Re: [Teas] WG I-D Status actn-vm

From: Dhruv Dhody <dhruv.ietf@gmail.com>
Sent: 15 July 2020 11:31

Hi Tom,

Thanks for keeping us on our toes :)
I will consider these in the next update.

Regarding uri, I was influenced by RFC 8345 and thought it made sense
here as well (with my reading being the datastore to be related to
NMDA). I will get that confirmed.

<tp>
I see what you mean about RFC8345.  I do not understand what it says either and disagree strongly with its recommendation to embody semantics, a regular source of grief.

uri does cover a lot, dozens of schemes, complex syntax multiple elements, plenty of scope for poor choices.  YANG does not even restrict the length which SMI does.  I have not seen it used much, rather string gets used without restriction.

Tom Petch
Thanks!
Dhruv

On Wed, Jul 15, 2020 at 3:32 PM tom petch <ietfa@btconnect.com> wrote:
>
> From: Dhruv Dhody <dhruv.ietf@gmail.com>
> Sent: 13 July 2020 19:53
> Hi Tom,
>
> Apologies for the late reply. I have made an update and handled all your pending comments! Please have a look - https://datatracker.ietf.org/doc/draft-ietf-teas-actn-vn-yang/
>
> <tp>
> Getting there.  I had not realised how much this leant on te-topo but now I am more familiar with the latter I find myself in less need of references but suspect that others will find them useful.  I note that there is a -24 for yang-te - I do not know what has changed.
>
> A general comment.  A much used and useful convention is that a list is plural and content singular so rather than
> list vn-ap
> uses vn-ap
> you could have
> list vn-aps
> uses vn-ap
> or
> container aps
> list ap
> container vns
> list vn
> I realise that other teas I-D use the suffix -list but for me that is clumsier than a simple 's'
>
> s.1.3 has the wrong prefix for te-topo
>
> I find the use of uri for ids interesting; any scheme? any length?  probably worth some advice here given people's predilection for disastrous identifiers:-)
> And do you mean datastores, or modules?  I struggle to see how the uri could differ  across datastores.
>
> LTP should be expanded in the YANG even if I do now recognise it.  It has a different meaning in other technologies.
>
> IANA /TDB/TBD/ !
>
> Tom Petch
> Thanks!
> Dhruv
>
> On Fri, Apr 24, 2020 at 4:03 PM tom petch <ietfa@btconnect.com<mailto:ietfa@btconnect.com>> wrote:
> From: Teas <teas-bounces@ietf.org<mailto:teas-bounces@ietf.org>> on behalf of Dhruv Dhody <dhruv.ietf@gmail.com<mailto:dhruv.ietf@gmail.com>>
> Sent: 22 April 2020 07:37
> To: TEAS WG Chairs; TEAS WG (teas@ietf.org<mailto:teas@ietf.org>)
> Subject: [Teas] WG I-D Status
>
> Hi Chairs, WG,
>
> Please find the status -
>
> ==
>
> A Yang Data Model for VN Operation (draft-ietf-teas-actn-vn-yang-08)
> Updates:
>   Mainly YANG guidelines update
>   All earlier issues resolved
>
> <tp>
> Looking at -08 I struggle to see a resolution of some of the issues I raised 3Jul2019.
>
> YANG deserves a reference in Introduction
>
>
> Done!
>
>
> YANG should be upper case, not yang
>
>
> Done!
>
>
> 1,3 places te-types in teas-yang-te
> The module imports it from teas-yang-te-types
>
>
> Updated to the RFC 8776
>
>
> identity  .. computatione-failed
>
>
> Fixed
>
>
> disjoint I cannot find in RFC8453 - a reference would be helpful
>
>
> I use the te-types typedef instead of defining a new one!
>
>
> ltp likewise
>
>
> Added
>
>
> vn-id seems underspecified.  Has it semantics or is it just a number?  Are there restrictions on the range?  Many I-D create a raft of YANG types
> which seem to complicate the model - here I wonder if a vn-id  type would help.
>
>
> Added typedef.
>
> Tom Petch
>
>
> Next Steps:
>   Yang Doctors Review
>
> ==
>
> YANG models for VN/TE Performance Monitoring Telemetry and Scaling
> Intent Autonomics (draft-ietf-teas-actn-pm-telemetry-autonomics-02)
> Updates:
>   Mainly YANG guidelines update
> Next Step:
>   Solicit more reviews
>
> ==
>
> Traffic Engineering (TE) and Service Mapping Yang Model
> (draft-ietf-teas-te-service-mapping-yang-03)
> Updates:
>   Mapping to Network Yang Models added (L3NM and L2NM)
>   YANG guidelines update
> Question to the WG:
>   Is this the right place to include mapping to the network model as well?
>
> ==
>
> Thanks!
> Dhruv
>
> _______________________________________________
> Teas mailing list
> Teas@ietf.org<mailto:Teas@ietf.org>
> https://www.ietf.org/mailman/listinfo/teas