Netconf Data Model Netmod BOF IETF 60

1
Netconf Data ModelNetmod BOF IETF 60

• Sharon Chisholm schishol_at_nortelnetworks,com
• Randy Presuhn - randy_presuhn_at_mindspring.com

2
Outline

• Preliminaries
• Problem Statement/Overview
• Charter Discussion

3
Preliminaries

• Administrivia
• Agenda
• Website, Mailing List
• BOF Status
• Additional Reading

4
Administrivia

• Blue sheets
• Minute Takers
• Jabber Scribe
• Instructions to Presenters
• http//www.ietf.org/instructions/instructions.html

5
Agenda

• Preliminaries - 15 minutes
• Problem Statement/Scope Overview - 10 minutes
• Framework for Netconf Data Models - Sharon
  Chisholm - 15 min
• http//www.ietf.org/internet-drafts/draft-chisholm
  -netconf-model-00.txt
• Netconf Data Model - Sandeep Adwanker, 20
  minutes
• http//www.ietf.org/internet-drafts/draft-adwankar
  -netconf-datamodel-00.txt
• Getting to a Netconf Data Model - Andrea
  Westerinen - 10 minutes
• http//standards.nortelnetworks.com/netconf/docs/G
  ettingToANetConfDataModel.txt
• Netconf Architecture Model Ray Atarashi et al.
  - 10 minutes
• General Discussion - 15 minutes
• Charter Discussion and Wrap-up - 10 minutes

6
BOF Coordinates

• Web page
• (which has a reference to the mailing list,
  archives, papers and presentations with proposed
  solutions)
• http//standards.nortelnetworks.com/netconf/
• Mailing List
• The mailing list address is netconfmodel_at_lyris.nor
  telnetworks.com
• To subscribe to the list send an email to
  lyris_at_lists.nortelnetworks.com - with the words
  "subscribe netconfmodel" (no quotes) in the body
  of the message (no subject necessary).
• MAIL ARCHIVES are maintained at the following
  URL http//standards.nortelnetworks.com/archives
  /netconfmodel.html

7
BOF Status

• Held Bar BOF for Netconf Data Model discussions
  during IETF 59
• Mailing List since March 2004
• Three Individual Internet Drafts, one informal
  discussion paper

8
Additional Reading

• Netconf Configuration Protocol http//www.ietf.org
  /internet-drafts/draft-ietf-netconf-prot-03.txt
• "On the Difference between Information Models and
  Data Models", RFC 3444 http//www.ietf.org/rfc/rfc
  3444.txt

9
Problem Statement Overview

• Netconf layering
• Problem Statement
• Building Blocks for Content
• Strategy
• Requirements CLRs
• Discussed Deliverables

10
Netconf Layering

• Layer
  Example
• -------------
  -----------------------------
• Content
  Configuration data
• -------------
  -----------------------------
• -------------
  -----------------------------
• Operations ltget-configgt,
  ltedit-configgt
• -------------
  -----------------------------
• -------------
  -----------------------------
• RPC ltrpcgt,
  ltrpc-replygt
• -------------
  -----------------------------
• -------------
  -----------------------------
• Application BEEP, SSH,
  SSL, console
• Protocol
  
• -------------
  -----------------------------

11
Problem Statement

• netconf working group is chartered to produce a
  protocol for network configuration
• the data models to be used with this new protocol
  are outside the scope of that discussion.
• netconf architecture proposes to be independent
  of data definition and data model,
• need to start talking about data models in more
  concrete terms to ensure there really aren't
  implications for either the protocol or the
  models as a result of combining them into system
  to provide management functionality.
• Need agreed to common ways of specifying
  compliance, maintaining backwards compatibility,
  defining relationships, naming, identification,
  access control, etc.
• In addition, many feel that the identification or
  creation of standard data models for use in
  netconf is critical for both the success of the
  protocol and the benefit of the industry.

