The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2002-Jun> msg00073



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

MPLS Management Overview

  • From: "Thomas D. Nadeau" <tnadeau@cisco.com>
  • Date: Thu, 13 Jun 2002 21:33:19 -0400
  • Cc: "Adrian Farrel" <afarrel@movaz.com>, "mpls@uu.net" <mpls@UU.NET>



>Thanks for the info. This overview document is quite helpful.
>
>One point to raise, though. In comparing TE-MIB (by te-wg) and
>MPLS-TE-MIB (by mpls-wg), the document failed to point out that
>the TE-MIB is structured better than the MPLS-TE-MIB.


         Rupert, the point of the document IS NOT to
have a spitting match over which MIB you think is
better that which.  We had a nice one about a year and a half
ago. If you feel like doing such a comparison on your own
feel free.

         --Tom



>Take a look at TE-MIB's structure.
>
>http://search.ietf.org/internet-drafts/draft-ietf-tewg-mib-02.txt
>
>     Table Name                            # of Fields
>     -------------------------------------------------
>     teAdminGroupTable                         2
>     teTunnelTable                            21
>     tePathTable                              16
>     tePathHopTable                            5
>     -------------------------------------------------
>
>The TE-MIB makes good distinction of "logical tunnel" from
>underlying "path", and models the two distinct entity classes
>with separate MIB tables.  In addition, the name of
>"tePathHopTable" clearly indicates that "hop list" is a property
>of "path" rather than "logical tunnel".  A very clean design.
>
>In contrast, the current MPLS-TE-MIB lumps attributes of "logical
>tunnel" together with attributes of underlying "path", and makes
>a big, monolithic table ("mplsTunnelTable").  This approach is
>problematic, IMHO.
>
>Well, I do recognize the excellent achievement in putting together
>the comprehensive MPLS-TE-MIB and appreciate the great effort by the
>authors.  Just want to point out that it is not perfect, yet.
>
>Regards,
>
>
>Rupert Zhu
>
>System Engineering
>Santera Systems Inc.
>972-461-6383 (TX, USA)
>
>
> > -----Original Message-----
> > From: Adrian Farrel [mailto:afarrel@movaz.com]
> > Sent: Thursday, June 13, 2002 2:13 PM
> > To: 'mpls@uu.net'
> > Subject: MPLS Management Overview
> >
> >
> > Hi,
> > A new version of the MPLS Management Overview has been
> > posted.  I didn't see a
> > notification to the mail exploder, but I may have been sleeping.
> >
> > Main changes are an increase in detail and a description of
> > the inter-relation
> > between MIB modules, and between MIB tables.
> >
> > http://search.ietf.org/internet-drafts/draft-ietf-mpls-mgmt-ov
>erview-02.txt
>
>Adrian



------------------------------------------------------------------------
Mathematics is the supreme nostalgia of our time.