[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Agent Capabilities and sysORTable

Hi all,
I second Steve's caution about protocol capabilities MIB and understanding
the dynamics around agent capabilites. My experience with agent capabilites
is limited to trying to understanding it.

We tried to recommend the use of Agent capabilites as a company wide
initiative at my previous place of work. Firstly it was not straight forward
to figure out. Each of us had our own interpretation with no example to
follow. Secondly, it was quite challenging trying to explain/sell it to
other engineers/managers who did not have the SNMP background (for that
matter, a lot of us with some background were having an interesting time
with it). We used toolkits for most agent developement. We looked at the
vendors for more explanation on agent-capabilites as most of the vendors
"support mib-2" as a part of the deal. When I hear mib-2 support, I expect
to be told, oh by the way you need to do "blah blah..." to add agent
capabilites, Since I am not going to be messing around with the system
group, no such luck:-)) We ended up not using agent-capabilites for all the
reasons David has mentioned and more.

Personally, I as an agent/EMS developer would like to see some more
defination around agent capabilites.

Aparna Amin

-----Original Message-----
From: David T. Perkins [mailto:dperkins@dsperkins.com]
Sent: Tuesday, April 17, 2001 4:42 PM
To: eos@ops.ietf.org
Subject: Re: Agent Capabilities and sysORTable


Reasons why AGENT-CAPs are not used:
1) there are no examples in IETF documents (other than RFC 2580),
     and so engineers have nothing from which to cut, paste,
     adapt, and publish
2) agent-caps have problems - there is a missing level of indirection,
     some of the clause don't make sense, etc
3) it is difficult to create correct ones
4) the value of them has not been shown
5) agent-caps don't provide all the information needed for an application
     to determine if it will interoperate with an agent. Thus, human
     directed testing is needed anyway, so no value having agent-caps.

At 09:19 AM 4/17/2001 -0400, Thomas, Steven C (Steve) wrote:
>Hey all,
>I need some help understanding here...
>When I first saw the Charter discussing capabilities, I thought this might
>be related to the agent capabilities addressed in the sysORTable.  Since
>then I understand that this part of the eos charter is referring to
>capabilities.  However, having the agent document its MIB capabilities
>be desirable also.
>My question is:  Why don't most agents (I haven't found any so far)
>implement or at least populate the sysORTable?  Is there some hidden
>difficulty in implementing it on the agent side?  Or have I misunderstood
>the purpose of sysORTable?
>My comment is: If the agents don't bother to populate the sysORTable to
>document their agent MIB capabilities, why would we expect them to populate
>some other MIB documenting protocol capabilities?  We need to understand
>dynamics in order to overcome them.
>I am not arguing that the protocol capabilities MIB or mechanism isn't
>needed, but I am cautious considering my experience so far with sysORTable.

/david t. perkins


Note: The information contained in this message may be privileged and
confidential and protected from disclosure. If the reader of this message is
not the intended recipient, or an employee or agent responsible for
delivering this message to the intended recipient, you are hereby notified
that any dissemination, distribution or copying of this communication is
strictly prohibited. If you have received this communication in error,
please notify us immediately by replying to the message and deleting it from
your computer. 
Thank you. Hatteras Networks, Inc.