Authors, WG, One further comment on this draft.
In the IANA considerations section you request the creation of a new registry for the "ARP/ND Extended Community Flags octet". The registry has a number of unassigned Flags. However, the draft does not specify a registration policy for this registry. Please can you propose a suitable registration policy. See section 4 of RFC 8126 for information on the different policies. Thanks Matthew On 26/06/2019, 17:46, "BESS on behalf of Bocci, Matthew (Nokia - GB)" <[email protected] on behalf of [email protected]> wrote: Authors I am the document shepherd for this draft. Please see below for my comments. Please treat these as you would any other WG last call comments. Best regards Matthew General: Why do you say the EVPN ARP/ND Extended Community is optional? Don’t you always use it if you comply to the procedures in the draft? Also, I am not clear why in section 3 you say the PE will follow the procedures (which I interpret as a MUST), but there are no MUSTs in e.g. 3a. Please clarify in the draft. Please also check the capitalization of RFC2119 keywords throughout. For example, there are a few instances where a 'may' should be a 'MAY' Abstract and 1. Introduction This document proposes an OPTIONAL extended community that is advertised ... Change to 'this document defines an extended community..." Section 1.1 Terminology and Conventions: "Address Resolution Protocol (ARP) Requests or Neighbor Discovery (ND) - or Neighbor Solicitation (NS) - messages are replied locally by the PE," This text seems to mix up Neighbor Discovery (the process) with Neighbor Solicitation (the message). I would suggest deleting Neighbor Discovery from this sentence. 1. Use of the EVPN ARP/ND Extended Community A PE may learn the IPv6->MAC pair and its associated ND Flags in the management plane or snooping Neighbor Advertisement messages coming from the CE. s/management plane or snooping Neighbor/management plane or by snooping Neighbor _______________________________________________ BESS mailing list [email protected] https://www.ietf.org/mailman/listinfo/bess _______________________________________________ BESS mailing list [email protected] https://www.ietf.org/mailman/listinfo/bess
