IAB, IESG and IEEE 802 Executive Committee Minutes of the 29 September 2015 Virtual Meeting Reported by: Amy Vezza, IETF Secretariat ATTENDEES ------------------- Mikael Abrahamsson Jari Arkko Alia Atlas Mary Barnes Deborah Brungard Ben Campbell Benoit Claise Alissa Cooper Subir Das Spencer Dawkins Ralph Droms Donald Eastlake Janos Farkas Eric Gray Bob Grow Craig Gunther Brian Haberman Ted Hardie Bob Heile Marc Holness Russ Housley Pat Kinney Jouni Korhonen David Law Barry Leiba John Messenger Kathleen Moriaty Paul Nikolich Erik Nordmark Alvaro Retana Dan Romascanu Jon Rosdahl Dorothy Stanley Andrew Sullivan Pat Thaler Pascal Thubert Brian Trammell Juan Carlos Zuniga REGRETS ------------------- Cindy Morgan Glenn Parsons 1. Roll Call, Agenda Bashing, Minutes Approval 2. Review of Other Action Items from Previous Meetings (not part of the shared items list) - Juan Carlos Zuniga to send mail to the coordination list about the next Privacy SG meeting. [DONE] - Benoit Claise to put the YANG doctors in touch with Marc Holness regarding shared item 27. [DONE] 3. 2016 Face to Face meeting Dan Romascanu started the discussion of a face-to-face meeting in January of 2016 to co-locate with the IEEE big interim meeting week in Atlanta, GA. The date proposed was January 16. Dan proposed a number of possible discussion topics for the proposed face-to-face. There was discussion whether a face-to-face meeting was necessary for the topics proposed, as well as if a face-to-face meeting just before the IESG and IAB leadership change would be useful. Russ Housley mentioned that face-to-face meetings are useful to create and maintain informal connections and collaborative efforts. Jari Arkko agreed that face-to-face meetings were valuable to both communities. Pat Thaler mentioned if the group wanted to meet face to face a different time of year it would need to be planned as a separate event and not in conjunction with something already on the calendar, unless the group looked to co-locate with the IAB/IESG retreat usually scheduled for spring. The group decided that a virtual meeting for January 2016 would be a better use of resources and Dan Romascanu agreed to run a doodlepoll for scheduling the meeting. 4. HOMENET and IEEE 802.11 (and other 802) multicast - Mikael Abrahamsson and Adrian Stephens (slides) with John Messenger standing in for Adrian Stephens Mikael Abrahamsson spoke about the IP and packet transport on multicast on 802.11 networks. He touched on the history of IP over 802.3 wired networks as well as on 802.2 ethernet in relation to packet loss. Mikael mentioned end users just want the Internet to work, most have no knowledge of how it works. Mikael mentioned that IPv6 Makes use of multicast for its control plane, and are delay tolerant, but sensitive to packet loss. In HOMENET from a routing protocol point of view some wanted to use IS-IS which is layer 2 multicast. Mikael talked about how the people doing mesh networking are putting pressure on the IETF because multicast doesn't work on wireless as they see it. They need to apply multicast reduction due to high packet loss and in order to achieve energy efficiency. They are forced to change the IP stack and use unicast instead. When questioned why they didn't talk to 802.11 about the issues they had making it work, they indicated they had no contact. Mikael spoke about IPv6 being more adversely affected by multicast and broadcast packet loss compared to IPv4 which caused some to voice opinions about limiting use of broadcast/multicast. John Messenger discussed the 802.11 properties about loss of packets through collision, interference or poor rate adaptation decisions or changes in the radio channel. Unicast traffic can be retried which improves reliability, but multicast does not get retried. John talked about how multicast/broadcast traffic is sent a basic rate which might be as low as 6 Mbps when unicast is operating at 600 Mbps. John mentioned new options not yet commonly implemented like groupcast with retries (GCR) and the flexible multicast service that allows a multicast rate higher than basic. These are being added to 802.118ak which will come out next year with the revision. John discussed how both unicast and multicast can be delayed by power saving mechanisms. Higher layer packets can get discarded due to buffer limitations in the AP and non-AP. John