RE: FW: I-D ACTION:draft-weijing-netconf-interface-00.txt

["Allen, Keith" <Keith_Allen@labs.sbc.com>]

> Juergen Schoenwaelder writes:

> From a very abstract viewpoint, I believe that a data-centric or
> object-centric management protocol must support the following set of
> essential management protocol primitives:
>
>    o  get, set, create, delete
>
>    o  search (or at the very least iterate)
>
>    o  lock, unlock, commit, rollback
>
>    o  notify
>
>    o  execute or invoke an operation

Juergen,

I think we would agree with you (not sure if the lock and unlock are
strictly needed as primitives).   I think the challenge, though, is how to
do this with the assumption that XML schema will be used to define the
"objects."

Keith Allen
SBC Labs
9505 Arboretum Blvd.
Austin, TX 78759
(512) 372-5741
keith_allen@labs.sbc.com
 
This e-mail and any files transmitted with it are the property of SBC, are
confidential, and are intended solely for the use of the individual or
entity to whom this e-mail is addressed. If you are not one of the named
recipient(s) or otherwise have reason to believe that you have received this
message in error, please notify the sender at (512) 372-5000 and delete this
message immediately from your computer. Any other use, retention,
dissemination, forwarding, printing, or copying of this e-mail is strictly
prohibited.


-----Original Message-----
From: Juergen Schoenwaelder [mailto:schoenw@ibr.cs.tu-bs.de] 
Sent: Friday, July 04, 2003 2:21 AM
To: Weijing_Chen@labs.sbc.com
Cc: netconf@ops.ietf.org
Subject: Re: FW: I-D ACTION:draft-weijing-netconf-interface-00.txt


>>>>> Chen, Weijing writes:

Weijing> I guess to summarize our philosophy, I would say we think
Weijing> that a management protocol should be divided into two parts,
Weijing> a basic set of management "operations" and an "information
Weijing> model" that defines the manageable capabilities of a device.
Weijing> Probably no argument there.  Where I think we may begin to
Weijing> disagree, however, is that we feel the management operations
Weijing> should be as simple and flexible as possible, with the
Weijing> information model being the focus for defining the complexity
Weijing> of the device.  Wherever possible, our preference would be to
Weijing> make the protocol simpler and the information model more
Weijing> complex.

The question is where you draw the line between protocol primitives
and what is being done in the information model. SNMP is a well known
example where the desire to stick to the very basic set of protocol
primitives and to do everything else in the data models led to thinks
like RowStatus which at the end have been more expensive then adding
create/delete operations to the protocol.

From a very abstract viewpoint, I believe that a data-centric or
object-centric management protocol must support the following set of
essential management protocol primitives:

   o  get, set, create, delete

   o  search (or at the very least iterate)

   o  lock, unlock, commit, rollback

   o  notify

   o  execute or invoke an operation

Introduction of additional primitives should be done with great care
because either (a) the abstract model above is wrong and needs to be
corrected or (b) the new primitives are solely justified because they
allow for significant optimizations.

/js

-- 
Juergen Schoenwaelder		    International University Bremen
<http://www.eecs.iu-bremen.de/>	    P.O. Box 750 561, 28725 Bremen, Germany

--
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/>

--
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/>