On Fri, 8 Feb 2002, Steve Deering wrote: > At 3:54 PM -0500 2/8/02, Keith Moore wrote: > > > >So if an ISP came along and said "we have a million customers signed up, > > > >we want to give them static /48 prefixes to their current home xDSL lines, > > > >and thus we'd like a /23", that should be approved? (2^25^0.8 ~= 1M) > > > > > > Yes, absolutely, assuming they can present some evidence that they > > > really do have that many customers. > > > >does the decision to use ASCII hex labels (rather than bitstrings) in ip6.arpa > >affect the allocation? would the RIRs be expected to dole out prefixes in > >4-bit increments in order to make DNS delegation for PTR lookups easier? > > I sure hope not. I have heard Randy and others claim that there is no > reason to impose such a restriction, regardless of the dropping of the > bitstring representation. Doesn't current IPv4 usage (which also doesn't > use bitstrings, and which doesn't restrict CIDR prefixes to be only > multiples of 4-bits long) provide a working example of how it can be done?
Allocations on non-nibble boundaries are possible of course.. it could just mean about 8 different almost identical delegations in the worst case. Or are you referring to "Class-less reverse delegation" (RFC2317)? I'm not sure if that'd help all that much. -- Pekka Savola "Tell me of difficulties surmounted, Netcore Oy not those you stumble over and fall" Systems. Networks. Security. -- Robert Jordan: A Crown of Swords -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