talked about the options implementers have. An 802.11 AP manufacturer could optimize multicast-related behaviors by turning multicast into unicast traffic or ARP proxying. Juan Carlos Zuniga commented that the GCR (groupcast with retries) is not commonly implemented or certified yet. He wanted to know if between the IETF and 02.11 if GCR would solve the issues for multicast over IPv6. Pascal Thubert mentioned there are Internet-Drafts in the IETF that discuss these issues and they are more complex. Erik Nordmark mentioned that it would be useful to understand the different parameters, not just who is working to solve the problem but what the problem is. Pat Thaler talked about a problem to be addressed with regard to multicast important for operation of basic protocols. For IP to work. The slides are available at: https://www.iab.org/wp-content/IAB-uploads/2013/01/multicast-problem-statement.pptx Juan Carlos Zuniga commented that the GCR (groupcast with retries) is not commonly implemented or certified yet. He wanted to know if between the IETF and 802.11 if GCR would solve the issues for multicast over IPv6. Pascal Thubert mentioned there are Internet-Drafts in the IETF that discuss these issues and they are more complex. Erik Nordmark mentioned that it would be useful to understand the different parameters, not just who is working to solve the problem but what the problem is. Pat Thaler talked about a problem to be addressed with regard to multicast important for operation of basic protocols; for IP to work. Which is a different problem from the streaming problem. The discovery protocols and the protocols necessary for IPv6 to work over wireless should be the focus. Pat wanted to know if there was an IETF working group that would be appropriate for this work. Dan Romascanu mentioned he thought the Internet Area would be appropriate. Ralph Droms suggested the 6MAN working group. Mikael Abrahamsson suggested V6OPS, but would need 802.11 experts to contribute. Russ Housley mentioned having a directorate with experts from each side to advise whichever working group ended up writing the document. Brian Haberman talked about the various working groups that have brought up the subject in their meetings. V6OPS, MBONED, INTAREA, and HOMENET. He mentioned that any would be appropriate to have someone come and discuss the 802.11 functionality that exists already. Dan Romascanu thought the INTAREA would be appropriate. NEW ACTION ITEM: Mikael Abrahamsson (IETF) and Dorothy Stanley (802.11) to coordinate for a presentation of a discussion of 802.11 functionality for the INTAREA at IETF 94. Juan Carlos Zuniga as INTAREA WG co-chair to put it on the WG's agenda for IETF 94. 5. November 2015 meetings - short pass on the IETF BOFs and IEEE 802 PARs that may be relevant to IEEE 802 and IETF coordination - http://trac.tools.ietf.org/bof/trac/wiki/WikiStart# - http://www.ieee802.org/PARs.shtml - Tutorials Barry Lieba gave a short description of the BoF ISS for IETF 94. ISS is looking to build a standard protocol for synchronization for Internet storage (for example dropbox). Dan Romascanu asked about new 802.11 PARS for November. Bob Heile spoke about a consolidated LLC for 15.4 in 802.15 for next March, not for the November meeting. He mentioned coordination needed between 802.15 and 802.1 as well as the higher level L2 stuff in IETF. Pat Thaler talked about new PAR in 802.1 for doing an address distribution for layer 2 for the local addresses. Janos Farkas mentioned URN name space in YANG that is being circulated for November. Pat Thaler discussed 802.3 putting forward PARs might be the study group for next generation ethernet passive optical networking. Bob Grow agreed they were ready to bring a PAR forward to target deliver data rates between 25 gigabits per second and 100 gigabits per second using the PON multi-drop architecture. Pat Thaler also mentioned the two and a half and five gigabits back plane and short reach study group bringing a PAR forward for November. 