need
> for 66nat)
>
> Eric Klein wrote:
> > If we are seriously discussing defining some flavor of NAT (and it
> > appears that more than half of the people writing are in favor of it)
> > then we need to resurect the address space that was allocated for
> site
> > locals fec0:: (and currently is likely not used do to early
> > implementations) and use it as the new "1918" type pool for v6.
>
> uh, no.
>
> the one thing that makes NAT even worth considering in IPv6 is not
> having conflicting use of the same address space, like we have with RFC
> 1918 addresses in IPv4.
>
>
> Keith
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www.ietf.org/mailman/listinfo/behave

_______________________________________________
Behave mailing list
Behave@ietf.org
https://www.ietf.org/mailman/listinfo/behave