Hi Jari, On 11-10-18 02:29 PM, Jari Arkko wrote: > The last call ended yesterday. We are fine with going ahead as proposed > standard, but there were two other issues raised during discussion. > > 1. Possible draft to update RFC 5453 / 5342 to say that allocations in either > one should not conflict with each other. I tend to agree with Thomas that we > are unlikely to see much use of these registries and we should not worry too > much about it. That being said, if Suresh finds enough time to write a draft > on this I wouldn't mind shepherding it forward. In any case, it is a separate > matter from Sri's draft.
I think this would be relatively simple to fix. We just need to add a new range entry into the IID registry to mark all of IIDs based on the IANA block as reserved and point to RFC5342. I can do this pretty quickly once the draft submission reopens. > > 2. Whether to allocate an EUI-64 from the IANA block and base the IID on > that, or to allocate just a reserved value per RFC 5453. Collisions are > extremely unlikely in either case. Personally, I'd prefer an EUI-64 based > approach though, because collisions based on random addresses are then ruled > out completely, and only manual or multiple MAG type collisions may occur. > That is a personal preference though. But we need to choose. Are we going > with the draft as is, or changing it to use EUI-64 allocation? I saw Suresh > support this approach, Sri had questions (were those answered and what was > your conclusion?), what about the rest of you? I am personally fine with either approach. Thanks Suresh -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
