Connection details ------------------ • Date: 7-8am US Pacific DST, 4pm CEST: https://www.worldtimebuddy.com/?qm=1&lid=100,12,5392171,1850147&h=100&date=2020-05-19&sln=14-15 JOIN WEBEX MEETING https://ietf.webex.com/ietf/j.php?MTID=m45866408ff12537a6fbaefaa37cf97a0 Meeting number (access code): 619 208 505 Meeting password: tG99jRSJgJ9 JOIN BY PHONE 1-650-479-3208 Call-in toll number (US/Canada) Tap here to call (mobile phones only, hosts not supported): tel:%2B1-650-479-3208,,*01*619208505%23%23*01* JOIN FROM A VIDEO SYSTEM OR APPLICATION Dial sip:619208505@ietf.webex.com You can also dial 173.243.2.68 and enter your meeting number. Join using Microsoft Lync or Microsoft Skype for Business Dial sip:619208505.ietf@lync.webex.com Can't join the meeting? https://collaborationhelp.cisco.com/article/WBX000029055 Participants: * Watson Ladd, Cloudflare * Laurent Toutain, IMT Atlantique * Ana Minaburo, ACKLIO * A Paventhan, ERNET India * Pascal Thubert * Laurent Toutain * Alexander Pelov * Dominique Barthel * Juan Carlos Zuniga, Sigfox * Ivaylo Petrov * Olivier Gimenez * Eric Vyncke, Cisco * Diego Dujovne, UDP * Vincent Audebert Previous for cc ------------------ - Dominique Barthel - Arunprabhu Kandasamy - Ricardo Andreasen - Diego Dujovne - Olivier Gimenez - Carles Gomez - Juan Carlos Zuniga - Ivaylo Petrov - Julien Catalano - Vincent Audebert - Ana MInaburo - Laurent Toutain - Alexander Pelov - Sergio Aguilar Agenda ------ [16:05] Administrivia [ 5min] o Note-Well, Scribes, Agenda Bashing o WG Status [16:10] SCHC-over-LoRaWAN Update [30min] [16:40] SCHC-over-Sigfox Update [10min] [16:50] AOB [ QS ] Minutes ------ [16:05] Administrivia [ 5min] o Note-Well, Scribes, Agenda Bashing o WG Status * Alexander presents the introduction, IPR BCP etc... * Meeting is not recorded, presence is recorded * 2 agenda items; no change proposed * minutes of last meeting are approved. * Ana reports she's working on Security Section for coap draft, to be published then. * WG adoption for the OAM draft? Authors would like to switch to XML v3 and SVG drawings. * let's decide to drop SVG drawings, convert the current text to xml v3 and then edit xml as source code. And use XML tables. * will be asked to report at next meeting. [16:19] SCHC-over-LoRaWAN Update [30min] * Olivier reports on version -08 * WGLC ended May 8th * Olivier browses through the comments * Terminology IETF vs. LoRaWAN Eric: Let's use IETF terminology and let's make sure the mapping is present in all WG documents. Let's be consistent with the terminology in all drafts from this WG. Pascal: Most reviews are going to be from this WG, so it makes sense to use IETF terminology. JCZ: There are some updates on the SF terminology that might need to be done if the table with the mapping is included in some draft. Pascal: object to the use of the term "session" for the fragments of a same datagram Laurent: Don't talk about implicit DTags as it might be confusing - just talk about RuleIDs. Olivier explains why retransmission timer are not used JCZ: I agree. The same issue applies to Sigfox. Laurent: in uplink for example, sending... what is the case? JCZ: It is application dependent as you don't know how it will behave as it depends on the device and the usecase. You can not have a precise timer. Pascal: So you are arguing that the timer should be big Dominique: could you still have a Retransmission Timer to rell that it is time to try to resend, and lower layer will send whenever available? Olivier: If you can not send it right away, what do you do? You drop it or you queue it. Pascal: You have different strategies, but people that create the devices should make sure there are not more applications than could be supported. Laurent: You don't want to do an ack-request? Olivier: For uplink you could have retransmission timer. Olivier describing Class A ACK-Always downlink problem with the timer. If ACK no received at core, can Laurent: You are aiming at supporting classes A, B and C, right? Olivier: We are aiming in something simple. Most people are using class A anyway Laurent: Maybe don't exclude the retramission timer from the draft, but recommend against its usage for class A devices. Pascal: ... Olivier: You are not getting stuck, because you have the inactivity timer. Dominique: It seems that you cancel everything when you lose one fragment?? Laurent: You don't want to fix value for the retransmission timer - it will be based on application and radio conditions. Can we put that it will be specified by implementers. JCZ: It is application dependent. Alex: In some cases you would be able to know how much traffic the device generates on average. In other you don't know. Can we have an estimate done on the GW. Laurent: Both ends need to agree on the value. Olivier: I don't think they need to have the same value - for uplink the GW does not need to have specific retransmission timer. Olivier: For uplink - set by the application. For downlink Pascal: Generally you generate the message, but you add a timebomb or something or you let the lower layer Laurent: Retransmission timer is the time during which we cannot send retransmission. We can send ACK REQ after this period. In a fluid network it will be just after the expiration. In a slotted network it is the next transmission opportuninity. Olivier - do we need to be explicit about All-1 Pascal: yes, please be explicit. [16:40] SCHC-over-Sigfox Update [10min] JCZ presenting the changes for 02 JCZ: Similar issues like the LoRaWAN draft regarding retransmission timers and DL Class-A restrictions. JCZ: All-1 using Seq # from Sigfox. Laurent: You do not interleave non-SCHC traffic, right? JCZ: I guess what you are asking is if the sequence counter is generic for all applications sending data and not necessarily SCHC specific? Dominique: Yes, it seems that you are re-using the counter, but if you don't know if there could have been other traffic, you can now that you did not lose anything, but if you lost something, you can not know if it was SCHC-related or not. JCZ: We will look into that and will provide details. [16:50] AOB [ QS ]