[arin-ppml] Draft Policy ARIN-2013-3: Tiny IPv6 Allocations for ISPs

[David Farmer]

On 3/29/13 14:05 , John Curran wrote:
> On Mar 29, 2013, at 2:47 PM, David Farmer <farmer at umn.edu> wrote:
>
>> The primary issue I want to avoid people going up and down several times.
>
> Why would this be expected, and/or a concern?

I'm not sure, I'd expect it.  I'm not even sure it is a real concern, 
its just making me uncomfortable for some reason.  Maybe I just need to 
get over it and not worry about it.

>>   I guess we don't need restrict it to people who received a /32 before the /36 or /40 were available.  So, I'd be OK everyone having one opportunity to reduce form /32 to /36 or /40.
>
> The challenge, of course, is that circumstances do change, and  (for example)
> a party that thought it wanted to move up to /32 might change its mind a year
> later...
>
> ARIN is here to serve the community, so the normal response to any request
> should be "Yes", unless there a clear reason (example, potential impacts
> to other parties) that something should be prohibited by policy.  Is that
> the case here, and can you elaborate on the policy concern that you see?
>
>> Unless you intended to create a generic ability to reduce your allocations that would apply to everyone and not just to the x-small and xx-small categories.  But that wasn't explicitly clear, and I took it that it was intended to allow low end adjustments.  I'd like other to weigh in on if there should be a generic ability to reduce your IPv6 allocation.
>
> Specific policy regarding return of IPv6 space might be necessary if you want
> ARIN not to accept returns (or only accept whole blocks and not partial, etc.)

As I see it this came up because we people asked to return all but a /36 
of their /32, they were initially told they had to return they /32 and 
the could get a new /36.  Business practice wise you fixed the immediate 
problem, I think.

After thinking about it for a while, if we even need any policy at all, 
we should just have a general policy describing how IPv6 block can be 
reduced by returning only part of a block, it should probably be very 
generic and apply to allocations and end user assignments.  It might 
even be better if it we a separate proposal all together.

What do others think?

-- 
================================================
David Farmer               Email: farmer at umn.edu
Office of Information Technology
University of Minnesota
2218 University Ave SE     Phone: 1-612-626-0815
Minneapolis, MN 55414-3029  Cell: 1-612-812-9952
================================================