More generally, I still don't see why there is a restriction on the prefix length for all IPv6 unicast addresses where the first 3 MSBs are other than 000. I could understand the wording in RFC 3513 (and RFC 3513bis) if the restriction was intended for "unicast addresses that are configured via stateless address autoconfiguration" (thus my initial comment about the need to update RFC 2462bis). But some operators may want to use DHCPv6 (stateful address autoconfiguration) where there is no concept of prefixes per se (just 128 bit addresses). Therefore, as an example, if an operator wanted to have a /65 subnet(or some other subnet where the prefix is greater than /64) (and I am not saying that this is a good idea), at least the RFCs wouldn't prohibit it in the present/future. It seems like an unnecessary/unwise limitation IMHO.
Regards, Pete -----Original Message----- From: Brian Haberman [mailto:[EMAIL PROTECTED] Sent: Saturday, December 20, 2003 8:52 AM To: Stephen Sprunk Cc: Barany, Pete; [EMAIL PROTECTED] Subject: Re: Question about Interface ID length I would suggest reading RFC 3627. Regards, Brian Stephen Sprunk wrote: > Thus spake "Barany, Pete" <[EMAIL PROTECTED]> > >>... RFC 3513 ... Section 2.5.1: >>----- >>"For all unicast addresses, except those that start with binary value >>000, Interface IDs are required to be 64 bits long and to be constructed >>in Modified EUI-64 format." >> >>... RFC 2526 ... Section 2: >>----- >>"For other IPv6 address types (that is, with format prefixes other than >>those listed above), the interface identifier is not in EUI-64 format >>and may be other than 64 bits in length; these reserved subnet anycast >>addresses for such address types are constructed as follows:" >>| n bits | 121-n bits | 7 bits | >>+---------------------------------+------------------+------------+ >>| subnet prefix | 1111111...111111 | anycast ID | >>+---------------------------------+------------------+------------+ >> | interface identifier field | >>----- > > > Different angle... > > RFC 3513 and RFC 2526 both seem to outlaw* the common and logical practice > of using /127 networks (and thus a 1-bit Interface ID) for point-to-point > and tunnel links, because the IID isn't long enough and there's no space for > the 7-bit anycast ID, respectively. Can we put some sort of exception for > /127 networks in the RFC updates? > > * at least in format prefix 001, which is what operators will want to use. > > S > > Stephen Sprunk "God does not play dice." --Albert Einstein > CCIE #3723 "God is an inveterate gambler, and He throws the > K5SSS dice at every possible opportunity." --Stephen Hawking > > > -------------------------------------------------------------------- > IETF IPv6 working group mailing list > [EMAIL PROTECTED] > Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 > -------------------------------------------------------------------- -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
