Re: Thoughts on NetConf Requirements

[Andy Bierman <abierman@cisco.com>]

At 06:38 AM 6/4/2003, Randy Bush wrote:
>> This NETCONF effort is multi-dimensional, and both simple and complex
>> at the same time. Yes, one of the primary issues is being able to
>> send and retrieve configuration; but it also must retrieve status,
>> statistics, and asynchronous notifications of events and trace output.
>
>imiho, if it unduly complicates things, ditch all this stuff about
>monitoring state, receiving [non-config] alters, ...  this is the
>netCONF wg, not the netMANAGEMENT wg.

I agree, but I want netconf to be capable of retrieving device state.
I think writing config is more complex than reading config or state
data.  I have not seen a single argument that proves otherwise.

So what's so complex about reads? The authorization model?
IMO, the authorization model amounts to a set of tuples:
   { user, operations-allowed, element-subtree }.
Read operations amount to 1 more bit in the operations-allowed field.

The only other issue related mostly to reads is the volume of
returned data.  IMO, the application streams data and the transport
handles fragmentation, so this is a non-issue.

Notifications are still an more TBD issue.  xmlconf punts the issue
by using RFC 3195.  The WG has not really discussed the details
of notifications yet, but I don't see why this needs to be complex.

I hope we all can agree that the essence of good engineering
is how simple you can make the solution, not how complex it
can be.  


>randy

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/>