possible metrics: - cluster wide - start time (cold start) - resource count / running count / etc. - controller resources (CPU, memory) - resource reconciliation time https://github.com/streamnative/function-mesh/pull/689 - resource - start time - component status (sts, service, hpa, vpa: ready, etc) - replicas references: - https://sdk.operatorframework.io/docs/best-practices/observability-best-practices/ - https://github.com/kubernetes/kube-state-metrics/blob/main/internal/store/pod.go - https://sdk.operatorframework.io/docs/overview/operator-capabilities/#monitoring