Re: [netmod] Draft Minutes for Virtual Interim

On Tue, Jan 30, 2024 at 8:55 AM Jason Sterne (Nokia) <jason.sterne=
40nokia.com@dmarc.ietf.org> wrote:

> Hi WG,
>
> (and in particular to those who attended the interim).
>
>
>
> The summary below mostly matches my memory of the discussions, but I don’t
> really remember us concluding on this:
>
>
>
>      The WG agreed to let 7950-bis "update" 8342 (NMDA) with the
>
>      clarification the <running> alone does not have to be valid.
>
>      E.g., clients may have to perform transforms to calculate
>
>      <intended>, which is subject to validation.
>
>
>

I do not have a problem with "<running> SHOULD always be valid" (instead of
MUST).
IMO the lack of standards for the "magic" that converts <running> into
<intended> should be fixed.
There was going to be metadata like an "enabled" flag on <running> data
nodes. This never happened.
Too bad, because this solution could support both NMDA and non-NMDA
deployments.

Andy

> (the rest of the minutes/summary below also seems to contradict that
> paragraph being a conclusion no?)
>
>
>
> I thought it was going to remain somewhat optional/indeterminate if
> running will be valid:
>
>    - Servers may or may not enforce running to be valid (i.e. they may
>    only validate intended as a proxy for validating running)
>    - Clients can’t necessarily expect to be able to offline validate
>    running, although it may work in circumstances where the operator doesn’t
>    use templates or inactive config **or** the client reproduces the
>    server logic for the running->intended transforms
>
>
>
> Jason
>
>
>
> *From:* netmod <netmod-bounces@ietf.org> *On Behalf Of *Kent Watsen
> *Sent:* Monday, January 29, 2024 7:21 PM
> *To:* netmod@ietf.org
> *Subject:* [netmod] Draft Minutes for Virtual Interim
>
>
>
>
>
> *CAUTION:* This is an external email. Please be very careful when
> clicking links or opening attachments. See the URL nok.it/ext for
> additional information.
>
>
>
> Link to minutes:
>
>
> https://datatracker.ietf.org/doc/minutes-interim-2024-netmod-01-202401231400/
>
>
>
> Reproduced below for convenience.
>
>
>
> Please report any updates needed here.
>
>
>
> Kent (and Lou)
>
>
>
>
>
>
>
> This virtual interim was soley focused on the "system-config" draft.
>
> Qiufang Ma presented.
>
>
>
> Draft: https://datatracker.ietf.org/doc/draft-ietf-netmod-system-config
>
>
>
> In the course of two hours, there was a lot of discussion.  So much so
>
> that trying to capture all the points verbatim would take too long. A
>
> link to the video is here: https://www.youtube.com/watch?v=sAF0fppqBGA.
>
>
>
> A high-level summary is:
>
>
>
>   Qiufang's presentation focused on two main questions?
>
>
>
>   1) The "origin" issue.
>
>
>
>      The WG agreed that <system> nodes copied into <running> should
>
>      have origin "intended".  The system-config draft will "update"
>
>      RFC 8342 (NMDA) to state this.
>
>
>
>      The WG agreed that data-migration is 1) not <system>-specific
>
>      concern and 2) is out-of-scope for this draft.
>
>
>
>   2) Validity of <running> alone.
>
>
>
>      The WG agreed to let 7950-bis "update" 8342 (NMDA) with the
>
>      clarification the <running> alone does not have to be valid.
>
>      E.g., clients may have to perform transforms to calculate
>
>      <intended>, which is subject to validation.
>
>
>
>      The WG agreed on a new Option 4: this document doesn't say
>
>      anything at all about the validity of <running>.  That is,
>
>      fully rely on existing 7950 and 8342 statements.
>
>
>
>      This leaves it up to interpretation.
>
>
>
>      Templates and inactive configuration are nice for humans, but
>
>      unnecessary for machine-to-machine interfaces.  That is, the
>
>      issues arounds such mechanisms are largely moot in environments
>
>      using a controller.
> _______________________________________________
> netmod mailing list
> netmod@ietf.org
> https://www.ietf.org/mailman/listinfo/netmod
>