On Tue, 26 Nov 2002, JINMEI Tatuya / [ISO-2022-JP] 神明達哉 wrote:
> >>>>> On Mon, 25 Nov 2002 19:42:53 +0200 (EET), 
> >>>>> Pekka Savola <[EMAIL PROTECTED]> said:
> 
> > There seems to be an assumption that provably unique SL addresses would be 
> > needed.
> 
> Perhaps this is a silly comment, but excuse me, I remember a very
> similar proposal by Paul Francis at an IETF meeting a few years ago.
> At that time the wg should have rejected the idea, but I didn't
> remember the details about the discussion and the reason why the idea
> was rejected.  

Thanks for the pointer.

> Anyway, we may be able to learn something from the
> experience (not to waste time, not necessarily to reject the idea
> again).  I'm sorry, but I don't have time to dig the minutes or ML
> archives on the discussion for now.

It's definitely worthwhile to learn from the past.

The draft is available at:
http://www.join.uni-muenster.de/drafts/draft-francis-ipngwg-unique-site-local-00.txt

The meeting was March 2001.

Text from minutes 
http://playground.sun.com/ipv6/minutes/ipng-minutes-mar2001.txt :

====
IPv6 Near-Unique Site-Local Addresses / Paul Francis
   <draft-francis-ipngwg-unique-site-local-00.txt>
----------------------------------------------------

  [See slides at http://playground.sun.com/pub/ipng/html/meetings.html]

Proposing to make site local prefixes non-local by adding a random number
to "zeros part" in site local addresses.

Now thinks that this doesn't help with the main problems stated in the
draft.  When merging sites, half will still have to renumber.  Not any
different from using the current site local prefix.  In both cases some
of the subnets will have to be renumbered.

Crawford:  Not quite as bad.  When merging each half could maintain their
original unique prefixes.  

Deering noted that notion of sites is not whole companies.  More like a
notion of a campus.  He also thinks we should think about non-routable
global address space.  Thinks there may need for that.  Something like a
special TLA that ISP's will never route in the global Internet.

Alain Durand, think this is unnecessary as we have lots of global
addresses and we don't need to create way to use local addresses for
non-local communication.

Additional discussion.
=====

But, when merging two sites, you don't need to renumber unless you want 
to.

-- 
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]
--------------------------------------------------------------------

Reply via email to