On Sun, 28 Dec 2003 [EMAIL PROTECTED] wrote: > > We know that EDNS0 works. Mark did a good description of the present > > state. I see nothing to the contrary. It is A Good Thing to endorse > > EDNS0 in a document like this, because it could speed up deployment. > > I am not an expert with EDNS0, but reviewing this thread, it seems > that the client support for EDNS0 is what is lacking, most servers > support it already. What I think this means is that a SHOULD in > the requirements shouldn't cause operational difficulties. If I > am correct, then, what would be the problem with listing EDNS0 > as a SHOULD? If we do this, we may get more clients to support > EDNS0.
As practically zero stub resolvers enable EDNS0 by default (I don't know of any) and only a few implement it at all, I think it is premature to say that we know it "just plain works" (without any adverse effects etc.). I don't think we have much real deployment experience, from the node perspective, yet. -- Pekka Savola "You each name yourselves king, yet the Netcore Oy kingdom bleeds." Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
