Interim #4 2023

Data / Time

• Date: February 14th, 2023
• Time: 7-8am US Pacific PST, 4pm CET

Meeting Information

• Agenda: on datatracker
• Meeting material: on datatracker
• Meeting link: Meetecho
• Live Minutes: CODIMD Link
• ical: ICS

Interim Agenda

[16:00] Administrivia [10min]

• Note-Well, Scribes, Agenda Bashing
• WG Status

[16:10] Rechartering [20 min]

• Presenter: Pascal Thubert
• Topics: Review proposed work item

[16:30] SCHC Data Model (auth48 changes) [10 min]

• Presenter: Laurent Toutain
• Associated draft: draft-ietf-lpwan-schc-yang-data-model

[16:40] Architecture Next Steps [10 min]

• Presenter: Chairs, Ana Minaburo
• Associated draft: draft-ietf-lpwan-architecture
• Topics: Ivan’s comments

[16:xx] AOB [ QS ]

Interim #4 2023

Data / Time

• Date: February 14th, 2023
• Time: 7-8am US Pacific PST, 4pm CET

Interim Minutes (times in CET)

[16:07] Administrivia [10min]

• Note-Well, Scribes, Agenda Bashing
• WG Status

Quentin would like to discuss the SCHC specs, some stuff to polish to
make it easier to use SCHC in over foo scenarios (eye opener, will
continue on the ML). No rush, today if time left.
Alex: very good please request a slot for the next meeting to have
dedicated time.

Alex: Is there something more we need to do on the SCHC over NB-IoT
document?
Pascal: No

Alex: I finished the shepherd review of SCHC Compound Ack, I don't know
what is the next step
Eric: in datatracker, a hyperlink to change "WG Status", then select
"Request Publication".
Alex: OK, I'll look at that and push it.

Laurent: I just had a meeting with Marco Tiloca, he has reviewed 8824
and will file an Errata of the RFC.

Eric: Compound ACK has not gone through Working Group last call yet, it
seems per datatracker.
Pascal: pretty sure WG LC was done. Launched October 2021, completed Jan
2022.
Eric: if the mailing list has a track record for the WGLC, then it is
all good. Kind reminder to the chairs to launch the WGLC by the
datatracker
Eric: Also about SCHC over Sigfox, there is still one minor comment to
be addressed.
Eric (after rereading the email thread after the interim): -21 appears
to address the remaining blocking issues raised by Roman Danyliw in his
DISCUSS ballot. Eric to nudge Roman to review his ballot.

[16:10] Rechartering [20 min]

• Presenter: Pascal Thubert
• Topics: Review proposed work item

Pascal : Third discussion on rechartering text at LPWAN meetings
(already at IETF 115 and previous interim).
Proposed text is shown, on 3 slides.
Laurent: does the SCHC Ethertype need to go on the charter? Solved soon
and not LPWAN's business.
Pascal: we discussed that with Bob, the goal is to have an RFC
discussing the need of this, but it is not to be done in our group, more
an Int Area document.
Laurent: Make the second sentence more genreal so that it is clear that
we work with different areas.
Quentin: With the proposed charter, is it still possible to amend SCHC?

Pascal: maintenance is alway possible, see next slide.
Laurent: remote end point is not clear, maybe too technical?, change
that for technologies?
Pascal: remote refers to endpoints that are over multiple hops, as
opposed to on link.

Laurent: For example if we want to extend that fo HTTP, it will not be
in the charter.

Pascal: we want to extent hop to multihop.

Laurent: yes I agree, but we are still missing
Pascal: need to further update the charter to reflect this? we know we
need dataplane signalling to address peer-to-peer.

Carles Gomez (in chat): perhaps s/larger L2 and L3/beyond LPWAN
Pascal: to edit further talk with Alex to include it in the dropbox.

Pascal: slide 2 of charter should allow what Quentin has in mind.
Quentin: need to extend the charter to update SCHC in order to support
multihop OAM
Pascal: multihop is on slide 3 of text.
Quentin: allow tuneling is also on the charter?
Pascal: yes, included in the Foos.
Alex: this looks clear and complete. We'll send it to the ML, and
collect feedback for next interim.
Eric: Once the dicussions are finished inside the group, let the chairs
me know, I will push new text for IESG+IAB review for two weeks, then 2
more weeks for the IETF community feedback.

[16:35] SCHC Data Model (auth48 changes) [10 min]

• Presenter: Laurent Toutain
• Associated draft: draft-ietf-lpwan-schc-yang-data-model

Pascal: Laurent, what changes still to be done?
Laurent: Sent on the ML last week. (i) It is about the way you arrange
it to validation, (2) coap options.
Laurent: We've got the RFC number with some questions, I'll wait for Ana
to address it but it should not be difficult. I already ask in the ML
but as you'll see in the github there are minor.
Eric: You have changes the YANG module syntax, not its semantic, so all
is good. Also, be sure to change the YANG module revision date.

Laurent: It's time to request sids for this model, but I believe that we
need to reorganize the order of sids. How to proceed? Write a draft to
explain what we need, or just write a sid file?
Alex: sid registry does not exist at IANA (yet), pick some seeds
provisionnally, and we'll ask IANA later to allocate them.
Eric: IANA is not planning to handle/allocate sids.
Alex: two registry levels. Upper level, not managed by IANA. For some
RFCs, subregistry is requested to IANA, who will allocate.
Alex: in any case, no place to request allocation as of today.

[16:43] Architecture Next Steps [10 min]

• Presenter: Chairs, Ana Minaburo
• Associated draft: draft-ietf-lpwan-architecture

• Topics: Ivan’s comments
  Ivan commented on seven sections.
  Ivan: with Laurent, read again carefully. Suggested to rearrange
  sections.
  Main point is to suggest one-hop is specialization of multi-hop.
  Pascal: indeed, we are going to cover multihop topologies. But
  wheter the two end points are onnected via single link or multihop
  does not matter.
  Pascal: one-hop is a collapsed version of multihop.
  Laurent: think better to start with what we know already, which is
  star topology.
  Pascal: real architecture starts in section 5. Current section is
  about LPWAN technologies. What you are proposing as section 2 is
  essentially current section 3.
  Maybe section 2 could go into intro, agreed.
  Pascal: I don't mind to keep 3 separate sections (2, 3 and 4) or
  collapse into one.
  Alex: maybe start with most general architecture, and reduce to star
  topology in section 2, and detail non-star topologies in section 3?

  Pascal: rather have a single architecture. Star is just a collapsed
  version.
  Alex: agree.
  Pascal: rule managed is currently covered in section 7 (deployment).
  Just not structured the same way.
  Ivan: provisioning but also onboarding.
  Pascal: current section 7 is lifecycle.
  Alex: what are the next steps?
  Pascal: decision on how to present the archicture (two descriptions
  vs. more one complete architecture and collapsed version).
  Alex: second the latter form
  Alex: running out of time. Laurent, Ivan, get this on the ML.
  Pascal: not convinced that we need to reorganise structure of doc.
  Alex: we'll discuss on the ML and at the next interim

[17:03] meeting adjourns