Thus spake Iljitsch van Beijnum
On 12 nov 2007, at 15:53, JORDI PALET MARTINEZ wrote:
If that happens (RIRs not having policies for using ULA), then IETF
can ask IANA to create a new registry for that, and IANA need to
do it, no choice against that IETF decision.
I object to the assumption that ULAs would be subject to RIR
involvement.
I presume you mean ULA-C/G, not ULA.
ULAs (RFC 4193) are deliberately unregistered.
RIRs deal with ISPs and the addresses ISPs use on the internet.
The holders of these addresses aren't ISPs and can't even be
necessarily assumed to be a customer of an ISP, and the
addresses aren't used on the internet. As such, it would be
unnatural to involve the RIRs here.
I haven't surveyed all of the RIRs, but at least one (ARIN) has explicit
policy allowing registration of addresses _not_ for use on the public
internet, either in isolation or as part of completely private network(s).
One is encouraged to use private/local addressing in such cases when
feasible, but not required.
Since PIv6 has happened, the _only_ remaining justification for ULA-C/G is
that it can be handed out to anyone who asks without meeting any bar for
"need" (based on DFZ slots), because ULA-Xs contains an evil bit in the
address that says "I hope nobody routes this". However, if LISP gets
anywhere, there is no reason for the RIRs to enforce _any_ bar on PI
addresses (v4 or v6) because they won't need to show up in the DFZ anymore.
As it stands, ARIN's rejection rate for PIv6 requests is under 3%; is
solving this "problem" really worth our time, or should we be focusing on
things like LISP that solve _real_ problems for _many_ people?
S
Stephen Sprunk "God does not play dice." --Albert Einstein
CCIE #3723 "God is an inveterate gambler, and He throws the
K5SSS dice at every possible opportunity." --Stephen Hawking
--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
