Agenda/Materials for the NETMOD 114 WG Session

https://datatracker.ietf.org/meeting/114/materials/agenda-114-netmod
https://datatracker.ietf.org/meeting/114/session/netmod

Session:

Wednesday, July 27, 2022
15:00-17:00 Wednesday Session III
Room: Independence C

WG Chairs:

Lou Berger (lberger at labs dot net)
Kent Watsen (kent plus ietf at watsen dot net)
Joel Jaeggli (joelja at bogus dot com)

Available During Session:

ICS: https://datatracker.ietf.org/meeting/114/sessions/netmod.ics
MeetEcho: https://meetings.conf.meetecho.com/ietf114/?group=netmod
Onsite tool:
https://meetings.conf.meetecho.com/onsite114/?group=netmod&short=&item=1

Audio Only: https://mp3.conf.meetecho.com/ietf114/netmod/1.m3u
Jabber: xmpp:netmod@jabber.ietf.org?join

Available During and After Session:

Notes: https://notes.ietf.org/notes-ietf-114-netmod#both
Slides: https://datatracker.ietf.org/meeting/114/session/netmod
Drafts (TGZ):
https://datatracker.ietf.org/meeting/114/agenda/netmod-drafts.tgz
Drafts (PDF):
https://datatracker.ietf.org/meeting/114/agenda/netmod-drafts.pdf
Datatracker: https://datatracker.ietf.org/group/netmod/about/

Available After Session:

Recording: http://www.meetecho.com/ietf114/recordings#NETMOD
recording and transcript at https://www.youtube.com/watch?v=qIIQwK3qNgc

Jabber Logs:https://www.ietf.org/jabber/logs/netmod

1) Session Intro & WG Status

Chairs (10 minutes)

Commenced 1500 EDT

Lou Berger - much of the time this time will be spent dicussiong yang
versioning

status updates

waiting for an update that will require another last call.

we had 25 minutes left in the schedule we had two topics.

Topic 1 yang next - discussion to be led by the chairs

Topic 2 and a proposal for how to move faster that was present to the
IESSG by our AD

Rob Wilton - aim is to get updated versions out quite quickly for yang
versioning.

Lou Berger: - 6991 bis dock is ready for last call will go out after
meeting.

syslog needs an update.

Liason from O-RAN - Balazas will help with the response

From: O-RAN-WG10
Received:
2022-07-20
Title: LS on O-RAN – IETF YANG Module Versioning
“O-RAN needs a robust module versioning system for YANG modules.“
“ORAN would like to use a ratified IETF-provided solution for YANG
module versioning”
Action requested: “O-RAN would like to ask when the two IETF drafts
draft-ietf-netmod-yang-module-versioning and
draft-ietf-netmod-yang-semver will be approved”
Desired response date: 2022-09-30

Chartered items:

2) A YANG Data Model for Syslog Configuration (5 minutes)

Presenter: Joe Clarke

Draft: https://datatracker.ietf.org/doc/html/draft-ietf-netmod-syslog-model-27

15:09 commences
Joe Clark presenting

Kent Watson - asymetric keypair you want the singular not plural

Joe Clark: I picked this based on my reading of the crypto draft.

if nothing else we're ready for last call.

3) Node Tags in YANG Modules (10 minutes)

Presenter: Qin Wu

Draft: https://datatracker.ietf.org/doc/html/draft-ietf-netmod-node-tags-08

15:13 commences
Qin Wu presenting

Rob Wilton: question about getting data with xpath filter, why do you
choose to do this versus a filter path operation?

Qin Wu: we try to use a standard selection tool due to complexity?

Rob Wilton: xpath filtering is quite expensive. wheres if you look at
nmba options they provide tools to limit the amount of data returned
based on filtering tags

Qin Wu: we don't use filtering

Rob Wilton: let me syync up offline

Balazas Lengyl: is there a prioty between these tags and masking

Qin Wu: regarding masking, We follow module tag concept in RFC8819

Balazas Lengyel: this will make some of our use cases much harder
because it will make schema tags unreliable

Lou Berger: clearly some comments in the room, not clear me that
juergens comments were addressed need to clear those.

