Minutes IAB Teleconference 2008-06-25 1. Roll-call, agenda-bash, approval of minutes, administrivia 1.1. Agenda 1.2. Attendance PRESENT Loa Andersson Gonzalo Camarillo Stuart Cheshire Russ Housley (IETF Chair) Olaf Kolkman (IAB Chair) Gregory Lebovitz Barry Leiba Kurtis Lindqvist Andy Malis Danny McPherson Dave Oran Dow Street (IAB Executive Director) Dave Thaler Mark Townsley (IESG liaison) Lixia Zhang APOLOGIES Aaron Falk (IRTF Chair) Sandy Ginoza (RFC Editor Liaison) Lynn St. Amour (ISOC Liaison) 2. IETF 72 Plenary Olaf began the meeting with a few IETF 72 coordination items (e.g. IAB BOF coverage), and then the discussion moved to the Technical Plenary. Gregory had previously summarized the plan in an email, and during the meeting he proposed that the plenary focus on IPv6 adoption successes, barriers, and IPv4 contingency planning. The overall goal would be to get the community to further engage in IPv6 adoption. A notable element of the transition debate involves the possibility of carrier grade NATs. The board intends to use a panel format for the plenary discussion, and will seek speakers from multiple perspectives: carrier, content provider, OS vendor, home gateway, etc. Danny and Dave Thaler will work to identify speakers; the board will review status at the next business meeting. At the end of this agenda item there was a short discussion about the possible need for an IAB workshop on carrier grade NATs, 6-4, and some related impacts to BEHAVE work. The board will revisit this question after the Dublin meeting. 3. Architecture Topic - Securing Inter-domain Routing Danny gave an update on the current lay-of-the-land in the world of securing inter-domain routing. The SIDR WG is working primarily on PKI and repositories, not really on the routing component itself. There are several additional touch-points needed between the certificate infrastructure and the routing protocol(s) before routing itself could actually be secured. These steps will need to involve RPSEC, IDR, GROW, and possibly OPSEC. Unfortunately, much of the discussion of authentication infrastructure is currently taking place outside the purview of the IETF 'proper', namely on the RESCERT mailing list: http://mailman.apnic.net/mailman/listinfo/rescert At some level, though the 'design work' is occurring via RESCERT, the expectation is that any output would be published via the IETF. Russ also noted that the discussion is likely to shift to IETF channels once the focus moves toward the routing mechanisms. All in all, there is a fair amount of work ongoing, but specific IAB next steps (if any) are not obvious. The group discussed a possible IAB document that would provide a taxonomy of the components involved in securing inter-domain routing, in turn providing some context to how work might be spread across various WGs. At previous IAB meetings the group had discussed the architectural ramifications (e.g. security vs autonomy) for securing routing in the manner that is currently being pursued. This IAB document could also raise awareness of such architectural considerations. Danny volunteered to draft a document outline for consideration by the group. 4. Architecture Topic - Peer-to-Peer Architectures Gonzalo led a discussion of the IAB peer-to-peer work item. The plan is to draft a document that provides a taxonomy of P2P architectures and relates P2P to server farms, grid computing, and client-server models. As agreed to previously by the IAB, the document would also emphasize that P2P approaches are useful for a variety of legitimate and legal purposes, helping to dispel any negative connotations of the technology. Gonzalo has completed a literature survey and recently added a detailed document proposal to the internal IAB wiki. The draft would likely start as an individual document, and then transition to an IAB document prior to publication. Specific near-term feed- back was requested of all IAB members. Gonzalo will use this feedback to write an initial draft, which the IAB will discuss at the Sunday IETF meeting (Dublin). 5. AOB There was a short discussion of an IPv6 experiment at the upcoming IETF. Stuart noted that such an experiment would be useful in continuing to motivate deployment of IPv6 services at Apple and elsewhere. It is likely that the Dublin meeting will not have a period where IPv4 connectivity is disabled (as was the case in Philadelphia), but that the IPv6-only network would be available throughout the conference sessions. Minneapolis was targeted as a first possibility for another "no v4" experiment, allowing more time between experiments for implementation of IPv6 services. Olaf noted that there is some follow-up work needed with IANA over the next few weeks. He will send an email to list. 6. Conclude Call