RE: [nv-l] Configuration of Event Threshold

[James Shanks <jshanks@us.ibm.com>]

I, for one, am having a very hard time following this, and based on the seven years I have spent working with distributed NetView, I am having an even harder time believing that something is broken here.  I really do work in this area every day.

All NetView traps, including  the snmpCollect Threshold Exceeded trap 58720263 should have the hostname of the device they are about in varbind 2, which you reference in trapd.conf and pass to ovactiond as "$2".  And they all have a minimum of 5 variables.  Your snmptrap script has neither of those and so I doubt that it truly represents what is being sent by NetView itself.  The "threshold exceeded" message should be $3 not $2.   This is as documented in Appendix A of the NetView for UNIX Administrator's Guide.

My suggestion is that you configure this trap in trapd.conf (using xnmtrap) so that it says $* on the Log Message line.  This will cause all the variables to be echoed to the log and the events window in numerical order.    If it truly is that case that a real NetView event does not meet both these criteria as I have stated then you indeed have a bug and should contact Support immediately.  

James Shanks
Level 3 Support  for Tivoli NetView for UNIX and Windows
Tivoli Software / IBM Software Group

Peter.Ingleson@gerlingncm.com Sent by: owner-nv-l-digest@lists.us.ibm.com 10/15/2003 07:22 AM Please respond to nv-l

To:        nv-l@lists.us.ibm.com         cc:                 Subject:        RE: [nv-l] Configuration of Event Threshold

List, 

Ive done some testing on this issue and I keep getting differences between
the variables.

Im using the following to simulate traps entering into the system 

snmptrap 1.1.1.1  .1.3.6.1.4.1.2.6.3 2.2.2.2 6 58720263 0
.1.3.6.1.4.1.9.2.2.1.1.12 integer 1 0.0 octetstring "threshold exceeded" 0.0
octetstring "locIfInCRC"   .1.3.6.1.4.1.9.2.2.1.1.12 integer 1   

Where 1.1.1.1 is my netview box and 2.2.2.2 is my monitored device

Using the $2 variable I get threshold exceed 
Using the $A variable I get the correct host name

When threshold monitoring fires (ie not using the above script), using $A I
get discrepancies in the value, and $2 simply carries threshold exceed. The
popup notification contained the host where the collection was targeting,
whilst the value passed over to automatic action contained the Netview Host
box. 

I've double checked the sequence of variables all they are both the same. 

Any thoughts would be appreciated, alternatively if someone can think of
another way to generate an email based on a threshold being passed I would
appreciate it. 

Regards
Peter 

      



-----Original Message-----
From: Binder, Karin [mailto:karin.binder@nwa.com]
Sent: 10 October 2003 22:15
To: nv-l@lists.us.ibm.com
Subject: RE: [nv-l] Configuration of Event Threshold


Ran into something similar.  This is from memory.  My test box is
overwhelmed at the moment so cannot verify.  IRRC, $A is the name of the
node that sent the event.  For a Netview event, that's the Netview server;
For an externally generated trap, it's the hostname of the device issuing
the trap.  As James suggested, use $2 as appropriate, or $NVATTR_2 in a
script called by a ruleset.  

Karin

-----Original Message-----
From: Kevin C Li [mailto:kevinli@us.ibm.com]
Sent: Friday, October 10, 2003 1:40 PM
To: nv-l@lists.us.ibm.com
Subject: Re: [nv-l] Configuration of Event Threshold



I've seen this kind of problem at least a couple of times.  One time when I
saw this I traced it to a name resolution problem.  I don't remember the
exact details anymore, but the situation was something like this: NetView
had the hostname of the agent in its database, but the DNS server was down
and NetView couldn't resolve it back to an IP address.  As a result,
NetView substituted its own hostname/IP in the agent address field.  This
was with NetView 5.x on AIX.

There has been one other time when I've seen this but wasn't able to find a
cause.


Kevin Li
IBM Global Services





 

                      Peter.Ingleson@gerlingn

                      cm.com                         To:
nv-l@lists.us.ibm.com

                      Sent by:                       cc:

                      owner-nv-l-digest@lists        Subject:  [nv-l]
Configuration of Event Threshold                                       
                      .us.ibm.com

 

 

                      10/10/2003 08:52 AM

                      Please respond to nv-l

 

 





List,

I have AIX5.1 with NV7.1.3,

I'm using threshold checking to monitoring errors on an array of cisco
routers. This works correctly and re-arms the monitor when the condition
meets, however I seem to be having a slight problem with trap customisation
(IBM_NVDCOL_EV).

In the pop up notification I have the following

$A $3 $4 $+1 which works correctly, ie $A displays the hostname of the
device with the errors, a router.

In command for automatic action I have the following

/home/$path/script.sh $A $3 $4 $+1

But the value in $A is the hostname of my Netview box and the not the
monitored device.

Any thoughts would be appreciated.

Regards
Peter



----------------------------------------------------------------------------
--------------------------------

This message is intended for the addressee or its representative only.
Any form of unauthorized use, publication, reproduction, copying or
disclosure of the content of this e-mail is not permitted. If you are
not the intended recipient of this e-mail message and its contents,
please notify the sender immediately and delete this message and
all its attachments subsequently.