Skip to main content

Minutes interim-2018-6tisch-02: Fri 16:00
minutes-interim-2018-6tisch-02-201802161600-00

The information below is for an old version of the document.
Meeting Minutes IPv6 over the TSCH mode of IEEE 802.15.4e (6tisch) WG Snapshot
Date and time 2018-02-16 15:00
Title Minutes interim-2018-6tisch-02: Fri 16:00
State Active
Other versions markdown
Last updated 2018-02-16

minutes-interim-2018-6tisch-02-201802161600-00

DRAFT

Agenda, 16 February 2018 interim, 6TiSCH WG

Connection details

Present

  1. Xavi Vilajosana
  2. Thomas Watteyne
  3. Pascal Thubert
  4. Keoma Brun-Laguna
  5. Georgios Papadopoulos
  6. Yasuyuki Tanaka
  7. Mitsuru Kanda
  8. Malisa Vucinic
  9. Michael Richardson
  10. Fabrice Theoleyre
  11. Vishnu.n
  12. Dominique Barthel

Present last time

  1. Diego Dujovne
  2. Jonathan Munoz
  3. Lotte Steenbrink
  4. Mitsuru Kanda
  5. Remous-Aris Koutsiamanis
  6. Tengfei Chang
  7. Yasuyuki Tanaka
  8. Simon Duquennoy
  9. Remy Liu
  10. Rahul Jadhav
  11. Carsten Boorman

Taking notes (using Etherpad)

  1. Xavier Vilajosana
  2. Pascal Thubert

Action Items

Agenda

  • admin [ 5min]
  • News from ROLL and 6Lo [ 5min]
  • Network and DODAG selection (Pascal) [25min]
  • security update (Michael Richardson) [10min]
  • 6LoWPAN Fragmentation (Thomas Watteyne) [10min]
  • Any Other Business [ QS ]

Minutes

7:05 * admin [ 5min]
* meeting starts
* agenda is approved. Last call minutes are approved
* preliminary agenda for ietf 101 published today
* 90 minutes slot at the iet101
* 5th of March cutoff date
* calling for presentations at the ietf meeting
* MR: plans for future interop events?
* TW: will be announced during the ietf 101.
* will be somewhere in europe between 101 and 102

7:10 * News from ROLL and 6Lo [ 5min]
* Network and DODAG selection (Pascal) [25min]
* Blah
* Blah

* rank used to determine to what node to join.  
* the goal is to determine what EB belongs to what network. 
* PANID? is not good because networks do not negotiate about PANIDs and hence can collide.  There is no guarantee that they are different if they are from different networks.
* MR: question mark needs to say it is the pledge.
* MR: green and blue they are different networks.
* the goal is to identify the network in the EBs, this is similar 
  to adding a signature in the network.
* when we have different dodags coexisting we need to make sure that a node
  can join to both according to its needs.
* The DIO may send some information about the traffic is handling.
* MR: 6Tisch is proposing a problem statement to ROLL. 
* use something else that the Join Proxy selection  and this requires to enhance the EB content. This can be metrics that cannot be exposed by ROLL. 
* We need to turn metrics into DODAG rank,
* TW: we need to split the problems. First one is network selection based on the PANID. The second point is to enrich the join metric with information that includes other metrics such as the load of the network. The first is a problem that needs to be addressed. The second one is a very complex problem.
*PT: we need to announce metrics that provide awareness of the dodag level conditions.
*PT: From a ROLL perspective we can provide metric containers.
*PT: we need also to provide some information about the capabilities of the JP so a pledge can decide which one to join. 
*
  • security update (Michael Richardson) [10min]

    • MR: updates on the security work
    • MR presents a picture
  • 6LoWPAN Fragmentation (Thomas Watteyne) [10min]

    • TW presents fragmentation design team update
    • produce 2 documents that summarize fragmentation and to add fragment recovery.
    • draft published today. Discussion whether should go to LWIG or 6Lo.
    • submit them at 6lo as an initial step;
    • present them at IETF 101.
    • presents the draft just pushed: 6lo-minimal-fragment-00
    • reassembly is done at every hop in current fragmentaiton impl.
    • fragments are reassembled at aevery hop.
    • virtual reassembly buffer a stores the mapping for input and output fragments with their tag.
    • tags are changed at every hop, because you cannot enforce unique tags in the whole network.
    • the outgoing tag is selected by the node relaying it.
  • Any Other Business [ QS ]
    No OB