Re: [lp-wan] augment coupond-ack

Thanks for the clarification,  Laurent !

(chair hat off)

+1 on Eric's position - option 1)

Cheers,
Alex

Le lun. 15 nov. 2021 à 09:32, Eric Vyncke (evyncke) <evyncke@cisco.com> a
écrit :

> And without any hat except some familiarity with YANG, I also support the
> approach of having the data model pristine and let the YANG augmentation in
> the compound-ack I-D
>
>
>
> -éric
>
>
>
> *From: *lp-wan <lp-wan-bounces@ietf.org> on behalf of Laurent Toutain <
> laurent.toutain@imt-atlantique.fr>
> *Date: *Monday, 15 November 2021 at 09:30
> *To: *Alexander Pelov <a@ackl.io>
> *Cc: *lp-wan <lp-wan@ietf.org>
> *Subject: *Re: [lp-wan] augment coupond-ack
>
>
>
> yes
>
>
>
> On Mon, Nov 15, 2021 at 9:08 AM Alexander Pelov <a@ackl.io> wrote:
>
> Dear Laurent,
>
>
>
> Thanks for getting the ball rolling on this one so quickly after the IETF!
>
>
>
> On 1) - when you say "ignore the compound-ack from RFC8724" - there is no
> compound-ack in RFC8724.
>
> Do you mean this to be - keep the Yang Data Model aligned with RFC8724 and
> not include compound-ack in it?
>
>
>
> (and thus -> define an extension in the SCHC-Compound-Ack draft ?)
>
>
>
> Cheers,
>
> Alex
>
>
>
>
>
>
>
> On Mon, Nov 15, 2021 at 8:51 AM Laurent Toutain <
> laurent.toutain@imt-atlantique.fr> wrote:
>
> Dear lpwaners,
>
>
>
> During the last IETF meeting, it was asked to detach the compound-ack from
> the RFC8724 model. I see to ways of doing it.
>
>
>
> 1) ignore completely the compound-ack from the RFC8724 data model and
> augment this model such as:
>
>
>
> augment
> "/schc:schc/schc:rule/schc:nature/schc:fragmentation/schc:mode/schc:ack-on-error"
> {
>
> leaf bitmap-format {
>
> type compound-ack:bitmap-format-type;
>
> when "derived-from(../schc:fragmentation-mode,
> 'schc:fragmentation-mode-ack-on-error')";
>
> default "compound-ack:bitmap-RFC8724";
>
> description
>
> "How the bitmaps are included in the Ack message.";
>
> }
>
> }
>
>
>
>
>
> Where we create the leaf and a specific identityref for compound-ack.
>
>
>
> 2) create the leaf in RFC8724 datamodel but just a default value and just
> define the compound ack value in the compound ack model.
>
>
>
> Personally, I prefer the first approach, since we cannot guess all the
> time what will be extended, but I would like to have your point of view on
> the subject.
>
>
>
> Laurent
>
> _______________________________________________
> lp-wan mailing list
> lp-wan@ietf.org
> https://www.ietf.org/mailman/listinfo/lp-wan
>
>