Re: I-D Action:draft-ietf-v6ops-ipv6-cpe-router-03.txt

[Ole Troan <ot@cisco.com>]

Mark,

>>> I'm generally happy with the above text. I think the last sentence in
>>> (b) would be slightly better if it said "greater than" rather than
>> "different than".
>> 
>> Actually, "different than" captures the scenario where the PD is a
>> longer prefix than the hint indicated, in which case the CPE router may
>> not be able to number all of the interfaces in such a way as to enable
>> SLAAC.  What to do in this case is unclear, and we leave it up to the
>> vendor.
> 
> I'm not quite sure then I can see the value in this mechanism at all
> then. I've seen this mechanism as a "Hey ISP, I really need this amount
> of address space, otherwise I'm in trouble". If it's less
> significant than that, and the CPE will have to deal with a smaller
> prefix than requested (how is an interesting issue - interface
> priorities?), then I'm not sure I can see that much value in what it
> provides, other than some slightly useful information to the ISP about
> how many downstream interfaces their customers' CPE has. That might be
> useful to an ISP for addressing allocation purposes, although I'd think
> they're being unnecessarily conservative if they're counting how many
> individual subnets are required downstream of their customers' CPE.
> 
> Have I completely misunderstood the intent of this request/mechanism?

it is a "MAY" and a "hint".
a typical scenario would for example be that an SP gave a /56 regardless of the hint.

does the working group feel a need to specify the case where the CE router doesn't get a large enough prefix delegated? or are we happy leaving that up to implementations. alternatives are at least "various degrees of bailing out", "NAT66", "proxy ND"...

Best regards,
Ole