[ppml] Policy Proposal: Changes to IPv6 policy - removal of "multiple /48" justification
Andrew Dul
andrew.dul at quark.net
Sat Jan 27 14:16:03 EST 2007
- Previous message: [ppml] FW: 2006-7 IPV6 Initial Allocation suggested changes-InputRequested
- Next message: [ppml] FW: 2006-7 IPV6 Initial Allocation suggested changes-InputRequested
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
At 12:01 AM 1/27/2007 +0100, JORDI PALET MARTINEZ wrote: >> De: Stephen Sprunk <stephen at sprunk.org> >> However, it's unclear that there is actually a problem here to be >> solved. Absent a comment from the ARIN Staff that a non-trivial number >> of larger-than-/48 assingments have been referred to them, I propose >> that we not solve this "problem" until it actually exists and we have >> the data to show what exactly needs solving. That seems reasonable to me too. > >The problem is no having an objective criteria. In absence of that, this >text on the policy is useless and it should be removed. > Or we should add objective criteria to the policy so that ARIN staff can make a determination if multiple /48 assignments should be permitted. How about the following... Remove the "Note:" paragraph from 6.5.4.2 and replace with the following. The RIR/NIR shall use the following criteria to evalulate multiple /48 assignments: New IPv6 end-user assignments: If the end user can show immediate need for at least 16,384 (25% of /48) /64 subnets they can be assigned additional /48 subnets such that their initial assignment usage is below 25% of initial /64 subnets. Existing IPv6 end-user assignments: If the end-user is currently using at least 32,768 (50% of a /48) /64 subnets they shall be permitted additional /48s such that their assignment usage is below 25% of /64 subnets for a two-year growth projection. Thoughts? To high? To low?
- Previous message: [ppml] FW: 2006-7 IPV6 Initial Allocation suggested changes-InputRequested
- Next message: [ppml] FW: 2006-7 IPV6 Initial Allocation suggested changes-InputRequested
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the PPML mailing list