I tried the suggestions made by Kevin and Dirk.  It greatly reduced the
false reports of problems on the server, but I'm not sure that the counter
is being reset correctly.

The test is as follows:
General: 192.168.7.19, connect via LAN, Keep connection alive
Check: NT perfmon, Processor,% Processor Time,0, > 99, timeout 30, second
knock
Alert: SMTP when 10 times down and then every 10 cycles
Schedule: every minute

I've gradually increased the number of down events before the alert is
supposed to trigger.  When I watch the CPU utilization with taskmgr, it is
almost always at 0%.  It will spike - perhaps every few minutes - but it
quickly drops back to 0%.

It's very unlikely that it actually hits 100% on each of 10 successive
checks.  Nevertheless, I've gotten several alerts today.  Perhaps the very
act of checking cpu utilization can cause it to hit 100% - without lasting
long enough to show up on taskmgr?

Is there some better way to do this?
Is there perhaps a counter that I can check for the average CPU utilization
over the last 5 minutes?

Thanks,
Gregg Nicholas

> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
> Behalf Of Kevin Stone
> Sent: Tuesday, June 24, 2003 17:12
> To: [EMAIL PROTECTED]
> Subject: RE: [SA-list] perfmon cpu utilization check
>
>
> When you set up the alert you can set it to alert after x number or
> checks.  Then at the very bottom of the box off to the right is the
> checkbox "and when back up too".  I do the same thing for CPU alerts.
>
> -Kevin

To unsubscribe from a list, send a mail message to [EMAIL PROTECTED]
With the following in the body of the message:
   unsubscribe SAlive

Reply via email to