Minutes IETF111: manet
minutes-111-manet-00

# MANET WG Agenda IETF-111
========================

Date: Thursday, July 29, 2021
Time: 1630-1730 PDT
Location: Meetecho Room 6
Chairs: Don Fedyk , Ronald in 't Velt
Responsible AD: Alvaro Retana

Session material: https://datatracker.ietf.org/meeting/111/session/manet
Note taking: https://codimd.ietf.org/notes-ietf-111-manet?both

## 1. Opening - 10 min (Chairs)
- Note Well etc.
- Welcoming new co-chair
- Agenda bashing
Slides:
https://datatracker.ietf.org/meeting/111/materials/slides-111-manet-chairs-slides-00.pdf

Don Fedyk is new co-chair of the group. Don is a longtime IETF participant. His
other current interests in the IETF are Detnet and IPsecme. He is also active
in IEEE, in 802.1 and security.

## 2. Document status - 10 min (Chairs, Lou Berger)

Lou presents latest status on credit-based flow control i-Ds; three out of
WGLC, one adopted as WG document (see agenda item 3) Lou Berger: How would you
like the updates handled on the ether-based credit extentions? Ronald in 't
Velt: Make the updates to the docuement then submit as WG document.

Hennings questions on the list answered on the list. (Why introduce Traffic
Class Identifier (TID) if we already have Link Identifier (LID)?) Lou Berger:
There was no LID when this work started. Also disagree on technical grounds. We
don't want to always mandate support of LID when doing credit-based flow
control. Lou Berger: Do not respin the drafts. Rick Taylor: I agree with Lou.
Do not want to put the extensions together and create dependencies. LID would
get two different meanings, causing confusion. I can see Henning's point from
an implementation perspective, but conceptually it would be a mistake

Lou: There is an implmentation of DLEP on Github - Opportunity to take over
this. There are old pull requests. Rick Taylor: Several of those Pull Requests
were from my team. There is a Fork of this but there is opportunity for others
to take over. Used for testbeds and interops in a large European consortium.
Ronald and Lou: Put the link on the list. Rick Taylor: Our merged fork is:
https://github.com/airbus-mipn/LL-DLEP1 Lou Berger: Elaborated on the
difference between the Flow Identifier and Traffic Identifier (on Ronald's
request).

## 3. WG adoption of draft-berger-manet-dlep-ether-credit-extension - 5 min
https://datatracker.ietf.org/doc/draft-berger-manet-dlep-ether-credit-extension/
Adoption call completed. Comment by Henning Rogge calls wisdom of splitting the
credit-based flow control documents into question. This decision was made
several years ago. Going back on this now would also affect the I-Ds that have
just completed WGLC. Ronald in 't Velt: Will announce adoption on the ML.
Intention is to start WGLC fairly soon, so that this I-D will not trail too far
behind the others. However, there is a window for further comments.

## 4. WG adoption of PHY-related DLEP extension I-Ds - 10 min
https://datatracker.ietf.org/doc/draft-rogge-manet-dlep-radio-band/
https://datatracker.ietf.org/doc/draft-rogge-manet-dlep-channel-utilization/
https://datatracker.ietf.org/doc/draft-rogge-manet-dlep-radio-quality/

PHY Related DLEP extensions
(The author, Henning Rogge, could not attend the meeting).

Ronald in 't Velt: Channel Utilization I-D; why does the modem not calculate
the channel utilization itself, instead of conveying low level parameters to
the router? Rick Taylor: I agree the information could be in a different
format. Ronald in 't Velt: Please say so (again) on the list. Lou Berger:
Applies to all of these. None of the drafts have motivation. What is the Router
supposed to do with these.

## 5. Future work - 25 min (Chairs, discussion)
- OLSRv2 & Routing Metrics maintenance and extensions
- DLEP clarifications & Lessons Learned I-D
(This topic was inadvertently skipped!)
- Multicast in MANETs
Rick Taylor: Mentioned BIER since BIER has arrived after multicast in MANETs
was last discussed shortly after last re-charter (2016). Rick Taylor: There are
Radios in the market that do multicast at the link layer. Lou Berger: There is
work in RAW that may end up countering this point, i.e., want to integrate IP
and radio support.  In general there is some potential work in RAW that may be
brought into MANET. Rick Taylor: Yes but it is different and this is not clear.
RAW is doing the use cases and the Architecture. This could be brought into
MANET. Alvero: Comented that the Managment may not be something that belongs in
MANET. Ronny Bull: It is a challenging problem to do this in MANET. Rick
Taylor: Agree the traditional Managmetn does not apply, DTN has working
mananging some of these aspect. Alvero Retana: Asked before if the mangment
work could move  work to MANET? Rick Taylor: Can handle this in DTN. Now in
scope. Ronald in 't Velt': Any interesting in YANG Models - No one spoke up
over what was previoulsy pointed out about mananagment. Ronald: Mulicast Would
like to have a small group to explore what is possible. - Work not covered by
current charter Inter-MANET routing (as presented at IETF-105) -