> From: Bill Fenner [mailto:[EMAIL PROTECTED]] [...] > >Yes, scopes have to be convex. This is not new in IPv6. > >See section 7 of the IPv4 scoped multicast RFC (2365). > > The worry is that people might look at routing protocols mechanisms > that keep the site convex wrt site-local addresses and consider > that sufficient. > > draft-ietf-ipngwg-scoping-arch-03.txt only says: > > o Each zone is required to be "convex" from a routing > perspective, i.e., packets sent from one interterface to > any other interface in the same zone are never routed > outside the zone. > > which does not seem to preclude my example topology, since the routing > protocol causes the site-local scope zone to be convex.
I believe it does preclude your example topology, since you have packets sent from one interface (H1's) to another interface (H2's) that are routed outside the zone. The fact that the destination address is global is orthogonal. -Dave -------------------------------------------------------------------- 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] --------------------------------------------------------------------
