> I would assume that any type of tunneling would be allowed (e.g., GRE), > and agree that it would be useful to enumerate at least general > situations where the SHOULD can be ignored.
Agreed. > I perused draft-hui-6man-rpl-headers and it appears to describe how to > insert/remove the RPL routing header at the RPL network border. Would > you prefer if the exception alluded to above via reference to that draft > be explicitly included in this draft? I don't immediatly see the value of the draft-hui-6man-rpl-headers at all. It seems incomplete. Is this intended to be standards track? Seems to me, there there are some (important) gaps. As such, the document isn't useful. For example, Section 3 talks about the mechanics of how to restore a packet (by removing headers) in some cases. How on earth is a node supposed to know it needs to do this? How can it distinguish between a packet that was properly created by the originating node (i.e., is fully compliant with existing routing header rules) vs. one where an intermediate router took shortcuts because it didn't want to do tunneling? Is there an assumption this is all manually configured somehow? Or? Thomas -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
