DNSSD WG Agenda IETF100, Singapore Wednesday, November 15, 2017 9:30-12:00 Local Time Canning Room Chairs: David Schinazi, Tim Wicinski (sitting in for Tim Chown) Minutes: Barbara Stark, Bjoern Metzdorf Jabber: Mikael Abrahamsson Chairs' Introduction Chairs, 5 mins Report on the Last Call for draft-ietf-dnssd-hybrid Stuart Cheshire, 10 mins https://tools.ietf.org/html/draft-ietf-dnssd-hybrid Update on draft-ietf-dnssd-push and draft-ietf-dnsop-session-signal Stuart Cheshire, 20 mins https://tools.ietf.org/html/draft-ietf-dnssd-push https://tools.ietf.org/html/draft-ietf-dnsop-session-signal (dnsop) Update on documents created before IETF 99 Stuart Cheshire, 20 mins https://tools.ietf.org/html/draft-cheshire-dnssd-roadmap https://tools.ietf.org/html/draft-sctl-service-registration https://tools.ietf.org/html/draft-sctl-discovery-broker https://tools.ietf.org/html/draft-sctl-dnssd-mdns-relay Report on the DNSSD testing at the IETF 100 Hackathon Stuart Cheshire, 15 mins draft-ietf-homenet-simple-naming Ted Lemon, 30 mins https://tools.ietf.org/html/draft-ietf-homenet-simple-naming draft-eckert-anima-grasp-dnssd Toerless Eckert, 15 mins https://tools.ietf.org/html/draft-eckert-anima-grasp-dnssd DNS-SD Privacy drafts Stuart Cheshire, 30 mins https://tools.ietf.org/html/draft-ietf-dnssd-privacy https://tools.ietf.org/html/draft-ietf-dnssd-pairing https://tools.ietf.org/html/draft-ietf-dnssd-pairing-info https://tools.ietf.org/html/draft-cheshire-dnssd-privacy-considerations Chairs' Conclusion Chairs, 5 mins -------------------------------------------------- Chair Slides https://datatracker.ietf.org/meeting/100/materials/slides-100-dnssd-00-chairs-intro/ David presented the slides. Round of applause from the room to thank Ralph for his years of service as chair. Agenda was not bashed. -------------------------------------------------- Report on the Last Call for draft-ietf-dnssd-hybrid Update on draft-ietf-dnssd-push and draft-ietf-dnsop-session-signal Update on documents created before IETF 99 Report on the DNSSD testing at the IETF 100 Hackathon https://datatracker.ietf.org/meeting/100/materials/slides-100-dnssd-01-stuart-doc-status/ Stuart Cheshire presented. There were no comments related to any of the status provided. It was noted this was a very fast 65 minutes of scheduled discussion, since it only lasted about 10 minutes. -------------------------------------------------- draft-ietf-homenet-simple-naming https://datatracker.ietf.org/meeting/100/materials/slides-100-dnssd-02-ted-homenet/ Ted Lemon presented. Toerless Eckert: Is there any insight that homenet might help to get more adoption of DNSSD? Stuart Cheshire / Ted: There may be some deployments happening. Stuart: Cautiously optimistic. Anecdotal evidence in Stuart's home shows adoption of multicast DNS. Toerless: The examples Stuart was giving aren't relevant to homenet specifically driving adoption. They would happen without homenet. Ted: Homenet is trying to make sure DNSSD works reliably in a home network. Mikael Abrahamsson: What homenet has is no professional management, and it's trying to define how devices trust each other in this unmanaged environmnent. Stuart: Homenet works in more than one way. There are many different physical layer technologies in use. There is a need for relays and proxies in gateways. But we need the capabilities defined so they can exist even if some gateways don't get updated. Link to source code of dnssd relay implementation: https://github.com/Abhayakara/dnssd-lite Note that the work that was done on Saturday and Sunday hasn't been committed to the repo yet; I will do that before I leave Singapore, but might not be until Friday, because we had a hard stop and the code doesn't compile at the moment. ---------------------------------------------------- draft-eckert-anima-grasp-dnssd https://datatracker.ietf.org/doc/slides-100-dnssd-03-toerless-grasp/ Toerless Eckert presented. Toerless mentioned that a new updated set of slides will be uploaded in half an hour. Mikael channeled jabber room: mic: Slide 2 is unclear. Are they saying link-local multicast is not available ? or merely that there is no multicast forwarding off-link. Toerless provided an answer which ran did not think answered the question: mic: did not answer the question. Is link-local multicast available or not ? mic: if link-local multicast is not available, how does IPv6 ND work ? Brian Carpenter: Stuart: It would be helpful if you could provide some background as to what types of devices are these? How are they connected? 802.15.4? Toerless: Anima was designed to be a well-managed network. Stuart: When you're talking about well-managed, are we talking about really expensive routers? Toerless: Yes. Toerless and Stuart then proceeded to have a discussion to try to better understand anima and how it differs from homenet. Brian: Anima is for the "Plug-n-play" enterprise. This may help as background for Toerless's talk in progress: https://github.com/becarpenter/graspy/blob/master/AN-overview.pdf Back to the slides. Toerless picked up going through the slides again at this point, starting with slide 4. Stuart: Offer, Discover and Use are the 3 basic operations provided by dns-sd. Stuart: Having service instance names as a unique identifier should not be optional. They are designed for machine consumption, as well. If you don't use service names, then you need another hidden identifier. Toerless: Are "host-names" actually needed? Stuart: Host-names were basically inherited from the DNS SRV record format and are helpful if a device has multiple interfaces with multiple addresses, even across address families. Did not want to re-invent SRV records. Toerless resumes the presentation (slide 5). Stuart: Thanks Toerless for the presentation and expresses gratitude that anima adopted the same IANA registry values. Brian Carpenter: GRASP will in addition need its own IANA namespace for its own specific purposes. ---------------------------------------------------- DNS-SD Privacy drafts https://datatracker.ietf.org/meeting/100/materials/slides-100-dnssd-04-stuart-privacy/ Stuart Cheshire presented. Dave Robin: Do message lengths also reveal information? Do we need to add padding/randomness or something similar like other protocols do? Mikael: Could you provide examples to the email list? Stuart: Examples are not in the draft right now. David Schinazi: Should we be looking into privacy? It sounds like we should. Ted Lemon: We should probably ask for a hum to see what people think. Tim Wicinski: The only doc not yet adopted is draft-cheshire-dnssd-privacy-considerations. Stuart: The authors of the other drafts said they would be ok to put some of the work on hold while we get more discussion. Aanchal Malhotra: Do the current drafts meet the goals for privacy? Stuart: I think they do not. Dave Thaler: Section 2 of the DNSSD privacy draft proves that this is interesting. Should the privacy considerations be put into their own document? Section 2 might not be sufficiently covering the subject of privacy. Alp Toker : Human rights issues are also important to consider here. Stuart agrees. David Schinazi: Should we be spending more time on privacy requirements? Humming is in favor of "yes". There were no hums against. Andrew: The humming for spending time was not very loud, so it would be good to know who is willing to work on it. Ted Lemon, Bernard Wolz, Andrew Sullivan, Alp Toker, Aanchal Malhotra raised their hands to volunteer to help work on this. ------------------------------------------------------ Chair's Conclusion David summarized the meeting. Stuart: We would like to do work at the hackathon again in London and would like more people.