When does Dynatrace consider an RDS instance to be undergoing a restart sequence?

In Dynatrace, an RDS instance is identified as undergoing a restart sequence when it experiences more than two resets within a 20-minute timeframe. This metric is critical as it helps users detect instability in their database instances early on. Frequent resets can indicate underlying issues within the database or the environment that could impact performance and availability. Monitoring for this specific threshold allows Dynatrace to provide reliable insights, enabling users to address potential problems before they escalate.

The chosen threshold is designed to balance sensitivity and specificity, ensuring that only significant instances of restarts are flagged, thus reducing noise from benign or inconsequential activity. This precise definition helps in maintaining the reliability of monitoring alerts and reports, enabling effective response and troubleshooting actions.