Re: [core] CoMI Cool draft splits

I agree that the CoMI approach is better in this respect.
From: Andy Bierman [mailto:andy@yumaworks.com]
Sent: Friday, November 20, 2015 11:49 AM
To: Rodney Cummings <rodney.cummings@ni.com>
Cc: Carsten Bormann <cabo@tzi.org>; Core <core@ietf.org>
Subject: Re: [core] CoMI Cool draft splits

Hi,

I think it will a challenge for the CORE WG to get IANA to manage
the manual number assignment of even IETF modules, let alone all
YANG modules from all vendors and all other SDOs. The cost of even 1 mistake
is rather high (like a wrong port assignment).

YANG Hash will allow any YANG module that has ever been written
or ever could be written to be used without modification or manual
number assignment.

I completely agree with you about CoMI, but not CoOL.

Andy

On Fri, Nov 20, 2015 at 9:36 AM, Rodney Cummings <rodney.cummings@ni.com<mailto:rodney.cummings@ni.com>> wrote:
I agree with Carsten on the point below.

As an example, the YANG development in progress for 802.1Q is using features, augments, and so on in order to build up modules from the basic components to more complex components (e.g. for providers). The basic components essentially represent the hardware registers of a switch. As constrained products explore integration of an 802.1Q switch in their product, the question of a constrained network management protocol comes up. IMO CoMI/CoOL provides an excellent solution for this use case, using the 802.1Q module(s) for basic components (i.e. modules not specifically targeted to constrained).

> -----Original Message-----
> From: core [mailto:core-bounces@ietf.org<mailto:core-bounces@ietf.org>] On Behalf Of Carsten Bormann
> Sent: Friday, November 20, 2015 12:36 AM
> To: Andy Bierman <andy@yumaworks.com<mailto:andy@yumaworks.com>>
> Cc: Core <core@ietf.org<mailto:core@ietf.org>>
> Subject: Re: [core] CoMI Cool draft splits
>
> Andy Bierman wrote:
> > [...] IMO constrained nodes
> > will never be using the same modules that are written for networking
> > equipment.
>
> Is that so?  Why wouldn't I use a couple of items from, say, RFC 7317
> (or 7277) in my constrained node?  In the end these are just resources
> that I add to my resource tree.  Of course, cherry-picking may require
> some form of subsetting that may not yet be part of the NETCONF approach.
>
> Grüße, Carsten
>
> _______________________________________________
> core mailing list
> core@ietf.org<mailto:core@ietf.org>
> https://www.ietf.org/mailman/listinfo/core