[arin-ppml] Policy Proposal (Global): Allocation of IPv4 Blocks to Regional Internet Registries - Revised

Martin Hannigan martin.hannigan at batelnet.bs
Thu Mar 12 00:25:43 EDT 2009

On Wed, Mar 11, 2009 at 8:24 PM, Scott Leibrand <scottleibrand at gmail.com>wrote:

> Marty,
> Thanks for your excellent input here.
> I seem to recall that the last time we went through the global PDP (with
> the "last /8" policy), that we had several RIRs who liked the general idea,
> but required substantial changes (namely N=1 instead of 2 or more) to
> support it.  IIRC, some RIRs passed it first with a higher value for N, then
> had to go back and pass the N=1 consensus version after that was what passed
> in the other RIRs.

Correct. That was in fact a result of the differences between the regions. A
contextually similar version was required to be passed and the initial
version was not able to meet that requirement based on regional changes.

> I'm wondering if we couldn't do something similar here.  While I think some
> sort of policy along these lines might be beneficial (and we need
> *something* to tell the IANA what to do after IANA free pool exhaustion), I
> think that the policy as written is unworkable, for all the reasons outlined
> by others.

It's perfectly fine if the IANA v4 responsibilities are deprecated at
exhaustion. They will not be going away as a result and not all of their
functions related to v4 would be deprecated. As far as doing something
similiar to the /8 policy you referenced, yes, the AC absolutely could. I
don't know that this is the most efficient way to proceed, all considered.
It looks like the APNIC last call will have concluded before the ARIN region
is able to return feedback/decision to the authors based on a meeting
consesus requirement. If that happens post ARIN meeting, significant changes
require them to start over[1]. If there is a compromise to be made, it's
likely better to make it sooner than later. If the AC has the ability to not
accept the policy now and allow them to resubmit a revision, that may
actually be beneficial to all.

> So, what if the ARIN AC were to go ahead and revise the policy along the
> lines of what Marla outlined earlier?  Assuming that a less restrictive
> policy is all we'll have consensus for, wouldn't that just mean that APNIC
> would have to pass the less restrictive version as well before it could
> advance to the ASO?

Marla indicated that she thought that the language was fuzzy. I assert that
that the language is not ambigous. You could insert language to make it
voluntary, for example. That would require APNIC to revisit the policy per
their PDP[1].



1. That took two years to pass its multiple iterations due to the
differences in the N=X equation. This demonstrates that if the current
policy being discused is passed and then deemed harmful or perhaps illegal,
it will take that long to fix it, or, that long to learn that one RIR
community will hold out and refuse to grant consensus and there will be no
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20090312/fed84b08/attachment-0001.html>

More information about the ARIN-PPML mailing list