Rob, Rob Austein wrote: > > At Tue, 30 Apr 2002 08:07:14 -0400, Brian Haberman wrote: > > > > Actually that is an incorrect statement. The IPv6 addressing > > architecture forbids the use of an anycast address as the source > > address. So, the response back from the anycast member will have > > one of its unicast addresses as the source address. So, it is > > similar to your multicast request/unicast response model. > > Sorry, back up. I assumed that this proposal intended to change that > restriction, since with that restriction in place the resolvers won't > be able to match up the responses they get with the queries they sent > and anycast DNS won't work at all. Perhaps this week's version of the > proposed solution for this is something that walks and quacks like an > anycast route but is called something else so that you can use it as a > source address. I don't really care either way, it's still an anycast > route in all but name.
Section 9.2 of draft-ietf-ipv6-dns-discovery-04.txt states: Since the destination address may be an anycast address, the reply will necessarily come from a different address. The host must not discard the reply simply because the source address is different. A more detailed discussion of this issue can be found in [ANYCAST]. So, in my reading it is not proposing to lift the restriction, rather engineer around it. Brian -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
