From: "Natale, Robert C (Bob)" <bnatale@lucent.com>
Subject: WG Last Call: AgentX Advancement 
Date: Mon, 1 Oct 2001 12:56:14 -0400 

Hi everyone,

I recently completed a review of the "issues list" on the AgentX web site 
and the e-mail discussion list for the past several months, in conjunction 
with a final review of the [13] implementation reports we received (and 
which were summarized on this list prior to the London IETF).  As a 
result, I would like to make a final request to the WG membership to 
comment as to whether:

1.  We should ask the IESG to advance the current
    RFCs (2741 and 2742) to Draft Standard status.

or

2.  Anyone wants to recommend specific changes to
    the documents that are *required* for successful
    implementation or interoperability.

My personal assessment is that the record argues for #1.  We received 13 
implementation reports, none of which identified spec defects preventing 
successful implementation or interoperability.  In addition, there are at 
least two known commercial implementations that did not file implementation 
reports.  A fair number of the known implementations are aimed at developers 
of SNMP-managed products and are known to have fairly wide distribution 
(e.g., several hundred development organizations).  Hence, distribution 
of AgentX technology is known to be quite wide and on-going discussions on 
several of the public e-mail lists indicate that development and deployment 
work is continuing at a fairly lively pace.

The major area (in my judgment) of potential new work for AgentX is in the 
area of sharing authentication information and/or responsibilities between 
the master agent and its body of sub-agents.  Most likely we would want 
this to be done in the context of SNMPv3 interactions between SNMPv3 command 
generator and/or notification receiver applications and the AgentX master 
agent.  If we want to pursue this, we should do so in the context of a new 
charter for the AgentX WG.  I just wanted you to know that that is on the 
table for future discussion.  It should not be a part of the determination 
as to whether to advance the current documents or modify them for advancement.
We will take it up once that primary issue is resolved.

If you feel that specific changes to the existing documents are *required* 
for successful implementation or interoperability, please post the details, 
including proposed replacement text, asap...but no later than Wed, 10/10/01 
(5pm EDT).

Questions, suggestions, comments welcomed...as always.

Cordially,

BobN