Re: 64share v2

The conclusion I personally came to after the PIO-X stuff was that I was
just trying to do it at the wrong layer.

I think the right way to do it is to have some information from the link
layer (or sub-IP layer) that tells the client that the RA is exclusive to
it.

In 3GPP, the architecture itself *is* that bit of information.  For wifi in
a /64 per host model (RFC 8273, which motivated the PIO-X work), it would
probably have to be some IEEE 802.11 bit somewhere.

On Tue, Nov 10, 2020 at 7:15 AM Ca By <cb.list6@gmail.com> wrote:

>
>
> On Tue, Nov 10, 2020 at 1:44 AM Lorenzo Colitti <lorenzo@google.com>
> wrote:
>
>> This looks pretty similar to the PIO-X bit which +Erik Kline
>> <ek@loon.com> and +Mikael Abrahamsson <swmike@swm.pp.se> wrote up in
>> 2017:
>>
>> https://tools.ietf.org/html/draft-pioxfolks-6man-pio-exclusive-bit
>>
>
> Interesting.
>
> Do you think we can achieve the goal without changing RA / adding a bit ?
>
>
>> I can't remember what happened to that. IIRC one of the primary
>> difficulties was that the scheme needs a strong guarantee from the link
>> layer that the RA is received by only one host. The minutes from IETF 98
>> <https://datatracker.ietf.org/doc/minutes-98-6man/00/> suggest that
>> there was also a concern that it duplicates existing functionality in
>> DHCPv6 PD.
>>
>
> I believe constraining this to 3gpp makes this achievable, as the link
> does not even run DAD. Duplicating PD functionality is a feature, not a
> bug.  As Joel pointed out, the ietf has already agreed to have 2 approaches
> for assign addresses. Adding DNS to RA has been a good thing.
>
> Also there was a question about the router being able to authenticate the
>> client.
>>
>
> In 3gpp, this is assured by low layer technologies
>
>
>> Maybe time to revive this?
>>
>
>> On Tue, Nov 10, 2020 at 5:10 PM Ca By <cb.list6@gmail.com> wrote:
>>
>>> Folks,
>>>
>>> In an effort to progress the conversation, i created a simple and rough
>>> pre-00 i-d (as ietf is not accepting submissions now) for your review and
>>> comment
>>>
>>> https://pastebin.com/duyYRkzG
>>>
>>>
>>>
>>>
>>> --------------------------------------------------------------------
>>> IETF IPv6 working group mailing list
>>> ipv6@ietf.org
>>> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
>>> --------------------------------------------------------------------
>>>
>>