Hi Hitoshi, Pierrick, I have 2 questions:
1) how is the operation of MAGs using a M-Tunnel different from a MR deployed in a MAG using MTMA mode? I didn't go through both drafts in detail, so I may be missing something..
2) The second issue is more specific and is related to the correct operation of your proposal (which I would sayz also applies to section 4.3 of "Multicast Mobility Routing Optimizations for Proxy Mobile IPv6").
It is said in the draft that "The MAG selects only one upstream interface (either M-Tunnel interface or physical interface) for a multicast channel by the Reverse Path Forwarding (RPF) algorithm." It is correct that RPF will lead to the selection of a single interface for any multicast channel. Although, the M-Tunnel will not be selected following RPF unless one of two options is followed: 1) a specific route entry to the RP (RPT) or to the source (SPT) exists in the RIB, and it "points to" the tunnel interface; - I assume such a tunnel won't be running a routing protocol like OSPFv3, so I find this case very improbable / impossible (?) 2) MRIB is (manually) configured with an entry stating that the RP's (or source's) prefix is accessed via the M-tunnel; - in RFC4601 it is stated regarding MRIB that "The routes in this table may be taken directly from the unicast routing table, or they may be different and provided by a separate routing protocol such as MBGP". In my opinion, this gives space to feeding MRIBs from MAGs in a "novel" way for PMIPv6 environments..
Please correct me if any of my conclusions / observations are wrong, but my main point is, the exact way how the MRIB is filled should be described, as it is crucial to the correct operation of the proposal.
Best regards, Sérgio _______________________________________________ multimob mailing list [email protected] https://www.ietf.org/mailman/listinfo/multimob
