Re: [v6ops] draft-byrne-v6ops-clatip-00

Is there a reason the 100.64.0.0/10 prefix reserved for transitional technology implementations would be inadequate or inappropriate to this task?

Owen

On Nov 5, 2013, at 9:48 AM, james woodyatt <jhw@conjury.org> wrote:

> On 2013-11-04, at 19:08 , Dave Michaud <Dave.Michaud@rci.rogers.com> wrote:
>> 
>> I think however, and this is my interpretation when I read the I-D, that the current intention is only to generalize a block already assigned for DS-lite so that it can be re-used for other transition technologies. The idea here is to provide an option for deploying 464XLAT. 
>> 
>> Mandating the use of this range may be inappropriate at this point but that is not what the current version of the draft does anyway.
> 
> I finally figured out why this draft made me say "huh?" when it was presented.
> 
> When I was working on the IPv4 dial-tone features a certain home gateway implementation, I ran into a very similar problem.  I needed some IPv4 addresses to number the PPP interface when it was in its "connect-on-demand" state.  These addresses would never appear on the wire, because they were only used for triggering the PPPoE Discovery protocol in response to "outbound" DNS query packets.
> 
> The way I dealt with this problem was to use the old 0.0.0.0/8 "unspecified" prefix from the old days before CIDR. These numbers have never been either global scope or routable. They're pretty much useless on the wire. I don't see why you'd need an RFC to use them for 464XLAT purposes.
> 
> --james woodyatt <jhw@conjury.org>
> 
> 
> 
> 
> _______________________________________________
> v6ops mailing list
> v6ops@ietf.org
> https://www.ietf.org/mailman/listinfo/v6ops