Create Monitoring Dashboard per Hosted Cluster

The HyperShift operator can create/destroy a separate monitoring dashboard in the console of the management cluster for each HostedCluster that it manages. This functionality can be optionally enabled on installation of the HyperShift operator.

Enable Monitoring Dashboards

To enable monitoring dashboards, use the --monitoring-dashboards flag when running hypershift install. Alternatively, to enable monitoring dashboards in an existing installation, set the MONITORING_DASHBOARDS environment variable to 1 on the hypershift operator deployment:

oc set env deployment/operator -n hypershift MONITORING_DASHBOARDS=1

Dashboards

When monitoring dashboards are enabled, the HyperShift operator creates a configmap named cp-[NAMESPACE]-[NAME] in the openshift-config-managed namespace (where NAMESPACE is the namespace of the HostedCluster and NAME is the name of the HostedCluster) for each HostedCluster that the operator manages. This results in a new dashboard getting added under Observe -> Dashboards in the administrative console of the management cluster. When a HostedCluster is deleted, its corresponding dashboard is also deleted.

Customize Monitoring Dashboards

To generate per-cluster dashboards, the HyperShift operator uses a template stored in a ConfigMap named monitoring-dashboard-template in the operator namespace (hypershift). This template contains a set of grafana panels that contain the various metrics that should go on the dashboard. Edit the content of this ConfigMap to customize the dashboards. When a particular HostedCluster's dashboard is generated, the following strings will be replaced with values that correspond to the specific HostedCluster: