[arin-ppml] ARIN-prop-136 Services Opt-out Allowed for Unaffiliated Address Blocks
> What do you do then? ARIN staff by now has determined the following:
> 1) The block is not used even in an unconnected network
> 2) The original holder is either ignorant of it's use and/or unable to
> use it, or is a proxy for the original holder.
> 3) Your not going to get permission, written or verbal, from anyone to
> take it back, based on them thinking that it might be worth something
> 4) The original holder doesn't give a damn what you mark for it in
> but is clearly never going to exert the effort to login to ARIN and
> modify the whois record.
> The reason I am asking is that I am thinking that perhaps ARIN needs
> some more clarity on what constitutes an abandoned legacy resource.
ARIN policy, as far as I know, is about the requirements of issuing new
IP addresses. If this entity went to ARIN to get new IP addresses,
considering their current assignment and usage according to your
scenario, they would be denied.
As far as I know, ARIN policy isn't about constantly monitoring usage
and yanking assignments when they are not currently in use. That would
result in a constant back and forth of
assignments/revocation/assignment/revocation as an entity changes. When
they got the assignment, they qualified for the IP addresses. They
might not qualify for that assignment *today* but they aren't making a
This isn't, to my knowledge, about making sure that everyone is always
within the usage guidelines of assignment, it is about being within the
usage guidelines when an assignment is made.
This is the sort of mess that is going to quickly make IPv4 useless and
probably ARIN-PPML useless if we continue to contort ourselves to chop
v4 into smaller and smaller pieces. Forget v4. It is out to pasture.
It is done. V4 is what it is. It will be around a while but I am
projecting that within 12 to 18 months time, my traffic will be about
85% v6 and 15% v4. In 24 to 36 months v4 traffic will be down to 5% of
my total bandwidth. It isn't worth spending a huge amount of time over.
Here's an idea: Give all of IPv4 North America to Benson to manage.
Let him handle it. The whole thing. Within 5 years it will still be
around but in the overall scheme of things, irrelevant, and Benson will
have managed to speed up the migration to v6 at least in the North