Re: How can I threshold Node

I like your idea on ways to eliminate erroneous events, but I have a
question.

Did you use rulesets to execute the shell script using an 'action' node?  Or
did you just have the trap execute the shell script?

When a storm of real node down events occur, does performance on your
Netview system degrade alot?  I would think that it would be effected even
more than normal due to having to spawn all the pings...

I currently use a ruleset for interface down events and page on every one of
those.  But of course I do get erroneous pages sometimes.


-----Original Message-----
From: Leonard Bocock [mailto:leonard.bocock AT NZ.UNISYS DOT COM]
Sent: Wednesday, February 10, 1999 10:20 PM
Subject: Re: How can I threshold Node_Down events ?


We use an event rule to catch all node down and node up events

A netview node down events rule calls  a shell script which pings the
device's
IP address ($NVA in the shell) 3 times to make sure the node down is
genuine,
(as at times timeouts cause erroneous node downs).  If the node down is
genuine,
we touch a file called $NVA in /var/tmp/netview and allow the event to
forward
(in this case to TEC)

A node up event calls a script which looks for /var/tmp/netview/$NVA.  If
the
file exists, it is deleted and the node event forwarded.

It may seem clandestine but it's the only way we've worked out of getting
rid of
erroneous node down events.  Also, the  nodes we monitor we' don't own the
routers so can get SNMP traps from them.

Rgds,





Mark van Kerkwyk <kerkwyk AT COMTECH.COM DOT AU> on 11/02/99 14:12:53

Please respond to Discussion of IBM NetView and POLYCENTER Manager on
NetView
      <NV-L AT UCSBVM.UCSB DOT EDU>

To:   NV-L AT UCSBVM.UCSB DOT EDU
cc:    (bcc: Leonard I. Bocock/NZ/Unisys)
Subject:  How can I threshold Node_Down events ?




Hi ,
     I must have surely missed something here , but I cannot work out how
to tell Netview not to send Node_Down events until a threshold (time) has
been met.
My customer wants Netview to only send a Node_Down alert when the node has
node responded to 2 or 3 ping attempts.
Since there is only on Node_Down event generated, I can't use a ruleset to
correlate the events together and seta threshold.

Anyone know of a simply way to do this ?

Mark :-)