Mastering CPU Throttling Alert Customization in Dynatrace for Kubernetes Environments

This episode focuses on customizing CPU throttling alerts in Dynatrace for Kubernetes environments to reduce alert fatigue and improve the actionability of notifications. It details how to tune built-in alerts, emphasizing their hierarchical application at tenant, cluster, or namespace levels, but noting their limitation at the workload level. The document extensively explains creating custom metric events with both static thresholds for predictable workloads and auto-adaptive thresholds, powered by Davis AI, for dynamic environments to learn normal behavior and reduce false positives. Finally, it describes using Alerting Profiles and Management Zones to filter and route problem notifications based on severity, scope, and custom event properties, ensuring that alerts reach the appropriate teams.