[arin-ppml] Policy Proposal: Protective UsageTransferPolicyforIPv4 Address

[Chris Malayter]

Scott,

I think this would work fantastic to be honest.  I'm not sure that
everyone else is going to be happy with it.  From an IX perspective, it
moves them into PI space and they are assigned from ARIN, which is what
the goal of renumbering would be if they are all forced down that path.

Speaking for myself,

-Chris

-----Original Message-----
From: Scott Leibrand [mailto:scottleibrand at gmail.com] 
Sent: Wednesday, February 11, 2009 7:28 PM
To: Chris Malayter
Cc: arin-ppml at arin.net
Subject: Re: [arin-ppml] Policy Proposal: Protective
UsageTransferPolicyforIPv4 Address

What if... <idea type=crazy> we simply asked EP.net to trade in their 
current IX-assigned space for new space, converted the returned EP.net 
space to critical infrastructure microallocation space, and convert all 
EP-IX reassignments to direct PI critical infrastructure assignments. 
</idea>

This would definitely be something we could do through the policy 
process, but it might be a way for ARIN to solve this problem in a way 
that makes everyone happy, and requires the minimum disruption
possible...

-Scott

Chris Malayter wrote:
> Leo,
>
> I guess the way I look at it is that the provider in question has long
> been (10 years+) a reliable broker of space for exchanges all over the
> world.  All of the IX's in the space have been blindsided by the idea
> that the space was now being shopped around for sale.  It would be the
> equivalent of ARIN deciding to pull back all the micro allocations and
> reuse them for something else.
>
> The point I'm making is that this is a non-trivial issue.  There are,
> from what I have been told, at least 40 or more IX's that are
> potentially affected.  
>
> Are there alternatives to a policy proposal, sure.  Are they the best
> way to maintain stability?  I'd have to say no.  It's going to be a
> rough few months if we have to renumber that many IX's globally, with
a
> bunch of them in the ARIN region.  
>
> I certainly think that this deserves the ability to move forward.
>
> -Chris
>
>
>
>
> -----Original Message-----
> From: arin-ppml-bounces at arin.net [mailto:arin-ppml-bounces at arin.net]
On
> Behalf Of Leo Bicknell
> Sent: Wednesday, February 11, 2009 6:43 PM
> To: arin-ppml at arin.net
> Subject: Re: [arin-ppml] Policy Proposal: Protective
> UsageTransferPolicyforIPv4 Address
>
> In a message written on Wed, Feb 11, 2009 at 06:13:51PM -0500, Martin
> Hannigan wrote:
>   
>>    What does that (EP or S/D) have to do with anything?
>>     
>
> Mr Malayter made the assertion that:  
>
> In a message written on Wed, Feb 11, 2009 at 01:26:15AM -0500, Chris
> Malayter wrote:
>   
>>    There are a large number of IX's in the North American region (as
>>     
> well
>   
>>    as other regions) that have address space allocated from a
provider
>>    that specializes in exchange allocations.
>>     
>
> Thus it is perfectly reasonable to quantify "a large number of

> IX's".  Since he works for Switch and Data, it seemed logical to

> begin the detective work with where their addressing blocks came
> from, which whois quickly locates as EP.NET.
>
> Mr Malayter further asserts that:
>
> In a message written on Wed, Feb 11, 2009 at 01:26:15AM -0500, Chris
> Malayter wrote:
>   
>>    The real issue is that if the current provider was to serve a
>>     
> majority
>   
>>    of the US IX's with a cease and desist order from using the space
>>     
> at
>   
>>    the term of all of the existing contracts at the end of 2009 that
>>    would force a massive renumber of most every IX in the North
>>     
> American
>   
>>    region, save one major IX.
>>     
>
> If the "real issue" is that the "current provider was to serve a

> majority of the US IX's with a cease and desist order" then looking

> at how many folks get space from the "current provider" would be 
> getting to the heart of the "real issue", now wouldn't it?  Since

> we know who that is, why don't we just look, rather than speaking

> in theoretical generalities?
>
> This is in fact critical to evaluating the policy.  Knowing how

> many folks might be affected by a policy change is one of the first

> things to evaluate a policy.
>
> This investigation has in fact been quite useful, as we now know if
> there is any problem, it is a contractual problem between a company
and
> its outsourcer, and there are already three solutions available today:
>
> 1) Renegotiate the contract to provide stronger protections.
>
> 2) Find another outsourcer who can provide addresses.
>
> 3) Come to ARIN and use the Micro Allocation for critical
infrastructure
>    policy to obtain addresses directly from ARIN.
>
> It appears the policy proposer would like a fourth option, of having
> ARIN step in the middle.
>
> To answer John Curran's question, "I am against the policy proposal as
> it appears there are ample other avenues for the requester to get what
> they want."
>
>