Re: FW: [PEPPERMINT] new BoF proposal

[Andy Bierman <ietf@andybierman.com>]

Joel M. Halpern wrote:
> You should only point them at NETCONF if we are prepared to help them 
> define a data model that actually works with NETCONF, instead of the 
> working groups perpetual arguments about what a data model might look 
> like, or why we won't define a data model, or ...
> A way of encapsulating CLI (which is all that NETCONF has so far managed 
> to provide) does not solve the problem at all.

I'm not writing off NETCONF a month after the RFCs are published.
I expect early NE implementations to be "wide and shallow" mappings
from proprietary CLI to proprietary encapsulated CLI.  More interesting
implementations will follow later.

IMO, the real issue is whether any WG in question can agree on
the information model and operational model for a particular feature.
If you can get that far, then many different protocols and encodings
will work.  We have very few standards for configuration because
vendors have not agreed on very many common 'knobs', let alone common
operational procedures.

The overhead of sessions, configuration databases, and other NETCONF-centric
details may be reason enough to create something new, but even NE provisioning
apps today have to integrate with CLI-based configuration management.
The non-volatile config file on the device usually needs to be updated,
regardless of the API.  NETCONF is designed to work with this type of device.

> Yours,
> Joel M. Halpern

Andy

--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>