Cluster Monitoring Operator

The Cluster Monitoring Operator manages and updates the Prometheus-based cluster monitoring stack deployed on top of OpenShift.

It contains the following components:

• Prometheus Operator
• Prometheus
• Alertmanager cluster for cluster and application level alerting
• kube-state-metrics
• node_exporter
• prometheus-adapter
• kubernetes-metrics-server only on TechPreviewNoUpgrade cluster

The deployed Prometheus instance (prometheus-k8s) is responsible for monitoring and alerting on cluster and OpenShift components; it should not be extended to monitor user applications. Users interested in leveraging Prometheus for application monitoring on OpenShift should consider enabling User Workload Monitoring to easily setup new Prometheus instances to monitor and alert on their applications.

Alertmanager is a cluster-global component for handling alerts generated by all Prometheus instances deployed in that cluster.

Adding new metrics to be sent via telemetry

To add new metrics to be sent via telemetry, simply add a selector that matches the time-series to be sent in manifests/0000_50_cluster-monitoring-operator_04-config.yaml.

Documentation on the data sent can be found in the data collection documentation.

Contributing

Please refer to the CONTRIBUTING.md document for information.

Release

Release checklist is available when creating new "Release Checklist" issue.