ARIN-prop-301: Remove Initial Small Assignment Requirements [Archived]

OUT OF DATE?

Here in the Vault, information is published in its final form and then not changed or updated. As a result, some content, specifically links to other pages and other references, may be out-of-date or no longer available.

Date: 10 September 2021

Proposal Originator: Larry Dockery

Problem Statement:

6.5.8.1 provides several criteria to determine eligibility for an IPv6 initial assignment:

[a] Already having an IPv4 assignment [b] IPv6 multihoming [c] actively using 2000 IPv6 addresses [d] actively using 200 /64 subnets [e] 13 active sites [f] justification as to why ISP/LIR addresses cannot be used.

The specific problems with this are:

  • /44’s and /48’s are exceedingly abundant. It is not necessary to have strict eligibility criteria for obtaining these smaller assignments. The fees should be enough to manage and maintain the assignments under the proposed change in light of the removed verification work.
  • There is no clear meaning of “active use” as said within [c] and [d]. Does active use mean “assigned to a host”, “assigned to a service”, or “responds to ICMP echo on the public Internet”?
  • Smaller organizations with less than 12 sites - the majority of all businesses and organizations, that would benefit from PI space may not currently qualify under 6.5.8.1 or may have difficulty proving they do under the current process.
  • The current eligibility restrictions for /44 and /48 slow IPv6 adoption by denying most business and organizations PI addressing, and encouraging business to remain with RFC1918 that they control, or adopt practices such as NAT-ULA that are not in line with end-to-end connectivity and GUA addressing of IPv6.
  • Provider lock-in with PA addresses is weak in IPv4 and strong in IPv6. A common network design for small and medium sized organizations is static IPv4 addresses provided by an ISP with NAT to RFC1918 for internal addresses. These organizations do not have to re-address their entire internal network when switching Internet providers. In contrast these organizations would have to re-address the entire internal network with IPv6 unless they obtained provider independent address space. This makes Internet provider lock-in strong in IPv6. This could be argued to satisfy [f] for all organizations.
  • To avoid provider lock-in, organizations that cannot qualify for PI space could consider NAT to ULA as an option to preserve the ability to easily switch providers and ensure consistent internal addressing on-par with IPv4 RFC1918. This legitimate need may confuse ULA as a suitable and equivalent replacement for preserving existing RFC1918 network designs.
  • The above point is not hyperbole, as it is already suggested as a solution in the most popular IPv6 design documentation on the web and printed. For example, Tom Coffeen, author of O’Reilly Media “IPv6 Address Planning” and host of IPv6 Buzz notes in his book “Some single-homed organizations may be concerned about a greater likelihood of having to renumber. This could be a consequence of changing ISPs. With such a concern, the deployment of ULA with NPTv6 may be an option.”

Policy statement:

Proposal:

Remove 6.5.8.1

Change the first sentence of 6.5.8.2 to “All end-user organizations are eligible to receive an initial assignment of a /48 for a single site, or a /44 for up to 12 sites.”

Timetable for implementation: Immediate

OUT OF DATE?

Here in the Vault, information is published in its final form and then not changed or updated. As a result, some content, specifically links to other pages and other references, may be out-of-date or no longer available.