I am curious to know about CPU Check parameters for configuring the alerts. we have installed uptime server and installed agents on Linux servers. we are getting CPU check CRIT alert from up-time but when we check in the Linux server the CPU utilization is normal. we can see "SAR" report most of all time the CPU idle state is around 88% and don't see any spikes.
we ran "sar -u" to check all CPU statistics and used "sar -P ALL" to check individual CPU statistics and average CPU idle time is 80 to 90%. so , we are not sure how uptime monitoring sent CPU utilization is high and what condition uptime will trigger the "CRIT" alert.
can you please help us to reconfigure monitoring parameters to know really when CPU span high utilization.