RE: In consideration of I-D.templin-6man-rio-redirect

Hi Alex,

> -----Original Message-----
> From: ipv6 [mailto:ipv6-bounces@ietf.org] On Behalf Of Alexandre Petrescu
> Sent: Wednesday, July 19, 2017 2:25 AM
> To: ipv6@ietf.org
> Subject: Re: In consideration of I-D.templin-6man-rio-redirect
> 
> 
> [...]
> > Question: Why are you using NS/NA for the RIO confirmation lockstep?
> > Why not use RS/RA instead?
> 
> I subscribe to this question.  Especially if the 'Target' in the figure
> is a Router, and especially if we talk about Target-to-Target communication.

Target is a node that receives a Prefix Delegation. The node can then act
as either a host that uses the prefix for its own multi-addressing purposes
or as a router that forwards packets on behalf of nodes on its downstream
attached networks. Our draft refers to these nodes as "Type D" hosts.

> In IPWAVE WG we develop a problem statement about this, for vehicle to
> vehicle communications.

This is for peer-to-peer communications so that two peer nodes on the link
can communicate directly without having to go through a default router. The
peer-to-peer communications are established through an NS/NA exchange
which may be initially triggered by a Redirect. 

> There are also some I-Ds with solutions doing this with RA extensions.

Our concept follows directly from RFC6706. But, we recognize the need
for involving RIOs in other IPv6 ND messages than just Redirects. We do
not see a need for any RA extensions.

Thanks - Fred

> Alex
> 
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> ipv6@ietf.org
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------