- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
|
|
|
From: David Barak (no email)
Date: Sun Jul 01 2007 - 00:14:49 EDT
--- JORDI PALET MARTINEZ <>
wrote:
>
> But as said, IPv6 was designed having in mind a
> smooth transition including
> dual-stack. Nothing is wrong when IPv6 "alone"
> doesn't work today. Is like
> trying to use only gas in an engine that requires a
> mix of gas and oil. It
> is something wrong ? No, it is the way you try to
> use the engine, because
> was not designed that way !
The problem is that turning on v6, while requiring
that v4 continue to work means accepting the
limitations and security risks of both protocols.
This is not a "transition" - this is another level of
indirection (c.f. RFC 1925). A "transition" has an
end-state which is clearly defined, and we are only
just starting to ferret out the end-state with regard
to v6.
> In fact, I have not talked about public IPv4
> addresses at all ! As explained
> in another message, we are doing large IPv6-only
> deployments (5.000 sites).
> The "only" applies to the core and access network,
> but we keep
> net10+NAT+IPv6 in the LANs.
That's what you mean by "IPv6-only"? When I talk
about IPv6-only, what I mean is "no other layer-3
protocols running: no IPv4, no Appletalk, no IPX,
etc."
I get that there is rough consensus. I'm waiting for
the running code.
-David Barak
David Barak
Need Geek Rock? Try The Franchise:
http://www.listentothefranchise.com
____________________________________________________________________________________
Looking for earth-friendly autos?
Browse Top Cars by "Green Rating" at Yahoo! Autos' Green Center.
http://autos.yahoo.com/green_center/
|
|
|