6BONE transition BOF

[Geoff Huston <gih@telstra.net>]

6bone Transition BOF

Current Issues
   - making the 6bone a robust testbed
   - transition the 6bone address registry (3FFE:: prefix) to the RIRs
   - planning the phaseout of the 6bone

Outlining of the address registry transition to the RIRs, noting that the 
phaseout is not an RIR-inspired issue.

There were reported to be comments from both the 6bone community and the 
RIR community. The 6bone community were qauestining the RIR fees associated 
with servicing 6bone addressing services. The price was an issue, 
irrespective of the amount. The RIR community was questioning the price 
discounts that the RIRs were proposing for 6bone services.

Bob Fink is saying that it is not clear that the RIR proposal should 
proceed given the comments being expressed, the 6bone phaseout plan, the 
decline of 6bone prefix requests, and the RIR's own initiatives in terms of 
making 'normal' V6 unicast addresses available. Also it was noted that the 
RIRs now have an experiment allocation policy.

The seocnd area is the transition of the 6bone reverse space to ip6.arpa. 
It is proposed that the 6bone operate eff3.arpa servers. This was noted as 
requiring an IETF document with an IANA considerations section to effect 
this delegation.

There is no migration of the 6bone registry to the RIRs, the 6bone would 
continue to manage its own allocations, and e.f.f.3.ip6.arpa is recommended 
to be delegated to 6bone servers.

Bob Hinden presented a 'phaseout plan', with the placeholder that "there is 
still some remaining need for the 6bone'. The logistics of the BOF were 
that the 6bone is not part of the v6ops charter, while it WAS part of the 
ngtrans charter. The proposed porcess is to continue the allocations for 
the next 16 months (1 July 2004) and thereafter make no further new 
allocations. All 6bone prefixes would be valid only until 1 July 2006, by 
which date there would be no 6bone prefixes on the Internet, and the prefix 
would at that stage be reclaimed in its entirety back into the registry 
pool (IANA registry) and the 6bone RFC would then be historic. The IANA 
prefix would be held 'for a period of time' before recirculation.

The commentary provided was that most people agree with the phaseout plan, 
but with comments requesting an accelerated timeframe of 1 January 2005 for 
an allocation cutoff and a 1 July 2005 shutdown. Further comments about 
preferred dates ensured, but noone arguing an indefinite extention of the 
6bone and the genreal tenor to stop allocating new prefixes sooner rather 
than later. Comments proposed the reuse of 3ffe:: as a MUST, rather than a 
SHOULD. A procedural request was to issue new IANA instructions to delegate 
e.f.f.3.ip6.arpa to 6bone servers by means of a dedicated Internet draft.

All sounds to me like a successful experiment being concluded in an orderly 
and accepted fashion. I like it!