Clone wiki

meetings / 140117_webex

Minutes Webex 17 January 2014, 6TiSCH WG

Note: timestamps in PST.

Taking notes (using Etherpad)

  1. Xavi Vilajosana
  2. Thomas Watteyne
  3. Pascal Thubert
  4. Dominique Barthel

Present (alphabetically)

  1. Alaeddine Weslati
  2. Diego Dujovne
  3. Dominique Barthel
  4. Guillaume Gaillard
  5. Elvis Vogli
  6. Kuor Hsin Chang
  7. Maria Rita Palattella
  8. Michael Richardson
  9. Pat Kinney
  10. Paul Duffy
  11. Pascal Thubert
  12. Pouria Zand
  13. Raghuram Sudhaakar
  14. Rene Struik
  15. Rouhollah Nabati
  16. Sedat Gormus
  17. Shitanshu Shah
  18. Tengfei Chang
  19. Thomas Watteyne
  20. Xavi Vilajosana



Action Items

  • Shitanshu to forward announcement of draft to 6TiSCH ML.


  • Administrivia [4min]
    • Approval agenda
    • Approval minutes last call
    • IETF 89: requests for sessions
  • updates [10min]
    • update security discussion
    • update plugfest
    • update YANG model and chunk definition
  • deterministic communication [Shitanshu Shah] [15min]
  • draft-ietf-roll-rpl-industrial-applicability [Michael Richardson] [10min]
  • use cases and ToS bits [Pascal Thubert] [10min]
  • AOB [1min]


  • [08.05] Meeting starts
  • [08.06] Administrivia
    • Approval agenda
      • [Thomas] Pat Kinney will make an announcement about the IEEE 802.15 "802.15.IG6T" IETF Liaison 6TiSCH Interest Group (IG 6T)

        No other issues raised. Agenda approved.

    • Approval minutes last call

      No issues raised. Agenda approved.

    • IETF 89: requests for sessions
      • Today is deadline for requesting slots at IETF89
      • 2 sessions requests:
        • WG meeting (90min)
        • plugfest (3h)
      • Suggested to be on the same day, but all depends on scheduling.
      • Preliminary agenda published on 01/31.
    • 802.15.IG6 [Pat]
      • Pat Kinney chairs different groups, including Vice Chair IEEE802.15.
      • IEEE 802.15 "802.15.IG6T" IETF Liaison 6TiSCH Interest Group (IG 6T) was created
      • see announcement at
      • Goals
        • Keep the interest going and act as liaison see how to help.
        • What 6TiSCH changes in 15.4e, help as much as IEEE can. Also relay to 6TiSCH, a 2 way street.
        • Discussing about 6tisch is and discuss what matches 802.15.4e, goal to enable 6tisch as much as they can
        • Listen for inputs and see what we can do
        • Get 6TiSCH requirements for changes on IEEE802.15.4e so both sides can improve.
      • First meeting on Monday.
    • [Thomas] Is the idea that most us join the reflector?
    • [Pat] Yes, the reflector is open, and the reflector will be used to inform about the activity of the IG.
  • updates
    • update security discussion [Michael Richardson]
      • Want to have a security process a least similar to ISA100 and WirelessHART. Actively looking at ZigBee IP profile and PANA, trying to understand byte space requirements and overhead.
      • Typical deployments are created in rooms before deployment so networks might collapse. This imposes more challenges on the security part.
    • update plugfest [Xavi Vilajosana]
      • [Thomas] Xavi offered to serve as a chair. Great support from mailing list for nomination. Rough consensus on nomination from WG.
      • Xavi e-mailed several people, already got confirmation that several people will come with hardware.
      • 3 types of activities
        • interoperation:
          • basic TSCH / minimal network formation
          • add RPL
          • add OTF scheduling
        • demonstration
        • tools:
          • Any tool that can be shared with the community is of interest
      • [Thomas] there is now a page linked from the home page in bitbucket
      • [Xavi] needs a description of the demo / test to organize an agenda for the plugfest.
    • update YANG model and chunk definition
      • Qin is not present (travelling). Thomas suggests to take the discussion to the mailing list
      • [Xavi] We uploaded yesterday the YANG model. It is in the bitbucket repo as a .yang file.
      • [Thomas] Deadline for all drafts including -00 is 2/14, one month away.
      • Thomas asks Xavi to submit a -00 soon and then make an update based on comment before deadline.
      • Xavi confirms that this will be done in that way.
  • [08.27] Deterministic Forwarding PHB [Shitanshu Shah]
    • See
      • draft presented at TSVWG WG (transport area WG)
    • goals:
      • agree on the requirements and need for a new DSCP (DiffServ Code Point)
      • if we agree, bring to TSV WG which will have to decide on adoption
    • motivation:
      • Need for a new DiffServ Code Point (DSCP) for deterministic that is different from low latency
      • forwarding requirements for time sensitive traffic
      • Expedited Forwarding (EF) PHB (Per-Hop Behavior) does not guarantee time scheduled service
      • New Deterministic Forwarding (DF) PHB proposed
    • use cases:
      • communication between nodes in a single LLN
      • communication between nodes in different LLNs, through the backbone
    • DF PHD
      • not mandating any specific method for provisioning (command line interface, signaling protocol, etc.)
      • provisioning of fixed/relative time for scheduling
      • provisioning of max data to be transmitted
    • Classification
      • packet inspection, classification in the DF class
      • maybe simple L2 or L3 header, maybe deep packet inspection
      • [Pascal] may be hard to do classification to find relevant information. Work happening at 802.1 TSN (Norm Finn) which general use MPLS label as flow identifier. Will find that early in the packet.
      • [Shitanshu] May require admission control. If there is a way not to have deep packet inspection, can help.
    • Scope:
      • in end-to-end path, if domain that does no support DF, proposal is to aggregate in EF PHB.
    • [Pascal] ECN set when congestion. We can combine with flow label and/or DSCP.
    • [Thomas] Can you send information to ML?

      Action item: Shitanshu to forward announcement of draft to 6TiSCH ML.

  • [08.49] draft-ietf-roll-rpl-industrial-applicability [Michael Richardson]
    • roll WG has some applicability statements in different env.
    • one is on industrial env.
    • security area review. 4 issues raised and logged, see this meeting slides.
    • draft says RPL can be used in a number of the phases
    • concern: state of the technology: operating best effort in industrial is hard. Applicability statement says that we can use RPL for boostrapping, emergency, but not for actual operation of the plant.
    • For control loops RPL does not seem adequate? It is adequate for monitoring applications including industrial.
    • The applicability statement does not make this statement clear enough.
    • ROLL WG Limited ability to review and analyse the document. Concern that ROLL is not able to provide an accompanying set of documents and efforts to support industry on that use case.
    • [Thomas] The statement that RPL is not applicable is not applicable in industrial settings? In more distributed approaches RPL is still useful.
    • [Michael] there are industries where there is no hard-control (never will use 6tisch).
    • [Pascal] RPL applies in industrial comes with 6tisch:
    • [Michael] Wants to know if there is any case where RPL can be used for industrial applications without 6TiSCH. If the answer is NO, then applicability statement in industrial scenarios of RPL won't make sense in the scope of ROLL but in the scope of 6TISCH.

      Action item: Michael to start thread on ML.

  • [09.14] use cases and ToS bits [Pascal Thubert]

    We are running out of time. Pushed to next call.

  • [09.15] AOB

    No other business raised.

  • [09.15] Meeting ends