The Cluster Agent monitors the health of Kubernetes and OpenShift clusters and is supported on all major Kubernetes distributions. The Cluster Agent collects metrics and metadata for the entire cluster, including every node and namespace down to the container level. When you instrument applications with APM Agents, you can use the Cluster Agent to view both Kubernetes and APM metrics for those pods (if both the Cluster Agent and the APM Agents are reporting data to the same account on a Controller).

You can use the Cluster Agent to:

• Gain visibility into key Kubernetes metrics and events, and detect uptime and availability issues
• Track resource usage of pods relative to the defined requests and limits
• Diagnose issues that may prevent uptime or scalability issues, such as: 
  • Pod failures and restarts
  • Node starvation
  • Pod eviction threats and pod quota violations
  • Image and storage failures
  • Pending or stuck pods

  • Bad endpoints (detects broken links between pods and application components)

  • Service endpoints in a failed state

  • Missing dependencies (Services, ConfigMaps, and Secrets)