Re: [Rtg-yang-coord] Operational State Modeling

[Juergen Schoenwaelder <[e-mail]>]

On Fri, May 15, 2015 at 09:53:17AM -0700, aldrin ietf wrote:
> It is totally misconstrued and wrong to think it is being done for the looks. It is clearly outlined in the draft on ‘why?'(<http://datatracker.ietf.org/doc/draft-openconfig-netmod-opstate/?include_text=1 <http://datatracker.ietf.org/doc/draft-openconfig-netmod-opstate/?include_text=1>>. 
> 
> We, the operator community, are requesting for a reason as outlined in the draft, to make it operationally efficient, when managing network @scale.
>

I read this I-D and I agree to some what it says. To some of what it
says I tend to disagree. My main concern is that trying to infer
semantics from structure is

- brittle, and

- it is for me more than unlikely that all YANG models will follow a
  common structure.

I also do not see anything wrong with the approach taken by RFC 7223
and on the other hand I do see problems with the example shown in
section 4 of the I-D (interface/name points to config and hence state
is bound to the lifetime of config, a limitation RFC 7223 does not
have - you can actually see unconfigured interfaces).

The bottom line is that I do not subscribe to all requirements stated
in the I-D and I do not subscribe to the solutions proposed. Nor have
I seen something that clearly tells me the structure we used in RFC
7223 is broken.

/js

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: [phone]         Campus Ring 1 | 28759 Bremen | Germany
Fax:   [phone]         <http://www.jacobs-university.de/>