James:
Thank you very much, your help was direct helpful and clear.
B. Regards
Alejandro Gabay
James Shanks
<jshanks@us.ibm.c To: nv-l@lists.us.ibm.com
om> cc:
Sent by: Subject: Re: [nv-l] Automatic
clearing events and map
owner-nv-l@lists. status changes
us.ibm.com
Header:
Internal Use
Only
15/04/2004 17:56
Please respond to
nv-l
The short answer to your questions is that you cannot duplicate the function
you are used to in NetView without some considerable effort.
Either you will have to learn to do without it or you can implement something
of your own, but even then, that may only approximate but not duplicate what
you are used to. It is difficult to judge at this point. But the answer to
both your questions involves using rulesets and customizing the event displays,
not the trap log.
1. There is only one trapd.log. You control what is logged there by altering
the log message in trapd.conf using the xnmtrap editor. But trap severity is
not a field that is logged, unless you add it yourself in the log message.
The trap severity is used rather to alter the color of the event in the user's
event display(s), which is where users typically watch events, in the events
window, which is a real-time, in-memory display and not a log. You can affect
a user's displays by using rulesets. It would be easy to write a ruleset which
would resolve an event, that is, remove it from the user's display, upon
receipt of another. But you must familiarize your self with what is written
about the ruleset editor in the Administrator's Guide and then play with it a
little. You can start the editor outside of the GUI using the command
/usr/OV/bin/nvrsEdit . Open the sampcorrNuNd.rs file for an example of how a
resolve works. You'd need only change the Trap Settings to !
be the
events you want, and the Pass-On-Match to indicate the matching attributes
those have and the hold time.
2. You can use the 58916871 status trap but there are several caveats when
doing so. First the event window must be open on a display having been started
with the read-write map, or the update will not occur. This is because the API
is in the event window GUI, a process called "nvevents", and it requires access
to the read-write map to update the status. And these objects you affect the
status on should be ones which you add to NetView yourself because, if netmon
is also updating them, then whenever the GUI is recycled, the status that
netmon has for the object, which is stored in the database will be used when
the map is reopened. Updating the status of netmon-controlled objects is much
more difficult. If that's what you want to do, then you had better take up
that issue in a separate note.
The rest of what you say you want for this second case is rather murky to me
because you are still using some terminology from your other system. But let
me see whether I can clarify a few points. You can set whatever severity you
want for a trap initially in trapd.conf. That's how it will display. And you
may override that severity with a ruleset which changes that severity for the
user viewing it . And your ruleset could further log that trap with it's new
severity to a new log file of your own devising. But the trapd.log does not
log event severities as a field in the log record. They are only used to alter
the color of the event in the display. So your new log could record a
severity, but the default one would not have it. I hope that it clear.
All this customization is just with NetView itself as a stand-alone product.
If you have plans to use the Tivoli Event Console as a replacement for the
NetView event display, then the correlation you describe can be achieved
within the TEC display and TEC log using TEC rules, but it is beyond the scope
of my reply to explain to you how to do that. The learning curve for
customizing TEC is rather steep and user's typically go to class or get help
from IBM Services to do anything out of the ordinary. You might want to ask
your IBM marketing rep about the availability of classes on NetView and TEC in
your area.
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Tivoli Software / IBM Software Group
!
alejandro.gabay@reuters.com
!
Sent by: owner-nv-l@lists.us.ibm.com
!
!
To
nv-l@lists.us.ibm.com
!
04/15/2004 04:33 PM
!
cc
!
!
Subject
Please respond to
[nv-l] Automatic
clearing events and map status changes
!
nv-l
!
!
!
!
!
!
!
Hi.
I have a Netview 7.1.3 running Solaris 8 on a SUN box.
As we are migrating from a different SNMP manager to Netview, we were
accustomed to have a current log and a history log. Some events were configured
to auto-clear others (this means to remove them from the current and send them
straight to the history). Do you know how can we implement this in Tivoli?.
Also, we have and external application that should set some objects status in
the Netview map. I am planning to use the Netview specific trap 58916871 to
change the object color?. I would like to see the traps in the netview event
history with a priority linked to them (per instance an "Up" event should be a
clear event, an "User2" event should be a warning one). How can I change the
trap severity depending on the trap content?.
B. Regards
Alejandro Gabay
Technical Center Supervisor, Argentina
Reuters Ltd
-----------------------------------------------------------------
Visit our Internet site at http://www.reuters.com
Get closer to the financial markets with Reuters Messaging - for more
information and to register, visit http://www.reuters.com/messaging
Any views expressed in this message are those of the individual
sender, except where the sender specifically states them to be
the views of Reuters Ltd.
-----------------------------------------------------------------
Visit our Internet site at http://www.reuters.com
Get closer to the financial markets with Reuters Messaging - for more
information and to register, visit http://www.reuters.com/messaging
Any views expressed in this message are those of the individual
sender, except where the sender specifically states them to be
the views of Reuters Ltd.
|