[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Advancing the Protocol and Morin Drafts



In your last email you say "..Misunderstanding: the draft says "MUST"
for both PIM-based and BGP-based C-multicast routing."

How is that going to help you standardizing one solution?

Well, if an operator deploys equipments with compliant implementations
(implementing both BGP-based and PIM-based C-multicast routing), then he
knows that he will be able to activate the alternative it prefers, and
that it will interoperate.

What you are saying is not very different from what we document in our profiles. You just call it alternatives with the same MVPN solution set, we spell out the alternatives and call it profiles.



I also pointed out earlier that the Morin draft does not say which
core tree building protocol needs to be used. This also does not help
in getting an interoperable solution.

You possibly haven't read last revision.
Based on Eric comments about P-tunnels, the draft know says: "[the
recommandation is that] implementations implement the P2MP variants of
the P2P protocols that they already implement [..]".
Hence, if you have interoperable unicast tunneling techniques, the
recommandation orient implementations to also provide a set of
interoperable unicast tunneling techniques.

You can argue that this doesn't give *guarantees*, but I think we can
agree that it helps getting interoperability and that we can't
realistically do more. Moreover, RFC2547/4364 doesn't do thing much
differently and doesn't mandate either any specific tunneling
technology.

I disagree, because I prefer an approach that guarantees interoperability, that is the first priority of the IETF. The second priority is to converge on a single solution. You are limiting the building blocks and hope the end result will be interoperable.

Thx,

Ice.