6lo WG Agenda - IETF 124, Montreal
22:00-00:00 (UTC) @ St Denis
17:00-19:00 (Meeting venue time) @ St Denis
Thursday, November 6, 2025
Chairs: Shwetha Bhandari, Carles Gomez
Responsible AD: Éric Vyncke
Minute takers: Laurent Toutain
Jabber scribe: TBD
Introduction and draft status Bhandari/Gomez 10 min
Agenda bashing; blue sheets; scribe; Jabber scribe
No comment on the Agenda
CG: gives the status of documents
- 2 in RFC editor queue
- PASA is waiting for GAAO draft
- SCHC of 802.14.5 has been updated
- 6lo for Optical communication has been updated
No comment on the current status.
Luigi Iannone 10 min
Path-Aware Semantic Addressing for LLNs
Slides
2 updates,
1 change the option format
wait until GAAO is ready
No comments or questions
Generic Address Assignment Option for 6LoWPAN ND
Slides
2 revisions: based on Pascal's comment and review from Carles
document is stable, with section renaming.
went back from "ratification" to "registration".
INTDIR and GENART are OK.
Shepherd is OK
CG: there is some change in PASA, shepherd writeup will need a slight update. GAAO shepherd writeup needed. Then, ready to submit both docs to the IESG.
No objections.
Carles Gomez 25 min
Transmission of SCHC-compressed Packets over IEEE 802.15.4
Slides
The goal is to use SCHC in 6lo.
Version 11 addresses the comments in MAdrid
- different Strata, one for Transition Protocol Stack.
One sub-section in section 5, to use 2 stratums:
- one for UDP, one for COAP and may be one for OSCORE .
CG presents 3 scenario involving a 6LN and an external host and the different stratums using or not DTLS and OSCORE.
In a transition Protocol Stack, 6Lo is used at layer 3 and UDP and CoAP are compressed by SCHC.
EV: why in the second scenario UDP and CoAP ?
LT: may be the scope of the compression stratum
CG: clarify this, at least it is for illustration
Christian on the chat:
Now I understand the strata separation: makes sense if the 6LBR is not a CoAP proxy.
I understood this to allow independent negotiation of the settings for UDP (between node and 6LBR) and CoAP (possibly varying per peer, and the 6LBR doesn't need to know how to decompress for each peer on the unconstrained side)
Statrum header is used to carry the protocol number, may be fully compressed. Format is provided in the draft.
Appendix B summarizes with mode is adapted to a environment. Bob proposes to add some examples.
Coexistance is possible, and may be address in another document if necessary.
Single-end point networks has to be clarified
AP: agree that the document should not be too much complex and a new document for coexistance is better.
AP: for terminilogy, there is a detailled SCHC architecture for Marion and Quentin, look at minimal architecture, to look for the common vocabulary.
Review from Esko:
- hybrid approach (mesh-under, Route over), for example with Thread, add some text to explain
- 6LR and 6LBR may generate traffic, add some example
- can a router can change ECN bits ? not possible in all mode. add to the next revision
- Check the consistency between 6LN and leaf
- in PRO address is pointed in the residue, is the pointer size is enough
Align with the SCHC architecture and wait that the architecture get stable.
AP: there is a consensus for the change, participate to the SCHC architecture evolution
CG: time to get review, and the WGLC is near.
ED: Christian's comment on the chat
ED: SCHC domain could be interesting to investigate
CG: no specification right now for DTLS, that's why it is not included in the document
AP: DTLS has be compressed, but there is no official RFC on that.
AP: Matter does not use CoAP, is there something that can benefit of SCHC
CG: Maybe something to investigate
Younghwan Choi 10 min
Transmission of IPv6 Packets over Short-Range OWC
Slides
version 5: renumbering sections and text clarifications
No technical chages introduced
Security considerations missing on this document but is covered by the next presentation.
CG: there was a review some time ago, but this version of draft should be reviewed.
Munhwan Choi 10 min
Security considerations for IPv6 over OWC
Slides
ED: if IEEE describes security at link level, include a ref
ED: is DTLS be usefull, or OSCORE for CoAP
CG: announce new revision on the mailing list.
Kerry Lynn 10 min
Update of RFC 8163
Slides
changes on RFC 8163 (MS/TP or 6LoBAC)
align the termonology with the new version of MS/TP
What is the best approach
EV: should obsolete it
CG: follow the usual process, with call for adoption
DR: support this change, replace the whole document
EV: move the previous authors to contributors (not in acknowledgment section)
AP: support, SCHC with 6lo, whould it be good to investigate the header compression ?
this will be the decision of the WG
TW: where is the spec
CG: some explicit support of the work, draft -01 is consistent
CG: sign of support from the room, go to the mailing list for call for adoption.
CG: No other business, the session ends here.
Ends: 85 min
Meeting chat
Christian Amsüss
03:54
Now I understand the strata separation: makes sense if the 6LBR is not a CoAP proxy.
Christian Amsüss
03:59
I understood this to allow independent negotiation of the settings for UDP (between node and 6LBR) and CoAP (possibly varying per peer, wnd the 6LBR doesn't need to know how to decompress for each peer on the unconstrained side)
Esko Dijk
04:06
@Christian that sounds like a possible use case. Did you mean the 6LBR in this case may need to insert the uncompressed UDP header before sending the packets out on the non-constrained network. In case the decompression on the end server only has rules for CoAP decompression? E.g. because this server OS needs uncompressed UDP to find the port and then send it to the right application process on that port.
Christian Amsüss
04:07
Exactly.
Marion Dumay
04:16
Here is the link to the draft mentioned by Alex: https://datatracker.ietf.org/doc/draft-lampin-schc-minimal-architecture/. In particular, we have introduced the concept of Domain, which could greatly simplify the terminology. Single-endpoint networks and multiple-endpoint networks could become Single-Domain networks and Multi-Domain networks respectively. We have also proposed terminology that we believe to be more consistent.
Alexander Pelov
04:34
I looked up the Matter spec. Quite extensive use of mDNS for service discovery and commissioning + use of group control for lighting scenes or coordinated control. -> maybe something to look at
Esko Dijk
04:39
@alexander One thing that comes to mind is the SRP protocol (RFC 9665). It's used in Matter; and was mentioned in the dnssd WG session today. It produces pretty big packet sizes in the way it's currently used in Matter.
Compression could help there. (It was actually discussed in a previous dnssd session - I could look it up.)
Alexander Pelov
04:39
yes, that, sounds great!
Éric Vyncke
04:40
https://author-tools.ietf.org/iddiff?url1=rfc8163&url2=draft-lynn-6lo-rfc8163-bis-01&difftype=--html
Christian Amsüss
04:41
I never used but liked reading 6lobac, would be nice to have this updated both w/rt BACnet and to 6lo enhancements.
And good to have you back :-)