Each monitor location is modelled, and when a monitor runs for an unusual amount of time, at a particular time, an anomaly is recorded and highlighted on the Monitor duration chart.
Create a machine learning job to detect anomalous monitor duration rates automatically.
- To access this page, go to Observability > Uptime, and then click a monitor to view its the details.
In the Monitor duration panel, click Enable anomaly detection.
- You are prompted to create a response duration anomaly alert for the machine learning job which will carry out the analysis, and you can configure which severity level to create the alert for.
When an anomaly is detected, the duration is displayed on the Monitor duration chart, along with the duration times. The colors represent the criticality of the anomaly: red (critical) and yellow (minor).