Thorsten,
You say you are getting hardware traps from IBM, Compaq (HP) and Dell, and
probably also HP and SUN. So let's say you have 5 Enterprises sending you
hardware traps.
If you SIMPLY want to send -- specific --- traps from those enterprises too
TEC (no correlation), there is a very simple and effective way to do that
in your ruleset. In the ruleset that is called by your
/usr/OV/conf/tecint.conf you would have the default action of 'BLOCK' on
your event pipeline (also known as a pizza box!). You will then
connect that to a trap definition node and fill in the IBM Enterprise
number. You now see a window with all the IBM traps defined for that
Enterprise. You can select MULTIPLE traps in this window, all of them that
you want to forward to TEC. You then connect that node to the 'Event
Forward' node. You repeat this process for the other Enterprises of
traps you want to forward. The end result should be a single Trap
Definition block for each Enterprise you are forwarding.
Make sure that your Baroc in TEC can handle them all.
Another hint. For events that you are correlating in other rulesets. If
you want the outcome of correlation to be forwarded to TEC, I would
recommend that your correlation (between 2 events), generate a 3rd event.
The 3rd event is what actually gets forwarded to TEC . You may have to
define a 'correlation' enterprise in NV and TEC. This gives you complete
control of what goes to TEC.
Stephen Hochstetler shochste@us.ibm.com
International Technical Support Organization at IBM
Office - 512-838-6198 (t/l 678) FAX - 512-838-6931
http://www.redbooks.ibm.com
|---------+------------------------------>
| | "Mildeberger, |
| | Thorsten" |
| | <thorsten.mildeberg|
| | er@eds.com> |
| | Sent by: |
| | owner-nv-l@lists.us|
| | .ibm.com |
| | |
| | |
| | 11/25/2003 10:12 AM|
| | Please respond to |
| | nv-l |
| | |
|---------+------------------------------>
>-------------------------------------------------------------------------------------------------------------------------------|
|
|
| To: nv-l@lists.us.ibm.com
|
| cc:
|
| Subject: [nv-l] [nv] special ruleset for hardware traps
|
|
|
>-------------------------------------------------------------------------------------------------------------------------------|
hi all,
we are in the phase of implementing a hardware monitoring and therefore
going to receive approx. 300 different traps from hardware platforms like
IBM, compaq and dell. These traps get sent to a "virtual interface" that
can be linked to either a backup netview server or a primary netview
server.
As there is no way to define 300 traps hardcoded inside of a ruleset, we
are looking to go for another way.
In one ruleset, which is defined in ESE.automation we would like to "mark"
incoming traps by using some appropriate "ruleset nodes" (MIB variable,
database field, set state..). In the final ruleset that is defined in
/urs/OV/conf/tecint.conf, we only want to let these traps pass the ruleset,
which got marked with "compaq, dell or IBM" before accordingly. This check
should be done again by using appropriate "ruleset nodes".
Can somebody please shine some light on me and give me a hint to get it
done using some of these "ruleset nodes" the right way? I already made some
testing, but didn´t succeed so far.
We are already using the "set state" functionality in one ruleset for the
correlation of interface up/down and node up/down traps.
Any help is greatly appreciated. Many thanks.
best regards,
> Thorsten Mildeberger
>
> EMS Solutions - SMC Tools & Automation
> GOSD - Central Region
> EDS Deutschland GmbH
> Eisenstr. 43
> 65428 Rüsselsheim
> Tel.: +49 (0) 6142 80-3706
> Fax.: +49 (0) 6142 80-3030
> mailto:thorsten.mildeberger@eds.com
>
|