Sorry, for the second append, but apparently some of my capital lettelrs
weren't typed correctly
Here's a corrected version.
Well, you can try building a ruleset and see if it will give you what you
want, but I don't see exactly how to give you all of it.
Start with an Event Stream which has BLOCK as the default action. You can
use a Node Down connected to a Reset-On-Match, with a time out of 5
minutes, connected to a Forward and the #2 side of the Reset is your Node
Up, matching on attribute 2 (the hostname). Then only those Node Down's
which don't get matched within 5 minutes will go to the window. But the
Node Ups will not go at all unless you explicitly connect them to the
Forward. But then you will get all Node Ups regardless of whether they
were used in a match or not. There is no getting around that so far as I
can see.
I don't understand what you are asking about attributes in the Events
window. Almost everything you see there is customizable. You use xnmtrap
to configure the Log Message in trapd.conf to change what you see. The man
page on trapd.conf will explain what all the variables such as $A, $C, and
$1 - $50 mean and how to use them to display whatever parts of the trap you
want. This is all explained in the Administrator's Guide, too.
If you are going to stay at Version 4, I recommend highly that you obtain
the redbook, "Examples of Using NetView for AIX Version 4, SG24-4515-00,
from the IBM website, http://www.redbooks.ibm.com. It has more examples of
rulesets and a bunch more things you might be interested in doing to
exploit the product more fully.
James Shanks
Team Leader, Level 3 Support
Tivoli NetView for UNIX and NT
Szymon Trocha <Szymon.Trocha@man.poznan.pl>@tkg.com on 06/15/2001 08:08:23
AM
Please respond to IBM NetView Discussion <nv-l@tkg.com>
Sent by: owner-nv-l@tkg.com
To: IBM NetView Discussion <nv-l@tkg.com>
cc:
Subject: Re: [NV-L] Node Down/Node Up events
Thx James. Of course it was NetView V4R1 for AIX (according to "Help"). I
know it's rather old one. I did what you advised and it worked.
Now I try to correlate NodeDown/Up events. I know that this topic has been
raised many times on this list but I didn't find a solution for my problem.
I'd like to see NodeDown event and alarm generated provided more than 5 min
has passed since NV discovered the node was down. If there will be NodeUp
event in some time (separated more than 5 min) I want to see it too. In the
other way if there is an event from the same node separated by less then 5
min I don't want to see them both and have generated any alarms.
And the last question: Is there any possibility for changing the way the
data is displayed in the Event window. I mean - can I choose a range of
attributes displayed?
Regards and thanks for any advice.
Szymon
At 11:05 AM 6/6/01 -0400, you wrote:
>
>Are you serious?
>You bring up the NetView GUI and click on "Help" and then "On Version" (or
>whatever) and it tells you that you have Version 4.0?
>I thought that the first level of V4 was 4.1.1, but I could be wrong
>because that was the first version I worked with when I joined the team in
>1996.
>If your assessment is accurate, then your code has been out of support for
>at least 5 years. NetView Version 4.1.2, the last level of NetView
Version
>4, went of out of support the end of November 1998.
>
>But to answer your first question, you should be setting the ping schedule
>by using xnmsnmpconf. Just type that in or do it from the GUI using
>Options --> SNMP Configuration to bring it up. What you set here is how
>frequently netmon pings the devices. This is done a regular, cyclical
>basis and is not done more frequently after a "Node Down" unless things
>have changed dramatically since the level of code you have. t is hard to
>tell.
>
>More than likely the reason that you get frequent "Node Down" - "Node Up"
>event for the same node is that the time-outs for your polling intervals
>are too short. If you adjust these then you may not have to build filters
>or rulesets to alter your event window. But you can alter what you see in
>the window with a filter based on Thresholds. I'd leave that until you
try
>adjusting the polling intervals first.
>
>
>James Shanks
>Team Leader, Level 3 Support
> Tivoli NetView for UNIX and NT
>
>
>
>Szymon Trocha <Szymon.Trocha@man.poznan.pl>@tkg.com on 06/06/2001 10:26:16
>AM
>
>Please respond to IBM NetView Discussion <nv-l@tkg.com>
>
>Sent by: owner-nv-l@tkg.com
>
>
>To: nv-l@tkg.com
>cc:
>Subject: [NV-L] Node Down/Node Up events
>
>
>
>Hi,
>
>I have two simple questions:
>- How to set the number of pings after which NV generates NodeDown event?
>- How can I avoid receiving consecutive Node Down and Node Up events in my
>main "Events" window within a small amount of time? I would like to have
>generated one Node Down event and then one Node Up if there was for
>instance 5 consecutive down/up events from the same node within the last 5
>minutes. Should I use filters?
>
>I have NV4.0
>
>Best regards,
________________________________________________________________
Szymon Trocha
POZNAN SUPERCOMPUTING & NETWORKING CENTER
Tel. (+48 61) 8582022
http://noc.man.poznan.pl
http://www.man.poznan.pl/~konserwa
_________________________________________
_________________________________________________________________________
NV-L List information and Archives: http://www.tkg.com/nv-l
_________________________________________________________________________
NV-L List information and Archives: http://www.tkg.com/nv-l
|