Juergen Schoenwaelder <[email protected]> wrote:
> If I put an origin on every config leaf and config p-container, why
> would I need another origin somewhere up in the np-container hierachy?
> We seem to make rules that to some extend miss the point we are really
> trying to make, namely that every config leaf and config p-container
> must have an origin (defined directly or inherited).

Yes, I think this is what we should say.


/martin


> If you choose to
> inherit origin attributes, it follows that there must be a place to
> inherit from (but that place does not have to be the root or a root).
> 
> /js
> 
> On Mon, May 04, 2020 at 02:04:41PM +0200, Martin Björklund wrote:
> > Kent Watsen <[email protected]> wrote:
> > > One small concern with the proposed NEW text is that it suggests that
> > > an NP-container is configuration, which I think is untrue.
> > 
> > An NP-container can represent config data, so I think that part of
> > Rob's suggested text is ok.
> > 
> > Here's Rob's proposed text:
> > 
> >     The origin for any top-level configuration data nodes, except
> >     non-presence containers, must be specified.
> > 
> > This doesn't say that a list within a top-level NP-container MUST have
> > "origin".
> > 
> > E.g.:
> > 
> >   container top {
> >     container second {
> >       list foo {
> >          ...
> >       }
> >     }
> >   }
> > 
> > Here /top/second/foo must have origin.
> > 
> > 
> > /martin
> > 
> > 
> > Thusly,
> > > maybe the following tweak is better?
> > > 
> > > s/except/which excludes/
> > > 
> > > NEWER:
> > >     The origin for any top-level configuration data nodes, which excludes
> > >     non-presence containers, must be specified.
> > > 
> > > Still, my preferred fix is captured at the end of the linked mail
> > > archive (i..e., fix the source definition for “data node” in RFC
> > > 7950....and reject this errata).
> > > 
> > > K.  // contributor 
> > > 
> > > 
> > > > On May 4, 2020, at 6:15 AM, Rob Wilton (rwilton)
> > > > <[email protected]> wrote:
> > > > 
> > > > Are there any other comments on the proposed resolution of this
> > > > erratum?
> > > > 
> > > > Regards,
> > > > Rob
> > > > 
> > > > 
> > > >> -----Original Message-----
> > > >> From: netmod <[email protected]> On Behalf Of Martin Björklund
> > > >> Sent: 28 April 2020 16:47
> > > >> To: [email protected]
> > > >> Cc: [email protected]
> > > >> Subject: Re: [netmod] Erratum 5514 on NMDA [RFC 8342]
> > > >> 
> > > >> "Rob Wilton \(rwilton\)" <[email protected]> wrote:
> > > >>> Hi,
> > > >>> 
> > > >>> There is one open erratum on NMDA from 2018 that I would like to
> > > >>> process.
> > > >>> 
> > > >>> The erratum is here: https://www.rfc-editor.org/errata/eid5514
> > > >>> 
> > > >>> There has been quite a lot of discussion on this erratum previously on
> > > >>> the NETMOD alias.  The last email in the thread was
> > > >>> 
> > > >> https://mailarchive.ietf.org/arch/msg/netmod/LHJZmf5gtESX6Nobwst0OwXbGG4/
> > > >>> 
> > > >>>> From my reading of the discussion, I don't think that there is clear
> > > >>>> WG consensus between the two competing concerns:
> > > >>> (1) The origin for any top-level configuration data nodes must be
> > > >>> specified (section 7, YANG annotation definition).
> > > >>> (2) The origin applies to all configuration nodes except non-presence
> > > >>> containers (section 5.3.4).
> > > >>> 
> > > >>> Hence my proposal is to mark this as "Hold for Document Update" with
> > > >>> Kent's proposed resolution of changing the description in the YANG
> > > >>> model.
> > > >>> 
> > > >>> OLD:
> > > >>>    The origin for any top-level configuration data nodes must be
> > > >>>    specified.
> > > >>> 
> > > >>> NEW:
> > > >>>    The origin for any top-level configuration data nodes, except
> > > >>>    non-presence containers, must be specified.
> > > >>> 
> > > >>> For reference, this will mean that the extension [NEW] is defined as:
> > > >>> 
> > > >>>     md:annotation origin {
> > > >>>       type origin-ref;
> > > >>>       description
> > > >>>         "The 'origin' annotation can be present on any configuration
> > > >>>          data node in the operational state datastore.  It specifies
> > > >>>          from where the node originated.  If not specified for a given
> > > >>>          configuration data node, then the origin is the same as the
> > > >>>          origin of its parent node in the data tree.  The origin for
> > > >>>          any top-level configuration data nodes, except non-presence
> > > >>>          containers,  must be specified.";
> > > >>>     }
> > > >>> 
> > > >>> Please can you let me know if you support or object to this
> > > >>> resolution.  I'll leave it a week to see if there is consensus before
> > > >>> processing the erratum.
> > > >> 
> > > >> I think this is ok.
> > > >> 
> > > >> 
> > > >> /martin
> > > >> 
> > > >> 
> > > >> 
> > > >> _______________________________________________
> > > >> netmod mailing list
> > > >> [email protected]
> > > >> https://www.ietf.org/mailman/listinfo/netmod
> > > > 
> > > > _______________________________________________
> > > > netmod mailing list
> > > > [email protected]
> > > > https://www.ietf.org/mailman/listinfo/netmod
> > _______________________________________________
> > netmod mailing list
> > [email protected]
> > https://www.ietf.org/mailman/listinfo/netmod
> 
> -- 
> Juergen Schoenwaelder           Jacobs University Bremen gGmbH
> Phone: +49 421 200 3587         Campus Ring 1 | 28759 Bremen | Germany
> Fax:   +49 421 200 3103         <https://www.jacobs-university.de/>
_______________________________________________
netmod mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/netmod

Reply via email to