> -----Original Message----- > From: [email protected] [mailto:[email protected]] On > Behalf Of Rémi Després > Sent: 09 December 2009 11:06 > To: Xu Xiaohu > Cc: [email protected]; [email protected]; > [email protected]; 'Brian E Carpenter'; [email protected] > Subject: Re: [BEHAVE] address-format: bits 64 to 71 > > Xu Xiaohu wrote : > > > The following statement is quoted from RFC 4291: "IPv6 > nodes are not > > required to validate that interface identifiers created > with modified > > EUI-64 tokens with the "u" bit set to universal are unique. > > > > The use of the universal/local bit in the Modified EUI-64 format > > identifier is to allow development of future technology > that can take > > advantage of interface identifiers with universal scope." > > > > Till now, we haven't known the real usage of this > constraint yet. So > > we should reverently obey such a constraint whose future usage is > > still uncertain. > > Two points: > > 1. > This sentence doesn't concern addresses that, like > IPv4-mappable addresses, are never subject to any neighbor > discovery protocol. > > > 2. > Relevance of this sentence seems negated in RFC 2462 where, > in the specification of IPv6 stateless autoconfiguration, we > have (upper cases > added): > > "To insure that all configured addresses are likely to be > unique on a given link, NODES RUN a "duplicate address > detection" algorithm on addresses before assigning them to an > interface. The Duplicate ddress Detection algorithm is > performed on ALL addresses, independent of whether they are > obtained via stateless or stateful autoconfiguration." > > Indeed, not checking uniqueness of IPv6 addresses when they > are MAC-address derived would be a bad idea: MAC addresses > cannot be assumed to be really universal because they can be > administratively configured in some hosts.
Under the EUI-64 scheme, a non unique IPv6 address is the result of a non-unique MAC address. When having the latter in a shared bridge domain is a bit of a problem to say the least and doing or not doing IPv6 DAD doesn't change things. Where checking for uniquness makes sense is whenever there is some involvment of randomness or administrative fiat in configuring/creating the IPv6 address. Hence in a perfect world where the EUI-64 interface-id rules apply, checking for uniquness wouldn't be needed. In mandating the check the authors evidently thought of the cases where the EUI-64 rule doesn't hold. -Woj. > > > Look forward to a clear conclusion after discussing it in the 6man > > mailing-list. > > Agreed. > > Best regards, > > RD > -------------------------------------------------------------------- > IETF IPv6 working group mailing list > [email protected] > Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 > -------------------------------------------------------------------- > -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
