Re: [v6ops] I-D Action: draft-ietf-v6ops-reducing-ra-energy-consumption-02.txt

[James Woodyatt <[e-mail]>]

On Nov 5, 2015, at 16:32, Lorenzo Colitti <[e-mail]> wrote:
> On Thu, Nov 5, 2015 at 3:00 PM, James Woodyatt <[e-mail] <mailto:[e-mail]>> wrote:
> I’m opposed to that suggestion. The practice described in section 5.2 is necessary today for dealing with behavior of deployed networks that conform to existing standards. If 6MAN wants to clarify or improve the ND6 specification to change the need for the behavior documented here, then that would be splendid, but please count me on the list of people who prefer not to wait for the deliberations in 6MAN to resolve successfully before a practical solution to real world problems can be applied.
> 
> Erik, James and I worked on this together and I believe we're OK with the following text:
> 
>    1.  Maintaining IPv6 connectivity requires that hosts be able to
>        receive periodic multicast RAs [RFC4861]. Therefore, hosts that
>        process unicast packets sent while they are asleep MUST also
>        process multicast RAs sent while they are asleep. Hosts MAY
>        rate-limit identical RAs if they are sent too frequently.
> 
>    2.  Mobile devices that do not intend to maintain IPv6 connectivity
>        while asleep SHOULD either disconnect from the network, abandoning
>        all IPv6 configuration, or perform DNAv6 procedures [RFC6059]
>        when waking up.
> 
> Erik, James, can you confirm?

I confirm. I’m tempted to suggest that inserting some subclause indicators in the second paragraph might improve the clarity. Like so:

  2.  Mobile devices that do not intend to maintain IPv6 connectivity
       while asleep SHOULD either A) disconnect from the network, abandoning
       all IPv6 configuration, or B) perform DNAv6 procedures [RFC6059]
       when waking up.

—james