12
Building Blocks for Content
Tools for Creating Content
W3C XML Schema
Framework for Netconf Data Models SMI for
Netconf
Meta-model or Information Model
Netconf Data Types
Content
Standard Data Models
Proprietary Data Models
13
Strategy

• Applicable to all content IETF Proprietary
• Leverage existing technology
• Prioritize on delivering the Framework Document
• Capture requirements without rat holing
• Framework
• We focus syntax restrictions on those that enable
  interoperability, implementability, parsability,
  backwards compatibility, readability, and other
  'bilities' as required.
• We should do a gap analysis compared to existing
  W3C XML methods and only innovate as necessary.
• We should be careful to not create CLRs.
• We should capture both our use of W3C XML methods
  to meeting specific requirements as well as our
  own innovations
• We should evaluate our innovations for possible
  inclusion back into W3C XML specifications

14
Requirements CLRs

• CLRs
• rules that get introduced with the best
  intensions, but later are felt to either place
  unnecessarily limiting restrictions on the
  solution or waste review cycles to ensure
  compliance.
• Requirements
• Necessary to understand to ensure reasonable
  solution and to form common understanding of
  goals.
• Historically has caused delays or working group
  death
• Proposal
• Capture need being met by each introduced portion
  of netconf data model along with the solution.
  Benefits include
• Puts requirements in context of the proposed
  solution which should help keep focus
• Documents reasoning behind all introduced rules
  to hopefully prevent creation of CLRs or to
  provide the historical context to deal with the
  in the future

15
Discussed Deliverables

• Data Types
• Minimal set of core data types for use in network
  elements
• IP Address, etc
• Syntax
• SMI equivalent
• compliance, backwards compatibility, etc
• Meta Modeling
• Framework to ensure
• Consistency,
• Constraints beyond syntax
• Referential Integrity?
• Individual data models combine together into a
  cohesive whole
• Well defined relationships
• Solution could range from a BCP or design
  patterns to a full fledged information model
• Mappings to other solutions, like SNMP

16
Charter Discussion

• Proposed Charter
• Next Steps

17
Proposed Charter

• The Netconf Data Model Working Group would be
  charted to provide an initial
• framework to create XML data models for use with
  the Netconf protocol.
• Abstract information models on which all XML data
  models would be based may
• also be discussed. The working group will also
  produce some initial XML data
• models as proof of concept examples, as well as
  to meet specific industry
• need.
• The network configuration (netconf -
• http//www.ietf.org/html.charters/netconf-charter.
  html ) working group in the
• IETF is chartered to produce a protocol for
  network configuration. The data
• models to be used with this new protocol are
  outside the scope of that
• discussion. Even though the netconf architecture
  proposes to be independent
• of data definition and data model, it is critical
  to start talking about
• data models in more concrete terms to ensure
  there really aren't
• implications for either the protocol or the
  models as a result of combining
• them into system to provide management
  functionality.

18
Proposed Charter - Continued

• In addition, many feel that the identification or
  creation of standard data
• models for use in netconf is critical for both
  the success of the protocol
• and the benefit of the industry.
• The working group will complete these tasks
• Define a 'Framework for Netconf Data Models'
  Document Exact Title TDB
• Define a 'Netconf Data Model for system,
  interfaces and physical entities'
• The working group will consider existing
  definitions, including
• o Netconf Protocol Work
• o W3C XML Schema
• o Existing Information Models (SNMP, CIM from
  DMTF, SID from TMF, etc)

19
Proposed Charter Goals Milestones

• Done Bar BOF IETF 59
• June 2004 Initial draft of Framework document
• Aug 2004 Netconf Data Model BOF at IETF 60
• Sept 2004 Version of Framework document as
  Working Group document
• Mar 2005 Working Group Last Call On Framework
  document
• Oct 2005 Working Group Last Call on Complex Data
  Types document(s)

20
Next Steps

• Do you think this is an area worth working on?
• Are you personally willing to invest time
  providing content, editing, reviewing, etc?
• Should we charter a working group with this
  charter?

21
(No Transcript)