Routing Area Director: o Joel Halpern: jhalpern@newbridge.com Area Summary reported by Joel Halpern/Newbridge Networks Corporation Inter-Domain Multicast Routing Working Group (IDMR) The focus of the Danvers meetings was whether there is evidence to substantiate the need for shortest-path multicast trees for inter-domain multicasting (it would seem that the available documentation suggests not). The simulation results carried out comparing CBT and PIM showed CBT incurred less average delay between receivers than had been previously thought. These results, of course, must be put into perspective given the somewhat connectivity-poor simulation topologies. In terms of protocol and bandwidth overhead, it was shown that both protocols perform similarly overall. Further simulations using more redundant topologies are to be conducted, and it is hoped these results will be available for the meetings in Stockholm. Implementation work is continuing on both protocols, and both are being implemented in `gated.' As the debate on whether multicast SPTs are appropriate for inter-domain multicasting could not be resolved, it would seem that implementation experience will be the deciding factor. A number of people voiced the opinion that the market should decide in any event. Inter-Domain Routing Working Group (IDR) The following items were discussed at the IDR Working Group meeting: o Selective Updates in Inter-Domain Routing o A Further Application for RIFs: De-aggregation in ATM Clouds o BGP/IDRP Route Server Proposal o BGP Communities o Routing Confederations o AS Guidelines o BGP-4 to Full Standard o IDRP for IPv4 and IPv6 o Performance Related Attributes for IDRP o BGP AS Path Metrics It was agreed to publish BGP/IDRP Route Server as an Experimental RFC, to publish BGP Communities as an Experimental RFC, to ask the IESG to advance the AS Guidelines document (after appropriate modifications) to a Proposed Standard, and to prepare all the necessary documents to move BGP4 to a full Standard by the next IETF. IP Routing for Wireless/Mobile Hosts Working Group (MOBILEIP) MOBILEIP met and the group has reached a point of consensus that should enable implementations to proceed. The group discussed broadcast and multicast issues, address usage in the protocol, and resolved several other minor problems with the -9 draft. The only outstanding issue involves two IBM patents and the lack of a clear and precise statement from IBM as to the terms (if any) with which IBM makes these technologies available. The perception of the co-chairs is that the underwhelming number of Mobile IP implementations is due in large part to this outstanding patent issue. Nonetheless, several working group members plan to implement Mobile IP for interoperability testing scheduled tentatively for the August time frame. Charlie Perkins, working group editor, hopes to have an updated (and quasi-final) draft available by the end of April. New Internet Routing and Addressing Architecture Working Group (NIMROD) o Charlie Lynn gave an overview of the Deployment Document. He discussed strategies for the gradual integration of Nimrod in the Internet. o Martha Steenstrup presented a list of information elements that could be obtained by either configuration or discovery. For each element, Martha discussed the trade-offs and the current thoughts on what to do with that element. o Isidro Castineyra described a bootstrap model for Nimrod. o Martha Steenstrup described an alternative bootstrap model. o Charlie Lynn presented the current software architecture for the BBN implementation of Nimrod. o Martha Steenstrup listed alternatives for path setup. Open Shortest Path First IGP Working Group (OSPF) At the Danvers IETF, the OSPF Working Group did the following: Proposed modifications to the base OSPF specification were reviewed. The current approach for OSPF for IPv6 was outlined. The current states of the OSPF MIB document, the CIDR routing table MIB and OSPF MD5 authentication were described. Possible ways to use OSPF to solve the ROLC problem were explored. The meeting ended with a discussion of a proposal to enhance OSPF security by signing individual LSAs. Router Requirements Working Group (RREQ) The Router Requirements Working Group met for one session. Minor editorial changes to the draft were discussed. A small change to wording to clarify certain security issues was made. The draft is in Last Call for becoming a Proposed Standard. Routing Information Protocol Working Group (RIP) The RIP Working Group met on Thursday. The group began with a presentation of the agenda and of the current documents associated with the group's charter. Fred Baker gave a brief presentation of the changes to his Internet-Draft on cryptographic authentication for RIP-2. The group agreed with the changes. The draft will be submitted for consideration as a Proposed Standard next week. Gerry Meyer gave a presentation on his Internet-Draft on demand circuit RIP. This draft updates the current Proposed Standard RFC-1582. There was significant discussion about the handling of slow links and devices with many more logical addresses than physical interfaces. It was decided that these issues should be resolved between Gerry and the interested parties prior to submitting into the standards track. Whether it goes in as Proposed or Draft will be determined by the extent of the changes. The group continued and completed discussions started on the mailing list by Paul Traina. The changes to the RFCs are detailed in the meeting minutes. The group discussed the RIPng draft. There was concern that it is the minimal possible change from RIP-1 and RIP-2. However, that was the condition set forth by the area director for the draft`s creation. It will be submitted for consideration as a Proposed Standard. Routing Over Large Clouds Working Group (ROLC) There were 125 attendees. The working group received an ATM Forum Multiprotocol over ATM report from George Swallow. Dave Katz led a review of the current NHRP draft, draft-ietf-rolc-nhrp-IV.txt, which had been distributed on the mailing list the previous week. A number of action items for the editors were generated. An official revision of the specification, to be version -04, will include changes generated at the meeting and is expected to be distributed as an Internet-Draft several weeks following the conclusion of the meeting. Derya Cansever led a review of the NHRP Applicability Statement, draft-ietf-rolc-nhrp-appl-01.txt. An updated version will be forthcoming. Mike Patrick led a review of the NHRP MIB, draft-ietf-rolc-nhrp-mib-00.txt. An updated version will be forthcoming. Kanan Shah will be producing a Protocol Analysis draft for the next working group meeting. The work plan was updated, and will be reflected in a charter update. Source Demand Routing Working Group (SDR) Yakov Rehkter presented the current version of Explicit Routing Protocol (ERP) and its recent changes. The forwarding algorithm for strict hops in the explicit route has been worked out more carefully and appears to operate correctly. Steve Deering pointed out that ERP is less space efficient than simple recursive encapsulation. However, this implies that a router's FIB must contain some indication of all immediately adjacent regions so that strict hops can be properly supported.