<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7654.12">
<TITLE>RE: [arin-ppml] IPv6 Transition Policy (aka Soft Landing)</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<P><FONT SIZE=2>Chris,<BR>
<BR>
As you know, at our lunch table discussion 'fighting over the crumbs of v4', I made mention that I believed it was beyond the scope of ARIN's mission to 'mandate a business practice change' like requiring v6 implementation as a condition of receiving future v4 allocations.<BR>
<BR>
In a brief discussion about the legality of such....I now believe(or interpret what I heard) that ARIN might arguably do this without undue risk....<BR>
<BR>
FYI...<BR>
<BR>
bd<BR>
-----Original Message-----<BR>
From: arin-ppml-bounces@arin.net on behalf of Chris Grundemann<BR>
Sent: Sat 10/9/2010 2:34 PM<BR>
To: arin-ppml@arin.net<BR>
Subject: Re: [arin-ppml] IPv6 Transition Policy (aka Soft Landing)<BR>
<BR>
Let me offer a general response to all of the great feedback:<BR>
<BR>
The basic idea here is to redefine efficient utilization. The fact is<BR>
that at this point (and for quite some time now) IPv4 addresses that<BR>
are not deployed along with IPv6 addresses are simply not being<BR>
efficiently utilized. Just like assigning a v4/24 to a PTP link is<BR>
wasteful, deploying IPv4 only is wasteful.<BR>
<BR>
Many folks have made comments that effectively say "this is hard" and<BR>
that in some cases, IPv6 deployment is still impossible. What do you<BR>
think is going to happen next year when there is no more IPv4? This<BR>
policy effectively places a new wall in front of the no-more-v4-at-all<BR>
wall. The effect is to apply real pressure now, so that we have less<BR>
pain then.<BR>
<BR>
Every crazy scenario that you can think of with regard to IPv4<BR>
depletion is going to happen - all of it. The real question is not how<BR>
bad it will be (bad) or when it will happen (soon) but rather how long<BR>
will it last? The answer depends on whether this community decides to<BR>
step up and provide real leadership or not.<BR>
<BR>
The only true soft-landing strategy that will allow the transition to<BR>
happen without pain is to dual-stack everything and simply let traffic<BR>
shift over from v4 to v6. That ship has sailed. So now we have two<BR>
choices:<BR>
<BR>
1) Change nothing and run into the wall. This idea has merit in that<BR>
we stick to our guns, and provide some predictability by not changing<BR>
the rules.<BR>
2) Require IPv6 deployment now. This idea has merit because it allows<BR>
IPv4 to keep growing while also softening the blow when new IPv4 runs<BR>
out.<BR>
<BR>
The way to implement option #1 is pretty obvious. The method to affect<BR>
option #2 is probably a bit less clear. What Jason, Marty and I have<BR>
come up with here is a start at defining that method. Due to the<BR>
applause at open mike, the in-person feedback in Atlanta and the<BR>
responses here on the list - I am going to assume that this is<BR>
valuable and that there are a lot of folks in the community who think<BR>
that option #2 (action) is preferred to option #1 (inaction).<BR>
<BR>
So - where do we go from here?<BR>
<BR>
I would like to propose that we start with a high level discussion of<BR>
whether or not this basic idea is what the community wants and then<BR>
(if we do want this) dive quickly into the details of how exactly to<BR>
do it.<BR>
<BR>
With that in mind, I set up a quick poll on doodle:<BR>
<A HREF="http://doodle.com/wewba7hnpkqhc9e2">http://doodle.com/wewba7hnpkqhc9e2</A><BR>
<BR>
I will post the poll to ppml separately as well to grab the widest<BR>
audience - please forgive the double post. Feel free to share the poll<BR>
with all other interested members of the Internet community.<BR>
<BR>
Cheers,<BR>
~Chris<BR>
<BR>
"Those who do not create the future they want must endure the future they get."<BR>
~Draper L. Kaufman, Jr.<BR>
<BR>
<BR>
<BR>
On Fri, Oct 8, 2010 at 10:00, Chris Grundemann <cgrundemann@gmail.com> wrote:<BR>
> Problem Statement:<BR>
> We have failed to deploy dual-stack in a meaningful way in time to<BR>
> avoid transition problems<BR>
><BR>
> Objectives:<BR>
> -1- Encourage IPv6 deployment prior to depletion<BR>
> -2- Enable growth of IPv4 where IPv6 is being deployed<BR>
> -3- Improve the utilization of IP addresses<BR>
><BR>
> High-Level Requirements:<BR>
> -1- ARIN will only make allocations and assignments for networks that<BR>
> have already deployed production IPv6<BR>
> -2- Any IPv4 addresses received under this policy, must be deployed<BR>
> along side of IPv6<BR>
> -3- This policy will encourage deployment of IPv6 in existing IPv4-only networks<BR>
><BR>
> Rough Policy Text:<BR>
> ~ Requester defines classes in their network - only classes where IPv6<BR>
> is in production qualify for IPv6<BR>
> ~ New addresses must be deployed on dual-stacked interfaces plus one<BR>
> additional existing IPv4-only interface must be dual stacked, up to<BR>
> 80% of all interfaces.<BR>
> ~ The service that the address is used to provide must be fully IPv6<BR>
> accessible (if you deploy an A record, you must also have a AAAA and<BR>
> both must answer)<BR>
> ~ All end-sites must dual-stack all Internet facing services before<BR>
> getting this space<BR>
> ~ For each down stream customer site where these addresses are<BR>
> deployed, another pre-existing IPv4 only down stream site must also be<BR>
> IPv6 enabled, up to 80% of the total customer base.<BR>
><BR>
><BR>
> This is an emergency, let's get something together ASAP. All feedback<BR>
> is extremely welcome and greatly appreciated; this problem is all of<BR>
> ours. If you are still here in Atlanta come find me, Marty Hannigan<BR>
> and/or Jason Schiller to discuss.<BR>
><BR>
> Thanks!<BR>
> ~Chris<BR>
<BR>
<BR>
--<BR>
@ChrisGrundemann<BR>
weblog.chrisgrundemann.com<BR>
www.burningwiththebush.com<BR>
www.coisoc.org<BR>
_______________________________________________<BR>
PPML<BR>
You are receiving this message because you are subscribed to<BR>
the ARIN Public Policy Mailing List (ARIN-PPML@arin.net).<BR>
Unsubscribe or manage your mailing list subscription at:<BR>
<A HREF="http://lists.arin.net/mailman/listinfo/arin-ppml">http://lists.arin.net/mailman/listinfo/arin-ppml</A><BR>
Please contact info@arin.net if you experience any issues.<BR>
<BR>
</FONT>
</P>
</BODY>
</HTML>