Hi, Paul,

You had a very good question - a question I wanted to ask but was afraid I would
be considered too careless in IPv6 specs. Seems 'site' in IPv6 is an elastic
concept - it all depends on how big your heart is ;-) . But in reality when I
tried to learn to configure the 6Bone connection, I was baffled too.
Hope you can give us a concise and practical definition very soon.

Best regards.

Kevin

Paul Francis wrote:

> When I presented the near-unique stuff in IETF50 yesterday, Deering stated
> from the mike that a site is a location in space (geographical region,
> whatever...I don't remember Steve's exact wording).  This had me totally
> baffled---and it isn't fun to be baffled in front of a couple hundred
> people.
>
> It had me baffled because it doesn't make a lot of sense (it begs certain
> nonsensical questions like, how big can it be (in square meters), if you
> have five routers, one each in New York, Maimi, Chicago, LA, and San
> Francisco, all connected by dedicated links, with only a single ISP
> connection point, is this a site, or is it too big or too sparse to be
> considered a site, etc. etc.)?
>
> Given that this statement came from somebody as authoritative as Steve, and
> that nobody disagreed with it, I naturally assumed that I must have hastily
> glossed over some important piece of text somewhere, and felt embarrased for
> apparently not having done my homework and for having such a fundamental
> mis-understanding of IPv6 (and it isn't fun to be embarrased in front of a
> couple hundred people).
>
> So I started doing some homework.  There is nothing about sites in 2460.
> There is nothing that defines sites in draft-ietf-ipngwg-addr-arch-v3-03.txt
> (though it talks a lot about sites).  draft-ietf-ipngwg-site-prefixes-05.txt
> has a section called "What is a Site?".  This section starts by saying that
> it "does not attempt to define the concept of a site", but then goes on to
> say "A site is an administratively controlled piece of topology that is well
> connected".  This is much closer to what I was thinking (though I probably
> would have defined it as a connected piece of topology whereby SLA
> assignment is coordinated or some such thing).  Erik's definition in any
> event clearly has nothing to do with geography.
>
> I could continue to wade through IPv6 specs to find the definition of site.
> But at this point I don't think the onus is on me to do so.  Rather Steve I
> think the burden is on you to point me to this illusive piece of text that
> defines what a site is.
>
> Thanks,
>
> PF
>
> --------------------------------------------------------------------
> 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]
> --------------------------------------------------------------------
begin:vcard 
n:Wang;Kuanjing
x-mozilla-html:FALSE
org:Motorola, Inc.;NAT, NSS
adr:;;1501 W. Shure Dr.;Arlington Heights;IL;60004;USA
version:2.1
email;internet:[EMAIL PROTECTED]
title:S.E.
fn:Kevin Wang
end:vcard

Reply via email to