Skip to main content

Metrics and Monitoring

Monitoring the cluster via Prometheus​

Most components of STRM Privacy expose Prometheus metrics and we've included the Prometheus Cluster Operator Helm chart. In order to define scrape targets to Prometheus, we deploy Service Monitors that expose the metrics endpoints in the Event Gateway.

The kubectl commands below assume a default namespace of strmprivacy, which you can set via kubens strmprivacy

Prometheus server​

Access the Prometheus server either via telepresence or via a port-forward

kubectl port-forward service/strmprivacy-kube-prometheu-prometheus 9090:9090&

You should see the Event Gateway in the service discovery

  • serviceMonitor/strmprivacy/event-gateway-envoy-service-monitor/0
  • serviceMonitor/strmprivacy/event-gateway-service-monitor/0

and the targets

http://localhost:9090/targets

If you have the simulator running, you could have somewhat interesting timing data:

latency-timing

Grafana​

The CCD quickstart also includes a Grafana instance, that we can port-forward

kubectl port-forward service/strmprivacy-grafana 4000:80&

Log in to Grafana with username admin and password prom-operator (the value of the secret named strmprivacy-grafana).

I've added two panels for the very simple dashboard we show here:

grafana