Re: [NGO] comments on draft-li-ngo-access-mib-00

[Andy Bierman <ietf at andybierman.com>]

Hi Dan,

> So, in your view, what should the IETF do next? 
>
> - do work on the 'a combination of common standard parameters, standard
> verbs'? and/or
> - provide guidance and example to vendors about how to 'do the
> groundwork by creating new models that utilize this approach'? or
> - none of the above? 

1) Finish, publish, and use draft-romascanu-netconf-datatypes-xx,
   so we can converge on encodings for IETF-related data types.

2) Standardize a direct encoding of SMIv2 modules, based on
   the output of smidump --format=xsd.  The application of
   NETCONF verbs to this data is helpful, but not critical.
   (E.g., if we have a simpleType or element definition for 'ifMtu',
   then it can be used by reference anywhere we want -- like an
   RPC parameter, a new data model, or a notification.)

3) Explore (by design team or individual I-D) the creation of a
   NETCONF management interface for the Interface MIB modules
   and possibly the REMOPS-MIB module as well.  Interfaces are
   important and special in that they have their own config section
   (i.e., each interface instance effectively has a nested <config> sub-root).
   The output would be verbs and data models, and hopefully some implementation experience,
   which could then be considered for standardization.  It is possible
   that multiple teams and solution approaches would need to be considered.


> Dan

Andy

_______________________________________________
NGO mailing list
NGO at ietf.org
https://www1.ietf.org/mailman/listinfo/ngo