Re: [BEHAVE] Review of draft-boucadair-behave-dns-a64-01

["Dan Wing" <dwing at cisco.com>]

> -----Original Message-----
> From: behave-bounces at ietf.org 
> [mailto:behave-bounces at ietf.org] On Behalf Of Andrew Sullivan
> Sent: Monday, November 09, 2009 6:00 PM
> To: behave at ietf.org
> Subject: Re: [BEHAVE] Review of draft-boucadair-behave-dns-a64-01
> 
> On Mon, Nov 09, 2009 at 09:33:44AM +1100, Mark Andrews wrote:
> > 
> > One could also publish prefixes via DHCP and insert them into to
> > address selection algorithms to de-preference these addresses.  This
> > should work for internal use and a ISP supplying a NAT64.  It also
> > provides a mechanism to push DNS64 back into the 
> application for when
> > DNSSEC is in use.
> 
> This wouldn't solve the problem of any random site publishing DNS64
> records, which is supposed to be one of the big advantages of the
> draft.

(Contrary to what the draft says,) I don't think a site, publishing an AAAA
record that points to their own translator (to cover the "the IPv6 Internet to
an IPv4 network" scenario), truely cares their translator will be "over-"used
by dual-stack nodes.  Dual-stack didn't happen (to any reasonable degree of
"happen") and due to v4 exhaustion we're going to have clients doing some sort
of BEHAVE scenario or clients doing some sort of Dual-Stack Lite scenario.

I had envisioned A64 being primarily useful for an ISP's own services, with
clients that it can control (e.g., mobile phones, DNS proxies built into CPE
devices).

-d


> 
> A
> 
> -- 
> Andrew Sullivan
> ajs at shinkuro.com
> Shinkuro, Inc.
> _______________________________________________
> Behave mailing list
> Behave at ietf.org
> https://www.ietf.org/mailman/listinfo/behave