Re: [Netconf] YangPush now

[Kent Watsen <[e-mail]>]

>> <chair hat on>
>> ...
>> Assuming no objections, to close the issues discussed in Montreal, we're waiting 
>> for the following updates:
>> 
>> ...
>>  netconf-notif: modify to only reflect netconf for dynamic subscriptions
>
> This would be a tweak from case A2 from the WG session slides.   Personally
> I have no issue with this version of A2.  You and I have already scoped the
> changes.  Plus I don't know of anyone who has an issue with A2.   But we
> left the WG session with case A1 being selected.

True, but recall we only did so because the previous A2 was deemed not 
viable, and no one stepped forward to work on A3.  A1 wasn't really
"selected" in my view.  But with the insertion of a mandatory transport
into SN config model, it becomes viable again from a spec perspective,
even if not immediately viable from a market perspective.


> Procedurally how long must we wait before such a scope change becomes WG
> consensus?  Are there mandatory people you feel must chime in to support
> the change?

As explained before the meeting, the goal was to have a snap-decision on
how to sequence the work.  Regardless which sequence chosen, the same
consensus-based result should be achieved in time.  As such, the chairs
didn't feel the need to confirm this "sequencing" decision on the list.

That said, it seems that Andy and Robert are today aligning with this
approach, and I know that you and myself (as a contributor) also support
this approach...and I know of no objectors, all of which looks like
consensus to me.  [Also note that this is probably the only path that
might get these drafts to LC before Bangkok.]

That said:

  1) my preference is to have *no* netconf-notif or restconf-notif drafts
     if possible and, if it is, then this reduced netconf-notif scope is
     a moot point...as dynamic subscriptions just work, for both NC and
     RC automatically.

  2) however, if netconf-notif MUST be defined, or else NC-based dynamic
     subscriptions are undefined, then it follows that RC-based dynamic
     subscriptions are also undefined, unless a restconf-notif is also
     published.  And since the WG treats both protocols symmetrically,
     I added RC-notif (for dynamic-only) as also being a needed.  Makes
     sense and should be fairly easy, right?  [hint: it's just restconf]

We should see if (1) is possible, and if not, go for (2).  This shouldn't
require any more consensus than we have already.  That said, as always, if
anyone objects, please speak now!


Thanks,
Kent