Re: [manet] Topic for discussion?

[Rick Taylor <[e-mail]>]

Hi Lou,

I think this makes sense.  I believe multicast is on the charter, and
although DLEP has some things to say about multicast, it is
intentionally a bit vague in places.

If we can work out a few work items of value that the WG wants to work
on, then I'm all in favour.  

I presented some very early ideas about a DLEP data item to indicate
the link-layer suitability for being a multicast rendezvous point, not
exactly the same topic, but I'm interested in feedback.

Cheers,

Rick

On Mon, 2019-07-22 at 16:25 -0400, Lou Berger wrote:
> Hi,
> 	If there is time on the agenda I'd like to raise/have a
> discussion on 
> dlep support for IP multicast.  These questions come out of looking
> at 
> the opensource dlep implementation posted at 
> https://github.com/mit-ll/LL-DLEP and RFC8175.  The questions are:
> 
> (a) How does a multicast receiver identify itself (and remove itself)
> 
> (b) how does a (potential) multicast sender know about a multicast
> group
> and it's related parameters
> 
> (c) how does a (potential) multicast sender know what endpoints are
> reachable via a multicast group
> 
> I raised this same question with a number of people including the 
> RFC8175 authors and we generally agreed that there are answers for
> (a) 
> and (b) that can be traced easily/obviously back to RFC8175, and a 
> fairly straight forward answer to (c) that we don't think falls
> within 
> the intent of the RFC. (At least the authors of RFC8175 gave the
> same 
> answers that we separately arrived at.)
> 
> I think it would be great if we could review the questions and
> ensure 
> that we all agree on the answers.  I can through some slides together
> on 
> this, or we Stan/Rick could do the same. - Or we can start with
> Stan's 
> (if he prefers) email on the topic.
> 
> Thoughts?
> 
> Thanks,
> Lou
> 
> 
>