ROUTING AREA Director: o Joel M. Halpern Area Summary reported by Joel M. Halpern, Newbridge Networks IP Routing for Wireless/Mobile Hosts Working Group The Mobile IP Working Group met twice in Dallas. The first meeting concentrated exclusively on IPv4 mobility; the second primarily on IPv6 mobility. In relation to Ipv4, all patent issues relating to Mobile IP have been solved. The working group was comfortable with the summary opinion of the counsel hired by the IESG to look into the scope of infringement of Mobile IP with respect to IBM patent #5,159,592 (Charles Perkins' patent) and decided to include this summary as an appendix in the draft. No changes in Mobile IP functionality were deemed necessary by the working group with respect to the forementioned patent. As for IBM patent #5,148,479 (Bird, et. al.), the working group decided to "work around" this patent by using timestamps as the mandatory mechanism for replay protection while allowing the optional use of timestamps. In other matters, the wholesale re-write of the draft was met favorably and, with minor editorial changes, a "final" draft will be produced in very short order. In an interim meeting in October, substantial interoperability among independent implementations of Mobile IP was demonstrated. In addition, an Applicability Statement and a MIB will be submitted as Internet-Drafts very shortly. With the completion of these tasks, all the requirements of RFC1256 will have been fulfilled, allowing the working group to go to "last call" and then hopefully pursue Proposed Standard RFC by the end of 1995. In dealing with Ipv6, the working group embraced the approach for the architectural framework for mobility in IPv6 proposed by Charlie Perkins (IBM) and Dave Johnson (CMU). Future versions of the Perkins/Johnson IPv6 mobility draft will be considered "official" working group documents. Community interest in IPv6 mobility seems disappointingly low at the current time. Inter-Domain Multicast Routing The IDMR Working Group met for two sessions in Dallas The meeting began with several presentations on the interaction of PIM with other multicast routing protocols, and with heirarchical multicast routing. A number of changes to PIM were discussed in this context. Then, there were presentations on implementation experience with PIM. There were discussion of the plusses and minusses, including some effort to understand what aspects were PIM specific, and what applied to any multicast routing protocol. The second session began with a presentation on multicast traceroute. The presentation explained both how it worked, and why it is useful. Work is needed to extend this to the shared tree routing protocols. The group has requested the advancement of the IGMPv2 spec to proposed standard. The Area Director will be dealing with this issue. There was a presentation on CBT. A number of changes have been made to streamline the protocol. It is hoped that the new specification is now stable. Inter-Domain Routing Working Group No information was submitted. Routing over Large Clouds Working Group The RoLC Working Group met in two sessions with about 145 attendees. In the first session, the RoLC and IP/ATM chairs presented a joint statement that ATMARP will transition to NHRP in well-defined steps that emphasize interoperability with the installed base. The RoLC Working Group asked that references to NHRP be removed from the IP/ATM Classic2 draft. Several NHRP specification issues were discussed, especially with regard to harmonization with the IP/ATM MARS specification. Several changes were made to the NHRP specification and, in one case, a corresponding change was also made in the MARS spec. The working group began to discuss server redundancy and synchronization issues; further discussion will be moved to the mailing list. There was consensus that server redundancy must be addressed before NHRP is ready for prime time, although the working group hopes to be able to leverage the redundancy work in the IP/ATM Working Group. Client autoconfiguration was also discussed, based upon the use of configuration servers. In the second session, the working group agreed upon a joint RoLC and IP/ATM liaison to the ATM Forum. The liaison includes two sections, one suggesting leveraging the LANE configuration server for IETF uses, the second asking for LAN Emulation Configuration Server redundancy. There were four presentations. Following Yakov Rehkter's Local/Remote presentation, the working group agreed to forward his APR draft as an informational RFC. As a result of Hiroshi Suzuki's Mobile NHRP presentation, the working group agreed to include registration forwarding in the NHRP spec. Rob Coltun presented his approach to ATMARP and NHRP coexistence. The working group agreed to use Yakov Rehkter's Router- Router NHRP presentation as the basis for future work; Yakov will be documenting it in a forthcoming Internet Draft. The working group discussed changes to the charter and workplan, this discussion will be continued on the mailing list. The working group expects to meet again for two sessions at the Los Angeles IETF. RIP Working Group The RIP Working Group met on Tuesday, December 5. We began with a presentation of the agenda and of the current documents associated with the group's charter. RIP-2 (RFC 1723) is still waiting for the inclusion of introductory material from RFC 1058. Once this is done, RIP-2 (all RFCs) can be submitted for consideration as full Standards. The RIPng Internet Draft is still waiting for text to resolve the "Dedicated Addresses" issue. Several people volunteered to contribute that text. Once this is done, the I-D will be resubmitted for consideration as a Proposed Standard. Gerry Meyer presented Triggered RIP as defined in Internet Draft draft-ietf- rip-trigger-rip-00.txt. It was decided that no further development effort will be expended on RFC 1582 and that it will not be advanced in the standards track. The applicability statements will be modified to indicate that Triggered RIP is the preferred method of handling Demand Circuits. The Triggered RIP I-D will be submitted for consideration as a Proposed Standard after a few editorial changes are made. Once Triggered RIP advances, RFC 1582 will be moved to Historical Status. Internet Draft draft-ietf-ripv2-md5-02.txt is still awaiting approval for acceptance as a Proposed Standard. This needs to proceed so that the RIP-2 documents can proceed along the standards track. Source Demand Routing Working Group Deborah Estrin presented a proposal for incremental deployment of SDR systems, using the Internet Routing Registry for route construction and a new probing mechanism. However, incremental deployment would still require effort on part of the source. It appears that because of the relative uniformity of provider services, pricing structures, and access policies, there is no current demand for alternate-path unicast forwarding. On the other hand, route construction and probing appear to be fruitful avenues to pursue for alternate path multicast joins. This should be pursued in the context of multicast routing protocol development and does not require a separate working group. In the meantime, we propose that the SDR Working Group should become dormant until the differences among service providers creates a need for such policy control, or until we determine that the group should be disbanded. Mobile Mesh Networking BOF Steve Batsell and others chaired the meeting. The Mobile Mesh Networking BOF held one session at the Dallas IETF. There were presentations on the requirements from the Army Research Lab, Electrospace Systems, The Air Force, and the Navy Research Lab. The presentations outline the range of requirements which the Department of Defense feels need to be addressed. All the requirements presentations agreed on the urgent need for solutions, and all of them indicated that research is being funded into the problem area. M. Scott Corson presented a set of architectural considerations for addressing the issues which were raised. There was then extensive discussion attempting to refine the problem space and the proposed starting point for solutions. A number of issues (including the need to appropriately consider the non- military applications) were raised, and it was agreed that a mailing list would be established. Further work is needed to refine the requirements before it will be possible to request chartering a working group to work on this problem. There is a repository at ftp://mars.itd.nrl.navy.mil/mmnet for information relating to this activity. The mailing list for discussing this topic is mmnet@itd.nrl.navy.mil.