Re: [v6ops] I-D Action: draft-ietf-v6ops-unique-ipv6-prefix-per-host-07.txt

[Lorenzo Colitti <[e-mail]>]

On Sat, Aug 19, 2017 at 12:07 PM, Alexandre Petrescu <
[e-mail]> wrote:

> Le 19/08/2017 à 03:17, Lorenzo Colitti a écrit :
>
>> If the prefix length changes away from 64 bits it will be a major change
>> that will affect this document in many ways.
>>
>
> Change? there is no current spec currently requiring the prefix len to be
> 64Ș 4291 and 2464 talk IID len, not prefix len, right?


Yes, change is the right word. This is because:

   1. This document presumes the use of SLAAC.
   2. The prefix length has to be 128-IID length for SLAAC to work.
   3. RFC 4291 says that IIDs are 64 bits (except for ::/3, which cannot
   currently be assigned to any real deployment).

Therefore if the prefix length was not /64, no hosts would get any
addresses.

"currently /64" - does it mean it is what this deployment does it now? In
> that case is it maybe INFORMATIONAL rather than BCP?
>

/64 is what the deployment must do in order to operate. If it doesn't work,
then we don't document it in an operational working group.