Re: [Last-Call] Last Call: <draft-ietf-6man-ra-pref64-07.txt> (Discovering PREF64 in Router Advertisements) to Proposed Standard - PLC consistency

[Alexandre Petrescu <[e-mail]>]

Le 06/11/2019 à 21:08, Jen Linkova a écrit :
> On Wed, Nov 6, 2019 at 6:19 PM Alexandre Petrescu
> <[e-mail]> wrote:
>> draft says:
>>> Scaled Lifetime field SHOULD by default be set to the lesser of 3 x
>>> MaxRtrAdvInterval divided by 8, or 8191
>>
>> If MaxRtrAdvInterval is 1 (RFC6275 'Mobile IPv6', radvd.conf) then the
>> division wont work, so the 'lesser' wouldnt be computed.
> 
> We can add the following text: 'if 3 x MaxRtrAdvInterval  is less than
> 8 then the Scaled Lifetime field SHOULD by default be set to 1',
> 
> On the side note: it looks to me that RFC6275 contradicts 'MUST' in RFC4861:
> 
> RFC6275: "Routers supporting mobility SHOULD be able to be configured
> with a smaller MinRtrAdvInterval value and MaxRtrAdvInterval value....
> MinRtrAdvInterval 0.03 seconds, MaxRtrAdvInterval 0.07 seconds"
> while RFC4861  states that  "MaxRtrAdvInterval ..MUST be no less than
> 4 seconds" and "MinRtrAdvInterval...MUST MUST be no less than 3
> seconds").
> 
> At the same time RFC6275 is not formally updates RFC4861..
> 
>> Then, by 'lifetime', in the cited text below and throughout the
>> document, one means the 'Scaled Lifetime' of this option, right?
> 
> It actually does not matter, as zero Scaled lifetime means zero
> lifetime, and non-zero Scaled Lifetime means non-zero actual lifetime.
> 
>>> Routers SHOULD check and compare the following information:
>>>
>>> o  set of PREF64 with non-zero lifetime;
>>>
>>> o  set of PREF64 with zero lifetime.
>>
>> Finally, for my curiosity, I wonder what kind of algorithm for checking
>> is consideredin each case.  Is it a precise checking, or more relaxed?
>> For example, how does it compare a /96 to a /64 (with a precise checking
>> they cant match, but with a longest match checking they could); would an
>> absolute or a partial majority in the set be needed to win the check, etc.
> 
> 2001:db8::/96 and 2001:db8::/64 are two different prefixes IMHO.
> 'Longest prefix match' etc applies when you are finding a prefix a
> particular IP address belongs to, not when you compare two prefixes.

Ok.

And with respect to the question of how many members in a set must be 
equal in order to declare consistency?  All of them?

Alex

> Also please note that it's just a check for configuration consistency
> within a LAN/PvD.
> 
>>>
>>> Abstract
>>>
>>>
>>> This document specifies a Neighbor Discovery option to be used in
>>> Router Advertisements to communicate NAT64 prefixes to hosts.
>>>
>>>
>>>
>>>
>>> The file can be obtained via
>>> https://datatracker.ietf.org/doc/draft-ietf-6man-ra-pref64/
>>>
>>> IESG discussion can be tracked via
>>> https://datatracker.ietf.org/doc/draft-ietf-6man-ra-pref64/ballot/
>>>
>>>
>>> No IPR declarations have been submitted directly on this I-D.
>>>
>>>
>>>
>>>
>>> _______________________________________________ IETF-Announce mailing
>>> list [e-mail]
>>> https://www.ietf.org/mailman/listinfo/ietf-announce
>>>
>>
>> --
>> last-call mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/last-call
> 
> 
>