Operational Requirements Area Directors o Scott Bradner: sob@harvard.edu o Mike O'Dell: mo@uunet.uu.net Area Summary reported by Scott Bradner/Harvard and Mike O'Dell/UUNET Meetings of five Operational Requirements Area working groups and four BOFs were held during the IETF meeting in Danvers, Massachusetts. IP Provider Metrics BOF (IPPM) The primary topic for the BOF was the formation of a working group, although many of the areas where new technology might be needed were introduced. It was the consensus of the attendees that the work is important and belongs in a working group. It should focus on the delivery and routing of datagrams (and not related services, such as NOC characteristics). There is an apparent overlap with the BMWG Working Group, but the common view was that the work would be stronger in a more narrowly focused effort. Some requirements for metrics and a wish list of useful metrics were drafted. Some existing performance tools were briefly surveyed. Matt presented a bulk transfer performance tool, ``treno,'' which he is working on. Bill Manning lead an extensive discussion on metrics that might be applied to routing. Protocol Testing BOF (PROTTEST) The session began with a discussion of goals; they included definitions and methodology for testing of protocols, and guidelines for testability of specifications and protocols. The discussion soon focused on testability, with the idea of producing a set of informational guidelines for RFC authors to assist in creating testable (and implementable) specifications. It was decided to proceed by chartering a working group, with the goal of producing an Informational RFC. Art Mellor volunteered to be chair, and Peter Desnoyers to be document editor. A charter will be drafted and discussed on the mailing list, stdguide@midnight.com. Remote Authentication Dial-In User Service BOF (RADIUS) The RADIUS BOF had 42 attendees. There was considerable interest in forming a working group to 1) create a standards track RFC documenting the RADIUS protocol, and 2) create an Experimental RFC documenting RADIUS Accounting as it exists now. Carl Rigney has the action item to send out the minutes, a charter proposal and a proposed schedule to the RADIUS mailing list and Operations Area co-Directors by 17 April, and has volunteered to edit both documents. Several people expressed interest in exploring the position of working group chair and are talking to Carl and Mike O'Dell, Operations Area co-Director. Routing Policy System BOF (RPS) Daniel Karrenberg presented RIPE-181 to form the motivation and the background for the working group. Cengiz Alaettinoglu presented RPSL technical details (proposed details). Daniel presented the proposed distributed registry model. Cengiz and Daniel presented the available tools that operate on policy data in the registries. Curtis Villamizar presented his results from specifying AS690 policies and made various suggestions. The attendees interacted a lot and made many suggestions. A consensus was reached that the work should be done as an IETF effort in a working group. CIDR Deployment Working Group (CIDRD) Presentations on IPv4 address space usage and routing table size were made. The latest projection on address space utilization gives 2018 +/- 8 years. The routing table size is approximately 25,000 routes, up significantly from a year ago, but still showing dramatic reductions from the old exponential growth curve. An experiment will be run to see if the routing system is working with subnets of class A address space. This will allow us to make use of the reserved class A space, which is a substantial portion of the remaining address space. A talk on ``address ownership'' was given. Address ownership is incompatible with hierarchical routing. A version of this talk will be published as an RFC. The working group agreed that RFCs 1597 and 1627 should be revised, combined, and issued as a Proposed Standard. The working group is also going to start a document to describe possible filtering of very long prefixes to prevent insane exceptions from being entered into backbone routing. Guidelines and Recommendations for Security Incident Processing Working Group (GRIP) The GRIP Working Group met twice during this IETF. During the two meetings, the content of a framework template for security incident response teams was agreed on. This template will be included as an appendix in the document that is being created (Guidelines and Recommendations for Security Incident Response Teams). The group then used the template to define an outline for the document itself. After settling on the outline, the material in the current draft was reviewed and the group defined the content they wanted to see in each of the sections of the document. All in all, the two meetings were very productive and the group expects to have a new draft out by 1 May. The group further plans to have a second draft out by 1 June with the goal of issuing a Last Call by the next IETF. The group is planning to meet at the Stockholm IETF. Network Status Reports Working Group (NETSTAT) Summary not submitted. Operational Statistics Working Group (OPSTAT) An update on the status of 1404 was given. The 1404 draft is currently out as an Internet-Draft. Minor changes are required before sending it to the RFC Editor as an Informational RFC. The client-server draft has incorporated all the changes submitted on the mailing list. Minor changes were discussed and consensus was reached on incorporating them. Work is progressing on the SNMP User's guide very slowly but hopefully will pick up soon. N. Brownlee discussed traffic studies that had been done at his site and the statistical analysis of said traffic. They discovered that the traffic had a self-similar nature. Simulations were performed utilizing distributions with finite mean and infinite variance, but more work needs to be done to determine the characteristics of the traffic in more detail. RWhois Operational Development Working Group (RWHOIS) A meeting of the RWhois Working Group was held on 5 April. The status of the RWhois release was discussed with an anticipated delivery of 1.0 BETA1 the following week. The issues of integrating an authentication mechanism for registration was discussed. It was recommended to look at the work done in the CAT Working Group in the Security Area. The group also discussed the requirements of an RWhois registration server including the minimum attributes, distributed record locking, and secondary server requirements.