Re: [netmod] WG Last Call: draft-ietf-netmod-rfc7223bis-00

[Martin Bjorklund <[e-mail]>]

Hi Eric,


"Eric Voit (evoit)" <[e-mail]> wrote:
> Hi Martin,
> 
> Several comments on the YANG model within rfc7223bis.
> 
> list interface {
>         key "name";
>         description
>           "The list of interfaces on the device.  The status of an interface
>           is available in this list in the
>            operational state...
> 
> A few questions on this.
> (a) The description of the list defines behaviors of various list
> nodes which might or might not exist in different NMDA datastores.  It
> also suggests when certain elements should be populated in various
> datastores.  Is the precedence being set that datastore specific
> behaviors may be placed into descriptions?

I don't know; I guess it depends on what people think about this
style.  I think that when the mapping between config and operational
state is not 1-1 and obvious, it needs to be somehow explained in the
description - just like the old descriptions did, except that where
the old descriptions refered to different *nodes*, we now refer to
different *intantiations* of the same node.


> Is this type of
> documentation guidance something which explored in
> draft-dsdt-nmda-guidelines?

I don't think so.

> (b) Does status mean 'admin-status', 'oper-status' or both?  (I think
> 'oper-status'.)

No, it is intended to mean the status of the interface in general.

> (c) should quotes be used around status?

No, since it's not the name of a specific node.

> leaf name {,   leaf type { ....
> There are NETCONF specific behaviors in the definition of these two
> leaves.  It would be great to have this transport agnostic.

One thing to note is that this also should apply to a RESTCONF
server.  But adding that doesn't really address your point :)
However, I don't know what the correct way of handling this would be.
We could be less specific and just write that the server MUST return
an "invalid-value" error, but less specific might also be more
confusing.

> I realize
> that such a transport segmentation dissociates transport error
> handling from the nodes being handled.
> 
> leaf admin-status {...
> incorrectly marked  as config false;

See the reply from Vladimir; this is corrcect.


/martin


> 
> Thanks,
> Eric
> 
> 
> > > -----邮件原件-----
> > > 发件人: netmod [mailto:[e-mail]] 代表 Kent Watsen
> > > 发送时间: 2017年11月29日 3:29
> > > 收件人: [e-mail]
> > > 抄送: [e-mail]
> > > 主题: [netmod] WG Last Call: draft-ietf-netmod-rfc7223bis-00
> > >
> > > All,
> > >
> > > This starts a two-week working group last call on draft-ietf-netmod-
> > rfc7223bis-00.
> > >
> > > Please recall that this update's intention is to modify the YANG
> > > module to
> > be in line with the NMDA guidelines [1].  Reviewing the diff between
> > the
> > two drafts [2] should reveal just this.
> > >
> > > The working group last call ends on December 12.
> > > Please send your comments to the netmod mailing list.
> > >
> > > Positive comments, e.g., "I've reviewed this document and believe it
> > > is
> > ready for publication", are welcome!
> > > This is useful and important, even from authors.
> > >
> > > [1] https://tools.ietf.org/html/draft-dsdt-nmda-guidelines-01
> > > [2]
> > > https://tools.ietf.org/rfcdiff?url2=draft-ietf-netmod-rfc7223bis-00.tx
> > > t
> > >
> > > Thank you,
> > > Netmod Chairs
> > >
> > >
> > > _______________________________________________
> > > netmod mailing list
> > > [e-mail]
> > > https://www.ietf.org/mailman/listinfo/netmod
> > > _______________________________________________
> > > netmod mailing list
> > > [e-mail]
> > > https://www.ietf.org/mailman/listinfo/netmod
> > _______________________________________________
> > netmod mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/netmod