[RRG] Re: Knee jerking

[Brian E Carpenter <brian.e.carpenter@gmail.com>]

On 2008-03-26 10:31, Dino Farinacci wrote:
>> Here's the thing... IPv4 is already irremediably broken by
>> NAT (i.e. all upper layer designs must already allow for
> 
> I know what you mean, but saying IPv4 is broken is a bit strong. I'd
> argue you should thank NATs for allowing the Internet to scale to what
> it is today in terms of the number of hosts that can talk to each other.

Indeed. Although there is a dream world in which NAT didn't
happen and we got IPv6 deployed in 1998 because we had to...

> 
>> IPv4 NAT, or even worse NAPT). So if the LISP transition
>> for IPv4 sites includes NAT, it really isn't a new problem
>> for the upper layers.
> 
> Right.
> 
>> However, it would be sad if LISP required NAT for IPv6,
>> since the reality is that we're looking at this whole
> 
> Well since we have more address space, and it's not out of the question
> of getting a /8 or /16 of EID-space, PTRs look attractive.
> 
> Do you think so?

Yes, I think we put 128 bits in IPv6 so that we could use
them :-)

> 
>> class of solutions (and shim6) precisely to avoid the need
>> for NAT. In fact, although I don't fully agree with
>> draft-rosenberg-internet-waist-hourglass, it does point
>> us in the direction of needing a complete new identity
>> system at layer 4, if we end up breaking IPv6 with NAT too.
> 
> Don't you think a mass PI allocation scheme is going to break IPv6 if we
> don't do anything about it?

Absolutely. That's partly why we're here, isn't it?

    Brian

--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg