[6tisch] Minutes, IETF 95 6TiSCH WG Meeting

["Pascal Thubert (pthubert)" <[e-mail]>]

Agenda and Meeting information

Meeting        :   IETF95 Monday, April 4, 2016 (ART)

Time           :   14:00-15:30 Monday Afternoon session I (90min)

Location       :   Room Buen Ayre B, Hilton Buenos Aires, Buenos Aires, Argentina

Chairs         :   Pascal Thubert <[e-mail]>

                   Thomas Watteyne <[e-mail]>

Responsible AD :   Brian Haberman

URLs           :   http://tools.ietf.org/wg/6tisch/

                   https://datatracker.ietf.org/wg/6tisch/

                   https://www.ietf.org/mailman/listinfo/6tisch

                   https://bitbucket.org/6tisch



Intro and Status                                 [5min]  (Chairs)



   Note-Well, Blue Sheets, Scribes, Agenda Bashing



New charter and status docs                      [20min] (Chairs)

   * Status Documents

   * Status 6lo / ROLL

   * New Charter

   * Milestones

   * Action Plan



Plugtests

   * Announcement ETSI 6TiSCH #2 Plugtests       [7min]  (Miguel Angel Reina Ortega)

   * Announcement ETSI 6TiSCH #3 Plugtests       [3min]  (Miguel Angel Reina Ortega)



Dynamic Scheduling

   * <draft-wang-6tisch-6top-protocol-00>        [15min] (Xavier Vilajosana)

   * <draft-dujovne-6tisch-6top-sf0-01>          [15min] (Diego Dujovne)

   * call for WG adoption



Security

   * status of the work and action plan          [10min] (Michael Richardson)



Unchartered items, time permitting

   * <draft-satish-6tisch-aodv-rpl-00>           [10min] (Satish Anamalamudi)



Any Other Business                               [2min]  (Chairs)

Resources

  *   agenda: https://datatracker.ietf.org/meeting/95/agenda/6tisch/
  *   presented slides: TODO
  *   Meetecho recording (audio+video): TODO
  *   Jabber log: http://www.ietf.org/jabber/logs/6tisch/2016-04-04.html

Volunteers

  *   Scribes
     *   Dominique Barthel
     *   Charlie Perkins
     *   Keoma Brun-Laguna
  *   Jabber
     *   Ines Robles

Minutes

  *   [14.00] [expected: 14.00] Meeting starts
  *   [14.00] [expected: 14.00] Intro and Status (Thomas Watteyne) [5min]
     *   change in agenda: Tero will present IEs
     *   see draft-kivinen-802-15-ie-00.txt
  *   [14.03] [expected: 14.05] (Tero Kivinen) [5min]
     *   16 IE codes available, IETF can get one, to share between WGs.
     *   subtypes to be allocated by IANA.
     *   Tero requests for comments, on 6lo or 6TiSCH mailing list.
     *   Q&A: Tero suggest one framework on how IETF uses its assigned code. Will chase AD's to find a sponsor for this draft.
     *   Action item: Chairs to coordinate about AD sponsoring
  *   [14.07] [expected: 14.10] New charter and status docs (Pascal Thubert) [20min]
     *   -minimal draft: answered all INT AREA comments. Expect feedback fron INT DIR.
     *   6lo routing header split into one draft at 6lo and one at ROLL. Both now in last call.
     *   Gabriel asking for feedback about 6lo-dispatch
     *   Hop by Hop at 6man: 6man considering preventing any new use of HbH
     *   6TiSCH WG has rechartered. Thanks Brian!
     *   Current work items 6P, SF0, secure bootstrap, updating architecture doc, requirements to Detnet. YANG data model differed.
     *   Plugtest in Berlin [very important], if you have an implementation, please join.
     *   need to really get momentum on the security aspect.
  *   [14.16] [expected: 14.30] ETSI 6TiSCH #2 and #3 Plugtests (Miguel Angel Reina Ortega, presented by Thomas Watteyne) [5min]
     *   overview of past event in Paris. 12 companies. 14 test cases. Main point was 6top protocol and 6lo RH.
     *   did not test SF0, considered was not mature enough for interop.
     *   next event in Berlin, scope not completely defined. Probably SF0 and BB routing (although not on the slide)
     *   Check actual dates: July 16-17?.
  *   [14.21] [expected: 14.35] draft-wang-6tisch-6top-protocol-00 (Xavier Vilajosana) [15min]
     *   renaming of -sublayer draft
     *   renamed some fields, added 3-way transaction
     *   6top sits between 15.4e and next upper layer
     *   intended use is for 6top to run a different slot frame from -minimal, concurrently.
     *   in 3-step mode, requester only requests number of cells, and receiver suggests list of cells.
     *   Thomas: this draft needs update to take into account new proposed management of IEs at IETF.
     *   see draft-kivinen-802-15-ie-00.txt
     *   metadata for extra info to be exchanged between SF, unknown to 6top
     *   3rd step is called Confirmation, same format as response
     *   draft has guidelines for SF designers. ANy SF has to specify which statistics 6top should gather for the SF to be able to make its decisions
     *   Slide about "IEEE Liaison Considerations" will need changes in view of new draft from Tero Kivinen re: IEs
     *   Thomas: this is -00 but actually renaming of a quite mature -sublayer draft. Will do preliminary call for adoption after next presentation.
     *   Q&A: [no questions]

*        [14.35] [expected: 14.50] draft-dujovne-6tisch-6top-sf0-01 (Diego Dujovne) [15min]

     *   no longer supposes that application explicitely requests bandwidth.
     *   Blacklist also possible in addition to white list of cells.
     *   Bandwidth Estimation algorithm: do a little over-provisioning because guessing from observed traffic.
     *   Relocation request
     *   Allocation policy translate bandwidth allocation into cells (taking into account each cell's PDR).
     *   CellList / WhiteList

o   TODO:

o   Timestamp

o   Metadata

o   node behaviour at boot. Extra cells for speeding up the boot process, to be removed later on.

*        Q&A: * Xavi on Jabber: will the draft provide recommended value for threshold? DIego: not at this point. Thomas on jabber: I believe we should have recommended values in there, to insure interoperability, although it might not satisfy everything, RFC6550 includes RECOMMENDED values * Pascal: over-provisioning. For each child or for all children? Diego: will think about it. * Thomas 256 values for SF ids. SF0 is expected to be the simplest one.

  *   [14.46] [expected: 15.05] call for WG adoption (Chairs) [5min]
     *   6top and SF0 are companion drafts. Been worked on for months, just recently appeared on the charter.
     *   Pascal: (6top) anybody opposes WG adoption for 6top adoption? Raise hand if you thinks it time to approve (15 hands). Any opposition? none same questions regarding SF0: Approve? ~7 hands. Opposed? none.[to be confirmed on the mailing list]

*        [14.49] [expected: 15.10] status security (Michael Richardson) [10min]

     *   From the 6tisch point of view non have changed
     *   JCE changed named to ANIMA register
     *   assumes manufacturer installed certificate
     *   A lot of the work have changed to Wordking Group ??

o   ANIMA and 6TiSCH contrasts:

        *   ANIMA use HTTPS while 6T prefers DTLS/CoAP + 6top

o   ANIMA is not really constrained compared to 6TiSCH

o   ANIMA and NETCONF contrasts

     *   6TiSCH vs NETCONF:
        *   both use YANG
        *   ...
     *   Join Problem:
        *   new, potentially malicious, nodes dropped in the network without breaking everything.
     *   How to tunnel traffic from new node to registrar through Join Assistant Discussed at length in ANIMA. see draft-richardson-anima-state-for-joinrouter-00
     *   Q&A:
     *   Thomas: when can we start using stuff from ANIMA?
     *   Michael: June.
     *   Pascal: reform Design Team?
     *   Michael: first have running code and see what happens. We have written enough documents.

*        [15.05] [expected: 15.20] draft-satish-6tisch-aodv-rpl-00 (Satish Anamalamudi) [10min] Presented by Mingui Zhang on Satish'es behalf

     *   Two routes instances for route discovery
     *   extends RPL. One instance for Route Discovery from Source do Dest, another one for the reverse direction.
     *   new "A" flag in DIO to tell if link is asymmetrical.
     *   if all links symmetrical, destination will receive DIO with A flag reset: instance 2 will use the same links as Instance1. If any link is asymmetrical, destination will receive DIO with A=1. Destination will multicast Instance2 message.
     *   no address vector in response message, bandwidth saving.

o   Q&A:

?  Pascal: 6TiSCH is INT area WG. This should be presented at ROLL for potential adoption. 6TiSCH can use it later.

?  Pascal: this is evolution of P2P-RPL (experimental track). This is standards track. Can you summarize changes, improvements? Charlie Perkins: standards track: we want to have implementations, P2P experimental has been out for some time, people understand it now.

?  Satish: we avoid address vector completely.

?  Ines: come to ROLL meeting Tuesday 4:20pm, we have an open floor slot at the end of the meeting, you can squeeze in and briefly mention your work.

?  Thomas: You addressed a lot asymmetric links, what exactly defines an asymetrical link, is it the path or the cells. => resource level asymetry

?  Pascal: now confused. Allocation of resource is after path has been established.

        *   [15.22] [expected: 15.30] Any Other Business (Chairs)
           *   Pascal expresses thanks to those who helped get the IE discussion done. Pat Kinney, Tero, Charlie, ..
        *   [15.23] [expected: 15.30] Meeting ends
Cheers,

Pascal