4) YANG Versioning Update and Discussion (45 min)

Presenters: Tom Hill, Balazs Lengyel, Joe Clarke

Draft: https://datatracker.ietf.org/doc/html/draft-ietf-netmod-yang-module-versioning-06

Draft: https://datatracker.ietf.org/doc/html/draft-ietf-netmod-yang-semver-07

15:32 tom hill presenting (overview)

Balazas presenting 15:36 draft-ietf-netmod-yang-module-versioning-06

Rob Wilton: we have an interpreation, which suggests a version to be
used but doesn't mandated or require it.

if you do choose a version that is compatible that's great but if cannot
you get a warning

this behavior would be consisent with not understanding the extension

Lou Berger: believe this was the consesnus of the last meeting
proceeding on versioning work in 1.1 vs next

Balazas Lengyel: I think we settled on most issues. We will publish an
update, including the text on node tags, and then ready for a second
last call.

Lou Berger: ok so plan is to move forward with presented recommendations
unless there are issues raised on the list, and issue an upatch which
can go through a second LC.

15:59 draft-ietf-netmod-yang-semver-07 - joe clarke presenting

Lou - are us assuming you never use the date in semver?

Joe - in the revision or derived

Balazas - you don't change the revision state so it must include the
date

Lou - if you update the version without making any changes does the date
change?

Joe - the date shouldn't change either

Anthony Somerset - is there any problem with gratuitous version bumping?

Joe - avoiding unecessary churn is the goal

Charles eckel - some of the consumers will really just care about the
yang modlues, it's better if they don't see a change.

Joe - I take the comment that we can be better about describing

Rob - limited stub branches, back porting are reasons why this might
exists.

Lou - just brought up the outstanding issues, do you think this will be
done and ready for last call?

Joe - I would like to do a new version, and have an interim to have the
passionate people close this out live?

Rob - The people rasing issues during wg last call are not here, but we
have made tweaks and we think we are on the right track, we will need
any working group last call.

Lou - look forward to seeing the update and then the last call

Non-Chartered items:

5) System-defined Configuration (10 minutes)

Presenter: Qiufang Ma

Draft: https://datatracker.ietf.org/doc/html/draft-ma-netmod-with-system-03

16:23 Quifang Ma presenting

Balazas - object to the last bullet on recap slide, there are people the
use this behavior (modifying running with automation)

Quifang Ma - when discussed on the list, this is recomendation that
people prefered.

Balazs - onboard clients chanage configuration. The difference between
configuration change by an onboard client and by the system is just
phylosophical, it is the same from an external clients point of view

Rob - automation running on the device gets treated as an other client

Kent - I also think this is right

Jason Sterne - worried about be so stricvt about the running (when
considering how we might eventually deal with configuration templates)

Lou - question to the group for the poll of whether the working group
would like to continue discussing the draft.
There is continued interest, so we look forward to the next revision and
continued discussion on the list. (poll showed about half the present
particpants expressed interest)

6) YANG Extension and Metadata Annotation for Immutable Flag (15 minutes)

Presenter: Qiufang Ma

Draft: https://datatracker.ietf.org/doc/html/draft-ma-netmod-immutable-flag-02

16:37 Qiufang Ma presenting

Anthony Somerset - immutable means cannot be deleted or changed and the
exceptions means that's not the case

Balazas - immutable with exceptions means you can create or modify but
not delete. Maybe the word is not the best and opened to suggestion

Anthony - "locked?"

Rob - conflict are we going to make server implmentations choose not to
do this.

the last point i want to make is um a game in like in interfaces and
things be very clear about the fact that you can delete the interface
and its type but you just can't delete the type

Balazs - This could encourage bad behavior, but it's better to describe
it since it is what people are actually doing
there are running code examples and non-IETF standards that prescribe
this behavior. This is about documenting existing behavior.

Joe - must reject is a clarification is that the server would have
rejected this anyway.

poll are you interested in seeing this progress in the working group
(modestly less favorable then more)

Unscheduled time (25 min)

YANG Next discussion?

16:48 Kent / Rob

managing the evolution of ietf modules (yang next)

Lou - rob are you thinking of carrying out a process experiment

Rob - yes exactly

Lou - how quickly do you think you can do this

Lou - in another working group I have a process where the chairs want to
move faster then author

Benoit Claise - support the idea - the yang catalog was done this way.
the process is actually easy

Lou - so I hear co-author

Kent - one party rasied their hand to say they would not they would not
like to see this work proceed i'm just wondering if that individual
would be willing to speak as to why they did not raise their hand?

Kou this concludes the two lightning talks, thank you all.

done at 16:01

---