Gerald Andrew Winters <gerald@citi.umich.edu> Thu, 27 April 1995 20:01 UTC

Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa08994; 27 Apr 95 16:01 EDT
Received: from CNRI.Reston.VA.US by IETF.CNRI.Reston.VA.US id aa08990; 27 Apr 95 16:01 EDT
Received: from wugate.wustl.edu by CNRI.Reston.VA.US id aa13391; 27 Apr 95 16:01 EDT
Received: from host (localhost [127.0.0.1]) by wugate.wustl.edu (8.6.11/8.6.11) with SMTP id PAA27173; Thu, 27 Apr 1995 15:01:38 -0500
Received: from citi.umich.edu (citi.umich.edu [141.211.128.16]) by wugate.wustl.edu (8.6.11/8.6.11) with SMTP id PAA27129 for <oswg-l@wugate.wustl.edu>; Thu, 27 Apr 1995 15:01:04 -0500
Message-Id: <199504272001.PAA27129@wugate.wustl.edu>
Date: Thu, 27 Apr 1995 16:00:16 -0400
Reply-To: oswg-l@wugate.wustl.edu
X-Orig-Sender: owner-oswg-l@wugate.wustl.edu
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Gerald Andrew Winters <gerald@citi.umich.edu>
To: oswg-l@wugate.wustl.edu, mail@citi.umich.edu
Cc: dmuntz@citi.umich.edu, jeffya@teleport.com
X-Listprocessor-Version: 7.1 -- ListProcessor by CREN

> 4. Can someone clarify the "link-name" in the storage format?

Because of the fact that RFC1404 is written with a bias
toward router stats, I think Mr. Stockman was referring to
an interface connected to a point-to-point link rather than
to an interface attached to a subnetwork.  Mr. Clark's example
is what I think the 1404 wg had in mind...

>It's the name of the thing connecting one device to another, like a
>T1 between two routers that you call "boston-newyork"...

In the opstat model, I believe "interface" is now used instead
of the operation model "linkname".  Someone please correct me if
I'm wrong.
 
[Another question] Variable names...
On page 15 of RFC1404 it states: "Variable names are the fully
qualified Internet MIB names."  But nowhere in RFC1404 or the
Opstat i-d will you find a fully qualified MIB name.  The popular
example seems to be "ifInOctets".  Can someone tell me if the
definition is wrong or the examples are wrong?
  Continuing on this same line, did the wg include in their design
the possibility of using a table prefix for the variable field  
in a device section tag-table (as suggested by Alex Bochannek)?       
For example...

BEGIN_DEVICE
   ....
UNI-1,total,iso.org.dod.internet.mgmt.mib-2.interfaces.iftable,60,60

The intent is to dump the entire interfaces table every 60 seconds.
If I can do something like this, then I am not limited to just
mib-II.  I can poll the RMON mib, the enterprises mib, etc...
The problem I see is in reconciling the data values in the data
section to the corresponding mib variables.  Whats the answer here?

--Gerald Winters 

ps Sorry about the previous post, my modem connection died
while I was writing the msg.
  •   Gerald Andrew Winters