On 11/29/2011 15:04, Mark Andrews wrote:
In message <4ED55726(_dot_)5090209(_at_)dougbarton(_dot_)us>, Doug Barton
writes:
On 11/29/2011 13:59, Chris Donley wrote:
And that's one of the reasons this draft updates 5735. If routers make
decisions as to whether or not to enable a feature based on whether
behind a public or private address, having a defined address range for
CGN space will be significantly easier to deal with than to have
arbitrary address ranges selected on a per-ISP basis.
But that's certainly not the only way to handle that problem, right? If
the router needs to be updated to recognize the new space anyway,
wouldn't it make more sense to update it with a more generic mechanism
to signal "You're behind a private address?" That way you can use the
same mechanism for IPv4 and IPv6.
It doesn't have to be one or the other.
Of course it doesn't have to be, it *should* be.
It can be both. Having address
space that the CPE can identify as non-public without the ISP having
to configure something is a a good thing.
How are the CPEs going to recognize the new space without something
being configured? They will need a software upgrade, right? So why is a
generic, boolean flag not a better idea than allocating a magic block?
As much as I would like to see IPv6 native everywhere as soon as
possible forcing each ISP to choose part of their allocated address
space or to use RFC 1918 address space behind CGN's is not good
resource management.
I don't see anything wrong with using 1918 space for this purpose.
RFC 1918 space is suppose to be used *within* a enterprise. Using
it to *connect* enterprises is out of bounds.
How are you defining enterprise in this context?
Doug
--
"We could put the whole Internet into a book."
"Too practical."
Breadth of IT experience, and depth of knowledge in the DNS.
Yours for the right price. :) http://SupersetSolutions.com/
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf