% From: Bill Manning <[EMAIL PROTECTED]>
%
% | Why shouldn't an address be in the DNS? And how can you tell
% | that an address, deprecated for your particular instance, has
% | not been re-issued and is valid for some other system?
%
% Bill, you're attempting to be overly pedantic - you know that isn't
% what Rich meant, but in this case, you actually goofed as well...
%
% A deprecated address is still assigned, it can't be re-issued yet,
% rather it is in the last stages of its validity to its previous
% assignee - but not yet revoked.
%
% So, deprecated addresses shouldn't be in the DNS - the time line should
% be to receive notification of an address change, start populating new
% addresses through the affected systems, update the DNS for those systems
% as their new addresses become available so the new ones are used instead
% of the old ones, then once that is done, mark the old addresses as deprecated
% for however long is possible to allow existing connections to continue
% to work, then finally, remove support for them altogether. Once that
% has happened the addresses can be returned, and then re-issued to someone
% else (but by that stage, they're no longer deprecated).
%
% kre
Ok, I goofed. By what process are addresses marked as "deprecated
for however long is possible to allow existing connections
to continue to work, then finally, remove support for them
altogether"? I still don't see why there is the strong requirement
to tie an addresses validity to its instantiation in the DNS.
--bill
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
