Actually – not exactly:

This is a multi-instance object (processor) as there are often multiple processors per OS.

That alone will cause alerts from each processor instance – as he has shown.

However – it gets worse:  If you check the box to monitor “all instances” – then the workflow will load and inspect EACH instance.  If the FIRST CPU instance is over the threshold, you go red, if the next is not – back to green.  This monitor will “flip-flop” like crazy and not deliver the intended results.

So – multiple alerts = bad target.  Use Windows OS.

“All instances” = bad selection – will cause the monitor to flip-flop.

For a good overall rollup of %Processor Time on a server – use the built in monitor in the base OS, which targets Operating System, and only the _Total instance.

One more thing – targeting “processor” is targeting an abstract class – a rollup of ALL CPU’s.  It is better to target the individual OS that you want to monitor, because processor will include ALL OS, including client.

For REALLY detailed individual processor monitoring – you need to apply a multi-instance discovered target path, like so:

This is true for performance on any multiple instance object.  Logical disk is another classic example.

I talk a little about this here:

http://blogs.technet.com/kevinholman/archive/2007/12/14/how-do-i-collect-data-from-a-multi-instance-object-like-a-sql-db-instance.aspx

Now – all that said – in general – you DON’T want to monitor individual processors… because it is very normal for a single threaded application to use 100% of one CPU, will all the other cores are dormant.  On a 4 core box – you would alert in this condition, even though the overall CPU time was 25%.  You should only discover and monitor individual CPU’s in the case where you want to detect this condition.  Be prepared – it will be a LOT noisier than monitoring for overall CPU time.

When you selected the performance counter properties you clicked the checkbox to include all instances. Therefore, each instance of cpu utilization is examined individually. That will result in an alert for each cpu in a server that exceeds the threshold.

Uncheck include all instances and select the _Total  instance. That will give you what you are looking for. An alert when overall cpu utilization is high.

I have created a monitor for cpu utilization….i am getting many alerts for single server.

By configurting the monitor  for windows server cpu utilization I am getting the following alerts from a single server. 

How can I get a signle alert for  cpu utilization….

Many Thanks,

Narendra.

DISCLAIMER:
This email (including any attachments) is intended for the sole use of the intended recipient/s and may contain material that is CONFIDENTIAL AND PRIVATE COMPANY INFORMATION. Any review or reliance by others or copying or distribution or forwarding of any or all of the contents in this message is STRICTLY PROHIBITED. If you are not the intended recipient, please contact the sender by email and delete all copies; your cooperation in this regard is appreciated.

==============
Missed an email? Check out the list archive:
http://myitforum.com/cs2/blogs/momlist/

==============
Missed an email? Check out the list archive:
http://myitforum.com/cs2/blogs/momlist/

==============
Missed an email? Check out the list archive:
http://myitforum.com/cs2/blogs/momlist/