Re: Interaction between ARs in lmm domain (was Re: [netlmm] Re: [Mipshop] comments on Mobile IPv6 Fast Handovers over IEEE 802.16eNetworks draft)

James Kempf wrote:

> Even if DNA isn't used, the MN can use oDAD for the src addr and the pAR for
> the dest addr. In this case, the MN would have had to configure the new CoA
> using prehandover prefix information or prefix information derived from the
> specialized link support.
>

I think we are digressing from the main thread..

FYI: there is a draft specifying FMIP and DNA interaction that should
appear soon.

draft-koodli-dna-fmip-00.txt

A protocol (FMIP) should be able to work on its own, while also being 
able to work well when some other protocol is available.

The FBU can be constructed as above. But we still need a way to inform
NAR to flush the buffer.

> So, I wonder, is FNA really needed anymore for FMIP? Seems like it could
> really simplify the protocol to have it concentrate on FBU.

See above. In cases where we can assume that an announcement that makes
the NAR's ND state REACHABLE (for NCoA) is available, we can make do 
without it.

-Rajeev

> 
>             jak
> 
> 

_______________________________________________
Mipshop mailing list
Mipshop@ietf.org
https://www1.ietf.org/mailman/listinfo/mipshop