Re: [v6ops] ULA draft revision #2 Regarding isolated networks

[Tim Chown <[e-mail]>]

On 27 May 2014, at 14:59, Ted Lemon <[e-mail]> wrote:

> On May 27, 2014, at 9:30 AM, Nick Hilliard <[e-mail]> wrote:
>> or use NAT.  I'm not saying this in order to throw fuel on an existing
>> fire, but simply because this is the reality for many organisations in the
>> ipv4 world, and I see little reason why it will change for ipv6.  The IETF
>> can make recommendations about whether it thinks this is a good idea or
>> not, but it is not productive to pretend that the elephant isn't in the room.
> 
> Right, the point is that if we provide advice on how to set up ULA networks so that future transitions of this sort do not require NAT, that's worth doing.   Actually, for the enterprise scenario, I think the advice should just be "get a GUA, use it like a ULA" because that excludes the possibility of a future clash when two behemoths merge.   But it makes source address selection harder.   And there was a document a while back about informal ULA registries, IIRC, which could also represent a good mitigation strategy if it were to happen (but I think that's outside our scope of work).

The enterprise IPv6 incremental deployment text is just going up for publication, see http://tools.ietf.org/html/draft-ietf-v6ops-enterprise-incremental-ipv6-05.
That pretty much captures what Ted says, but also points to the draft under discussion here.  See section 2.6 in particular.

Tim

> _______________________________________________
> v6ops mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/v6ops