Wiki
Clone wikimeetings / 140829_webex
Minutes Webex 29 August 2014, 6TiSCH WG
Note: timestamps in PDT.
Connection details
- Webex: https://ciscosales.webex.com/ciscosales/j.php?ED=219615007&UID=481905242&PW=NZTRkNDAwOTE1&RT=MiMyMw%3D%3D
- Etherpad: http://etherpad.tools.ietf.org:9000/p/6tisch?useMonospaceFont=true
- Topic: 6TiSCH Weekly
- Time: 8:00 am, Pacific Daylight Time (San Francisco, GMT-07:00)
- Meeting Number: 206 802 913
- Meeting Password: sixtus
- CCM: +14085256800x206802913
Resources
- Webex recording: https://cisco.webex.com/ciscosales/lsr.php?RCID=d0697433e3ae46dfbfe43173a3b74b80
- Wiki: https://bitbucket.org/6tisch/meetings/wiki/140829_webex
- Slides: https://bitbucket.org/6tisch/meetings/src/master/140829_webex/slides_140829_webex.ppt
Taking notes (using Etherpad)
- Xavi Vilajosana
Present
- Michael Richardson
- Ariton Xhafa
- Pascal Thubert
- Joaquin Cabezas
- Xavi Vilajosana
- Raghuram Sudhaakar
- Diego Dujovne
- Kazushi Muraoka
- Maria Rita Palattella
- Pat Kinney
- Pouria Zand
- Rene Struik
- Guillaume Gaillard
- Badis Djamaa
- Giuseppe Piro
- Martin Turon
Agenda
- Administrivia [3min]
- Approval agenda
- Approval minutes last call
- Reboot of this and security meetings
- Overview of Discussions on language in drafts and meetings [10min]
- Review process for draft to be submitted to IESG this year [20min]
- Share on the Flow Label discussion [20min]
- AOB [1min]
Minutes
- [08.05] Meeting starts
- Meeting starts.
- Thomas is not present.
- recording started 8:07
- [08.07] Administrivia [7min]
- Minutes of the last call are approved. No concerns are raised.
- Agenda is approved. No concerns are raised.
- Question: Is ok to have meetings bi-weekly? Yes consensus on that.
- Michael Richardson: Security meetings. Reboot them. Not on Monday this time.
- Need feedback from the documents. Need people to read and to discuss starting from the documents.
- Introduce high level description of the security architecture on the Architecture draft.
- this will require consensus from the Group. ML. *ACTION: send a request to the ML to restart the Security conference calls. Doodle is needed as the previous time does not work for some of the participants. Rene: We had list of open issues: Michael Richardson: this are implementation issues. Rene: not sure about that.
- Michael Richardson: I see all this issues as protocol (design) issues not architecture issues. Are those issues architectural questions?
- Rene: I think they are.
- Michael Richardson: If so, you need to propose some improvements to the architecture to improve the current document.
- Rene Struik: Is the architecture describing the problem statement?
- Michael Richardson: To be discussed in the new calls.
- Pascal: If there are other topics to be discussed, parallel calls can be setup during the week.
- [08.15] Overview of Discussions on language in drafts and meetings
- Need to check english in all drafts. Clean up language and technical aspects before submitting to the IESG.
- WG chairs will help on finding the most technical and grammatical sentences in drafts if this is required by IESG or reviewers.
- It would really be good to have draft reviews by native english speakers.
- Review process for draft to be submitted to IESG this year
- Yang data model 6top, reviewed during IETF meeting by YANG Model people.
- Security draft, will require some more time.
- Michael Richardson: We need to get the architecture text reviewed.
- Pascal Thubert: the current document can be the base of the document to be submitted to review?
- Michael Richardson: will need 10 days or more to get it ready.
- 3 documents to be shipped this year:
- minimal:
- 6top:
- CoAP draft:
- review English piece of the drafts.
ACTION: to call to the ML for draft reviewers.
- Rene Struik: volunters to the Minimal draft. Only language. Not technical.
- 6top interface draft: Pat Kinney volunteers to review
- Coap draft: to call to the ML.
- Share on the Flow Label discussion
- hop by hop header requires 8 bytes
- draft flow label uses the flow label to carry the same information as the hop by hop header. Saves 5 bytes.
- has a dependency on 6MAN.
- Michael Richardson: the draft is not being adopted yet. There was consensus that an exception for the Flow Label needs to be done for the LLN. As having zero there is a waste of space. It is not clear if the proposed flow label draft is the solution or use some sort of compression using 6lowpan for the hop-by-hop header.
- The previous Flow Label RFC, enabled us to use the flow label to place a number which identifies a flow. This was done in ISA100. And we want to do the same for 6tisch.
- The only way to be used now (the flow label) is using a random which makes it unusable to identify flows.
- Michael Richardson: any router observing a flow label with value 0 can select a new value. We want a rule so a packet arriving to an LLN can set it to zero.
- The 6lowpan method to compress the h-b-h header is more flexible that using the flow label and in addition we can use the flow label for other things.
- We want something to setup the interop test. In a 6tisch network the h-b-h does not exists.
- Michael Richardson: 6lo compresses the flow label. otherwise it is represented in 20bits. There is no consensus on weather using the flow label or not. There is appaty. We need technical arguments. *Pascal: whatever decision we take. What do we do in november. What minimal draft should say w.r.t to the use of the hop-by-hop option or flow label?
- [09.06] AOB
- No other business raised
- Diego. Asks about Hawaii meeting IETF.
- Most probably yes but most of people will not attend. Short meeting.
- [09.10] Meeting ends
Updated