Re: [sig-policy] prop-086: Global policy for IPv4 allocations by the IAN
Louis Lee said the following on 26/08/10 08:00 :
> I apologize personally for the delay in my response. I have
> consulted with the rest of the authors to craft the reply below.
That's okay, just means we probably end up repeating the discussion on
the floor in a few minutes. :-(
> It's expected that the RIR's, through regional policy, will do
> their best to cooperate. We have the example that right now, an
> RIR can take five /8s from the IANA. Yet, the RIRs have agreed
> to take up to only two /8s at a time.
Is there an APNIC policy which says that it has to return unused address
space to IANA?
> To clarify, 5.3 states:
> The Reclamation Pool will be divided on CIDR boundaries and
> distributed evenly to all eligible RIRs.
> The key word here is "eligible". If an RIR is eligible, then
> the space will go there "where it is needed".
> It's worth noting that an RIR can fall in and out of eligibility
> depending on its state of exhaustion.
Okay, so it is not being divided equally between RIRs. It will be given
to RIR regions who do not have carefully considered soft-landing policies.
> Ya know, that's not all that different from dividing the last
> five IANA /8s evenly. And that's triggered when only 1 RIR
> has demonstrated need while the other 4 have address space. :)
Yup, that's a unique situation.
> In your example, only when an RIR is considered to have exhausted
> its address space will it be considered to be eligible for the
> /19. So if an RIR doesn't need the space, they won't get it.
As per above, those who are profligate and have no run-out policy stand
to benefit at the expense of the others.
> I expect that if a new RIR is forming with global community
> support, we as a comunity will have ample time to update the
> global policy to include the new RIR.
Seems odd to explicitly exclude it.
> But if a new RIR just appears out of nowhere....
Not sure what you mean here. RIRs don't appear out of nowhere, they are
formed with their community support, and are part of the global community.
> We authors do not intend to interfere with intra-RIR transfer
> policies that cover all address space prior to the IANA
> exhaustion. The transfer restriction is applied only to
> IANA-reallocated addresses covered by this policy proposal.
But banning transfers of the addresses covered by this policy proposal
does interfere with intra-RIR transfer policies.
> We would like to encourage the RIR's to develop inter-RIR
> transfer policy that is fair to all regions.
Well, you've just proposed one that transfers address space to RIRs who
don't have good soft landing policies in places. ;-)
For many years I was under the impression that IPv6 was where we
eventually wanted to be, and that IPv4 is going to be phased out. We've
developed a soft-landing policy for our final /8 here in the APNIC
service region (I was a co-author), designed to ensure that the APNIC
community has sufficient IPv4 from the last /8 so they can fully
transition to IPv6 over the coming years. If organisations are giving
IPv4 addresses back to APNIC because they no longer need them (something
I don't see happening for many years yet), it means that the entire
industry will have successfully completed the transition to IPv6. So why
do we even need IPv4 addresses then?