Removing Routing Requirements from the IPv6 Address Allocation Policy
Summary of Proposal:
The IPv6 address allocation policy currently contains mandates about how an allocated address range should be announced into the routing table. Following discussion at RIPE 58, it is proposed that this is removed from the address policy as it does not relate to address allocation.
Arguments supporting the proposal
Policies on how the RIPE NCC must make IPv6 allocations to the LIRs are documented in the RIPE policy document, “IPv6 Address Allocation and Assignment Policy”. These allocation policies cannot control routing practices, which are decided by network operators. The policy already notes that routability of those allocations made by the RIPE NCC cannot be guaranteed (see section 4.2. Routability Not Guaranteed in RIPE-466).
There is also a current RIPE document, ripe-399, “RIPE Routing Working Group Recommendations on Route Aggregation”. This document
notes that there are occasions where there is a legitimate reason and a need for advertising more specific prefixes from an allocation. Recent discussion around policy proposal 2009-05, “Multiple IPv6 /32 Allocations for LIRs” has also highlighted one of those reasons.
Therefore it is proposed to remove the routing requirements from the IPv6 allocation and assignment policy.
If other guidelines are required, they can be produced by a more appropriate forum or working group.
Arguments Opposing the Proposal
Removing the requirement from the allocation policy may lead to a 'free for all' and explosion in the IPv6 routing table size, but this is unlikely as operators will decide what level of deaggregation they will support, and this is unlikely to allow for very large numbers of prefixes to be announced.
Note: In order to provide additional information related to the proposal, details of an impact analysis carried out by the RIPE NCC are documented below. The projections presented in this analysis are based on existing data and should be viewed only as an indication of the possible impact that the policy may have if the proposal is accepted and implemented.
A. Impact of Policy on Registry and Addressing System
Address/Internet Number Resource Consumption:
After analysing the data that is currently available, the RIPE NCC does not anticipate any significant impact on Internet number resource consumption if this proposal is implemented.
It is difficult to estimate how many of the current IPv6 prefixes would actually be deaggregated if this proposal were to be accepted. However, we have taken the factor of deaggregation for IPv4 prefixes and made an estimate for IPv6 prefixes, based on the assumption that deaggregation practices for IPv6 will be similar to IPv4.
The factor of deaggregation in IPv4, based on the current number of IPv4 entries in the RIPE NCC statistics and the number of more specific entries in the global routing tables (as seen by the RIPE NCC Routing Information Service), is roughly 3.4. At the time this analysis was conducted, about 750 IPv6 allocations appeared in the global routing tables. Applying the deaggregation factor of 3.4 to this number, we estimate that if this proposal were to be accepted, the number of new IPv6-related routing table entries would be approximately 1800.
B. Impact of Policy on RIPE NCC Operations/Services
Having analysed the data that is currently available, the RIPE NCC does not anticipate that this proposal will have a significant impact on RIPE NCC Operations/Services, if implemented.