Metrics

• High reconcile latency

  Individual reconcile calls are taking longer than the configured p99 threshold, indicating slow source fetches, heavy evaluation, or a sluggish storage backend

• Metrics

  The controller-runtime Prometheus endpoint, the custom jaas_ metric family, scraping with a ServiceMonitor or a plain scrape config, querying with PromQL, and the Helm chart keys that drive it.

• Observability

  How to watch JaaS in production — structured logs, OTLP traces, Prometheus metrics, and the shipped alert catalog with Kubernetes Events and Flux notification routing.

• Workqueue saturation

  The reconciler's workqueue depth exceeds the threshold because the operator is dequeuing reconciles slower than the apiserver enqueues them