>> So, just to ask... Suppose you have this: >> >> A---B---C---D >> | | >> +---E---+ >> >> 1. The existence of the link from B to E is a fact within the topology >> shown. > > nope, it is not. A is a peer of B and E is a peer of B, so there is no > path between B and E for A's prefixes. sorry charlie.
I can't even begin to make sense out of this --can you explain? If A/B, B/E, and E/D are valid peering pairs, then the path exist. If B chooses not to advertise some route from A to E, that's a local policy decision. If you modify the protocol so D can tell B didn't intend to send a specific set of destinations to E, then you've injected policy into the protocol. There's no way to escape the conclusion that BGPSEC injects policy into BGP itself. Russ _______________________________________________ sidr mailing list [email protected] https://www.ietf.org/mailman/listinfo/sidr
