I created a Incident with the 'cpu total time' and set the threshold to 90% severe and 80% warning, when this incident was triggerd, the cpu usage rate is only about 25%.
So I am wondering, how to understand the 'cpu total time', and how to use this measure correctly?
Thanks a lot!
Solved! Go to Solution.
CPU Total Time is the sum of all time spent in the CPU(system+User+IO+Other) but excluding Idle Time.
Go through the below link to understand clearly on CPU Total Time.
90% severe and 80% warning should be set to Upper Severe and Upper Warning in the measure. Is this correctly configured in that column?Also make sure that Unit is selected as "Percent" in details tab of measure.
Hope this may help!!
Adding to Soorya's answer, you can also use the Infrastructure settings to control these incidents as we already trigger host unhealthy incidents out-of-the-box based on the thresholds set in infrastructure.
You access these settings by going from the thick client > Settings > Dynatrace Servers... > Infrastructure. Here you can create entries for your hosts (if not already auto-detected), add labels and group hosts together. Thresholds can then be set for individual hosts or whole groups.