IPV6 reverse DNS

Bill Manning bmanning@ISI.EDU
Thu, 14 Sep 2000 01:10:41 -0700 (PDT)

 In Bindv9rc5, use of A6 records will generate an assert failure -UNLESS-
the A6 records are in the degeneate case where they can not be distingusished
between AAAA records.  Bitstrings (DNAME) lables share the same fate.
Hopefully some corrections will be made before Bindv9 is actually released.
There are backard compatability issues w/ earlier resolvers.

% Please disregard my previous mail, I clearly did not have enough coffee.
% This was fixed, so I will try to reformulate:
% 1. Do some participants on the 6bone use A6 records (and bind9), in 
% parallel  to (or without defining) AAAA records? RFC2874 predicts that AAAA 
% should be replaced by A6 in the long run.
% 2. http://www.6bone.net/6bone_reverse_dns.html prones delegation of reverse 
% resolution using NS records, and under the ip6.int tree.
% Bind9 introduces a new way of delegating through DNAMEs. Is anybody out 
% there using it? As far as I can see, it may break queries, if a participant 
% tries to use it, as many servers (or clients) may not understand DNAME 
% answers to a PTR query.
% So, if this is the future, wouldnt it make sense that everybody switch 
% progressively to bind9?
% Or do the majority not believe in A6 and DNAME records?
% Cheers,
% DH.
% ___________________________________________________________________
%              * *         David Harmelin  	Network Engineer
%            *     *				DANCERT Representative
%           *              Francis House
%          *               112 Hills Road       Tel +44 1223 302992
%          *               Cambridge CB2 1PQ    Fax +44 1223 303005
%       D  A  N  T  E      United Kingdom       WWW http://www.dante.net
% ____________________________________________________________________