On Thu, Nov 25, 2021 at 11:03:16AM +0800, [email protected] wrote: > I have no objection to the comparison, please go on. > One thing I want to emphasize is that, whether DC use case (brought up by > draft-wang-bfd-one-arm-use-case) or broadband access use case (brought up by > BBF TR-146), the key requirement is that the peer system is totally > BFD-Unaware, in other words, the peer system would never send or parse any > BFD Control packets, from the very beginning.
This has been consistent with everything I've seen over the last several years and will be a fine requirement. (I had been asked about a similar mechanism almost two years before someone showed me the BBF document.) > Another thing is that the Unaffiliated BFD Echo can only work between the > local system and its one-hop-away peer system. For IPv4/IPv6 unicast, that's consistent with BFD Echo procedures. -- Jeff