6. Areas of shared interest between the IETF and IEEE 802. Review of the list of items, status and action items - discussion led by Pat and Dan Item 3: IETF NVO3 and IEEE 802.1 DCB Pat Thaler reported that the PAR was approved and are working on developing the PAR. Currently she are working on identifying an editor to start work on the draft. NEW ACTION ITEM: Pat Thaler to identify an editor to begin work on the draft. Item 5: Enabling use of Local Addresses for virtualization and IoT (was: Effect of virtualization on IEEE 802 architecture) Nothing to report. Item 11: 11. IETF and IEEE 802.1 OmniRAN TG Juan Carlos Zuniga updated item 11 for Max Riegel. The presentation is available here: https://www.iab.org /wp-content/IAB-uploads/2015/09/omniran-15-0046-00-00TG-sept-2015-status-report-to-802wgs.pptx 19. Common OAM proposal / Layer Independent OAM The latest status, as of Sept 14th is http://www.ietf.org/mail-archive/web/lime/current/msg00347.html Also, from the charter, note: - "This usability and extensibility must be demonstrated, for example with IP Ping, traceroute, BFD, and LSP Ping." -> this is the basis for the applicability statement - "Therefore, it is anticipated that the working group will closely coordinate its activities with other SDOs (including, but not limited to the ITU-T, MEF, IEEE, BBF and 3GPP) to ensure that the generic models are harmonized with work done in those SDOs and are applicable to many technologies." Item 21: 6tsch Bob Heile for 802.15 updated on 6TISCH. He mentioned 15.4 is in sponsor ballot and will help with the work in the IETF 6TISCH WG. 802.15 is fully behind it. Pascal Thubert added with 802.15.4 work on the six top layer. He also discussed the need for IETF code point for IEEE. 22. CAPWAP extensions in OPSAWG draft-ietf-opsawg-capwap-alt-tunnel (Alternate Tunnel Encapsulation for Data Frames in CAPWAP) will merge with draft-you-opsawg-capwap-separation-for-mp (CAPWAP Control and Data Channel Separation for Multi-provider Scenario) Item 24: Area Name - coordination between the IETF and IEEE 802 on Pervasive Monitoring Juan Carlos Zuniga spoke about how on the IETF side they carried out trials on the wifi privacy and the set up is now permanent. The same thing happened in the IEEE 802 meetings to keep the set up as permanent. The work in the specific 802 study group, they wrote a PAR that got approved by NETCOM (?) and the standards board at IEEE so now there is project 802E about privacy recommendations for 802 protocol developers. The work will be done by the 802.1 security group meeting at the upcoming Dallas meeting in November. The 802E specification will be available to all 802 developers. Item 25: Area Name - Layer2/Layer 3 Interaction for Time-Sensitive Traffic The DETNET WG in IETF is up for approval at the next IESG Teleconference. Item 26: Area Name - IS-IS extensions for IEEE 802.1Qca Janos Farkas updated the group with the IEEE 802.1Qca document has been approved. They are waiting on the corresponding IETF documents to be approved and assigned RFC numbers. 27. Area Name - development of YANG models in the IEEE 802 - Mahesh, as YANG doctor, is contact with Mark Holness. All IEEE 802.1X and 802.1Q YANG [draft] models now compile cleanly (using the Ðietf option in Pyang v1.5), with the exception of the namespace naming warning. - Marc will post the IEEE [draft] YANG models into GitHub at https://github.com/YangModels/yang/tree/master/ this week - IETF draft for IEEE URN, by Mahesh. Glen wanted to have someone from within IEEE submit the draft, so at this point I have handed over the draft to him. This was discussed during the interim meeting two weeks ago, and could not agree on the tree structure of the URN. Anyway, it will be administered by the IEEE and a project will be opened in November - Need to get back to Robert Wilton on his ÒInterface VLANÓ YANG model draft. Quoting Mark Holness: "I believe there were initial concerns that this work overlapped with the work being done in 802.1. Mark needs to close the loop on this with Glenn Parsons. I donÕt think there is overlap, but Glenn has some concerns." - Formation of at least an initial team for IEEE 802.3 YANG model. Owner: Mahesh and David Law - There is a YANG tutorial at the next IETF meeting, on Sunday. 7. Review of Action Items New Action Items: Dan Romascanu will run a doodlepoll to choose a date for a virtual meeting in January 2016. Dan Romascanu and Pat Thaler to work on finding a time for a face to face later in 2016. Mikael Abrahamsson (IETF) and Dorothy Stanley (802.11) to coordinate for a presentation of a discussion of 802.11 functionality for the INTAREA WG at IETF 94. Juan Carlos Zuniga as INTAREA WG co-chair to put it on the WG's agenda for IETF 94.