IMO nothing intended to be tunable by any customer should require the
use of mdb or destructive dtrace.  There has to be a safe interface
for administrative actions and neither of those are it.

Today I can tell the services guys to ask the customer to perform
an ndd query to obtain settings, or set a specific variable using ndd.
(virtually a game of chinese whispers).

Such instructions are easily followed and errors are non fatal.

mdb in particular is like using a tactical nuke to kill a fly,
you don't want a typo to ruin your day, particularly on a production
server running your bank.

-George



*>Date: Wed, 16 Apr 2008 11:47:40 -0400
*>From: [EMAIL PROTECTED]
*>Subject: [networking-discuss] replacing/cleaning ndd at the tcp/ip layer
*>To: [email protected]
*>MIME-version: 1.0
*>Content-transfer-encoding: 7BIT
*>Content-disposition: inline
*>X-BeenThere: [email protected]
*>Delivered-to: [email protected]
*>X-PMX-Version: 5.4.1.325704
*>X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on 
oss-mail1.opensolaris.org
*>X-Original-To: [email protected]
*>X-Antispam: No, score=-2.6/5.0, scanned in 0.082sec at (localhost 
[127.0.0.1]) by smf-spamd v1.3.1 - http://smfs.sf.net/
*>X-Mailman-Version: 2.1.9
*>X-Authentication-warning: quasimodo.East.Sun.COM: sowmini set sender to 
[EMAIL PROTECTED] using -f
*>List-Post: <mailto:[email protected]>
*>List-Subscribe:  
<http://mail.opensolaris.org/mailman/listinfo/networking-discuss>, 
<mailto:[EMAIL PROTECTED]>
*>List-Unsubscribe:  
<http://mail.opensolaris.org/mailman/listinfo/networking-discuss>, 
<mailto:[EMAIL PROTECTED]>
*>List-Archive: <http://mail.opensolaris.org/pipermail/networking-discuss>
*>List-Help: <mailto:[EMAIL PROTECTED]>
*>List-Id: Networking General Discussion <networking-discuss.opensolaris.org>
*>User-Agent: Mutt/1.5.17 (2008-01-17)
*>X-Spam-Status: No, score=-1.0 required=5.0 tests=AWL,RCVD_IN_DNSWL_LOW 
autolearn=unavailable version=3.2.3
*>X-Spam-Level: 
*>
*>
*>Now that Brussels has provided an alternative for ndd at the
*>datalink layer, it is time to re-examine the appropriateness of
*>some of the ndd tunables in the TCP/IP layer. ndd has
*>taken on many "tunables" that should have been implemented
*>with mdb/dtrace, or implemented by other methods.
*>
*>I'm trying to compile a list of the current ndd vars
*>to identify the true tunables from the ones that
*>are confusing debug knobs or diagnostic hooks. 
*>
*>It's still work-in-progress (lots of empty cells 
*>waiting for content) but please send any comments my way,
*>so that we can clean up ndd and potentially evaluate what
*>ndd (or its replacement candidate) must/should/must-not 
*>deal with.
*>
*>Help with filling up the cells would also be welcome,
*>of course!
*>
*>Location of table:
*>  http://cr.opensolaris.org/~sowmini/ndd.html
*>
*>--Sowmini
*> 
*>_______________________________________________
*>networking-discuss mailing list
*>[email protected]


George Shepherd
http://clem.uk/~georges/
==============================================================================
   Solaris Revenue Product Engineering:    |  SUN Microsystems
       Core team  -Internet                |  Guillemont Park
   Email: [EMAIL PROTECTED]          |  Camberley GU17 9QG
   Disclaimer: Less is more, more or less  |  United Kingdom 
==============================================================================

_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to