QOS_AWS_RDS_CPU_UTILIZATION collected by AWS Probe displayed as unexpectedly high value
Article ID: 279939
Updated On:
Products
Issue/Introduction
When monitoring the RDS service with the AWS probe, the QOS_AWS_RDS_CPU_UTILIZATION metric is collected and displayed with a very high value (anything greater than 100% is unexpected for utilization). This is causing a false alarm to be generated.
Environment
- UIM 20.4+
- aws probe 5.44+
Cause
Selecting the Statistics as sum for the RDS Configrations (as shown below), returns the sum of the data points collected for the metrics. This is working as intended.
Resolution
Choosing the Sum option from the drop down doesn't make sense for the CPU_UTILIZATION metric values. Selecting (Statistics* as Average), provides the correct result as the average of CPU_UTILIZATION metric values would be most appropriate.
Additional Information
The different possible values for the Statistics configuration item is discussed in the documentation. The documentation also notes that the QoS graphs are affected as described in this article by changing the Statistics configuration item.
Feedback
