Date: Thu, 15 Feb 2001 12:36:25 -0800
From: "Paul Francis" <[EMAIL PROTECTED]>
Message-ID: <000e01c097d9$04efdda0$[EMAIL PROTECTED]>
| It is funny for you to say that with IPv6 we don't need to use re-used
| addresses, because a site local address *is* a kind of re-used address.
Yes, I know, as often occurs, my wording was not very good.
I agree with your summary of the way things are (or will be)...
| On a separate note, I know this is probably just digging up dead
| discussions, but have people considered creating a new IPv6 space which is
| "globally unique but not globally routable" for use in this sort of internal
| site communications?
Yes there has been a bunch of discussion about that from time to
time, it is a topic Christian Huitema raises from time to time...
There's no reason these need to be different from site locals
though (all the attributes are there - if only site locals could
have a field in them to add the "globally unique" part).
The chief objection to this, as I understand it, has always been
how such things would get assigned, who would do that, on what
basis, etc. I can't actually see the difficulty in assigning
consecutive integers to anyone who asks (perhaps using an automated
system much like many mailing lists use for subscriptions - you fill
in the template, send it in, get back e-mail containing a token/passwd
send that back, and your site ID gets returned).
There's 38 bits of empty space in site locals (250 billion numbers
close enough) which would do for this, and '0' could continue to
mean the "undetermined" (not globally unique) site local.
| Such an address space would be useful, for instance,
There are lots of uses - it would provide a stable I-D for security
associations, ...
| Such addresses (lets call them unique-local addresses) could be put in DNS
| without concern for thier being mis-interpreted by other nodes.
Yes, though they still add to the DNS payload when for the vast majority
of users (everyone non-local) they're useless information.
| Granted
| when a node did a DNS lookup, it wouldn't know if a given unique-local
| address was in the same site,
No, considering aggregated sites (with multiple I-Ds) but a simple compare
of her site ID token against mine would catch the majority of cases.
kre
--------------------------------------------------------------------
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]
--------------------------------------------------------------------