Draft minutes, please send corrections (to us or to the list, as appropriate). Thanks again to our scribes (Ted and itojun), who are not responsible for what we did with the notes they gave us.
--Rob and Dave == DNSOP WG, 58th IETF meeting, November 2003, Minneapolis First session: 2003.11.10 19:30-22:00 -0600 Summary minutes, not attempting to follow meeting chronology, which hopped, looped, and otherwise failed to follow a straight line. Active WG drafts: - draft-ietf-dnsop-bad-dns-res-01.txt. Important topic. Moving target, but doc out of date at the moment. Authors begged WG's indulgence to let them bring the draft up to date, with a promise that they will do so by IETF 59. - draft-ietf-dnsop-dnssec-operational-practices-00.txt. New document, not yet widely read. Important topic. Those who have read it think it's on the right track. May want to drop specific TTL recommendations in favor of discussion of tradeoffs. Discussion of key lengths needs review by crypto experts (request already made to security area). Good discussion of key rollover. - draft-ietf-dnsop-inaddr-required-04.txt. Title is misleading, should be something like "considerations for DNS reverse tree". Author is tired of fighting with people who have not read document and are just reacting to its title. Some feel that subject is still important. Jun-ichiro itojun Hagino kindly volunteered to work on this document if author is tired of holding the pen. - draft-ietf-dnsop-ipv6-transport-guidelines-00.txt. Has been done for a while, WG chairs have been lame about this one. Ready for WG last call, heading for BCP. - draft-ietf-dnsop-respsize-00.txt. Consensus of room was that this is good stuff and should ship without further ado. Ready for WG last call. Chairs forgot to ask WG whether this should be going for BCP or informational, will include that question in last call. - draft-ietf-dnsop-ipv6-dns-issues-02.txt. This one has turned into a grab bag of open issues related to DNS and IPv6. Somewhat dead in the water in its current form. Author has tried very hard to do whatever the WG wants with this, but lacks time to flog the WG at the moment (has more important things to do in real life). Pekka Savola kindly volunteered to help out with this document. Expired WG drafts: - draft-ietf-dnsop-serverid-02.txt. Significant interest in having something in this space, especially from root server operators. Known issues with mechanism proposed in current draft. Suzanne Woolf kindly volunteered to take over this project. - draft-ietf-dnsop-dontpublish-unreachable-04.txt. At least a few people thought this one was important and should not be dropped. - draft-ietf-dnsop-resolver-rollover-01.txt. Author has not been working on it recently, but with DNSSEC finally close to completion, we may have finally become topical again. Author will resubmit draft so that people can read it. - draft-ietf-dnsop-ohta-shared-root-server-03.txt. Author says that experiments with the techniques discussed in this draft are in progress. Author will resubmit draft so that people can read it. - draft-ietf-dnsop-keyhand-05.txt. Nobody spoke up for this draft. New work: - draft-guette-dnsop-key-rollover-requirements-00.txt. Some overlap with draft-ietf-dnsop-ipv6-transport-guidelines-00.txt, but this draft is about requirements in this space. Not many people in the room had read this draft. No objections to taking it on as WG work item. - Discussion of name server clock synchronization. Roy Arends performed some measurements in this space, sent summary to the mailing list. Has not generated a lot of discussion. This doesn't matter much for DNSSEC proper, but does matter for transaction signatures. Roy agreed to write up a brief draft describing the mechanism he used to perform these tests. DNS Discovery topic was left for second (Tuesday) session, per agenda as announced. Second session: 2003.11.11 13:00-14:00 -0600 Entire session devoted to DNS Discovery discussion. Summary: No clear consensus on anything. Largest single camp seems to think that just using DHCPv6 lite is enough and that we should pick a single mechanism and move on. Sizable (albeit smaller) camp believes that RA based discovery is important; for some reason there seems to be a fairly close correlation between folks who think that RA should be one of the solutions and folks who think that it's not necessary to pick a single solution. Well-known addresses have a few very vocal proponants as well as some determined critics. While other proposals have been surfaced on the list (as well as in previous discussions, some dating back many years), none received any serious discussion time at this meeting. All three of the proposals discussed at the meeting are documented. DHCPv6-lite is documented in a DHC WG (about more than just DNS discovery) which is in IETF last call. The several RA-based proposals have been consolidated into a single draft. There is also a draft for the current spin on the well-known address proposal. == . dnsop resources:_____________________________________________________ web user interface: http://darkwing.uoregon.edu/~llynch/dnsop.html mhonarc archive: http://darkwing.uoregon.edu/~llynch/dnsop/index.html
