1. Run kubectl create namespace monitoring to create a monitoring namespace.

2. Run kubectl create -f ./ -R to get kube-state-metrics, Grafana, and Prometheus resources in place.

3. Visit http://localhost:30000 to view Prometheus.

4. Click on Graph. Select a metric from the drop-down (next to the Execute button) and then click Execute. You can switch between the Graph view and Console view.

5. Visit http://localhost:32000 to view the Grafana dashboard.

6. Login with admin and admin for the username and password. You can reset the password on the next screen or click Skip.

7. On the Welcome to Grafana page click New dashboard.

8. Click the drop-down item named Home in the upper-left corner of the screen.

9. Select Import dashboard from the available options.

10. Enter an ID value of 10000 into the Grafana.com Dashboard textbox and click Load. This will load the Cluster Monitoring for Kubernetes dashboard.

    Note: You can find a list of pre-configured dashboards at https://grafana.com/grafana/dashboards. Each one will have an id you can copy into the Grafana Import dashboard screen.

11. On the next screen you'll see a Select a Prometheus data source drop-down. Select prometheus from the list and click Import. Your dashboard will now load.

12. Click on the Cluster Monitoring for Kubernetes drop-down in the uppper-left corner of the screen. Go back through the steps to import a dashboard but this time enter an ID of 8588. This will load the Kubernetes Deployment Statefulset Daemonset metrics dashboard.

13. If you click on the dashboard name drop-down in the upper-left corner of the screen you'll see both of your dashboards listed and can now switch between them.

14. Import the dashboard with an ID of 10694. This dashboard is called Container Statistics.

15. From here you can load other dashboards or create your own!

NOTE: The Metrics Server YAML has been modified to enable it to run in more simple Kubernetes scenarios like Docker Desktop. Details about the modifications made can be found at https://blog.codewithdan.com/enabling-metrics-server-for-kubernetes-on-docker-desktop. You'll want to remove the --kubelet-insecure-tls arg in metrics-server/kubernetes/metrics-server-deployment.yaml for more real-world scenarios.

To remove all of the monitoring resources run kubectl delete -f ./ -R.

Get more details on Prometheus at https://prometheus.io/docs/prometheus/latest/getting_started and https://devopscube.com/setup-prometheus-monitoring-on-kubernetes.

A list of Pod Metrics provided by kube-state-metrics can be found at https://github.com/kubernetes/kube-state-metrics/blob/master/docs/pod-metrics.md.

Another option for getting Prometheus going using only Docker Compose and Docker Swarm can be found at:

https://github.com/vegasbrianc/prometheus

The yaml files used here were originally created by Bibin Wilson (thanks for the great work Bibin!) and documented at:

• https://devopscube.com/setup-prometheus-monitoring-on-kubernetes
• https://devopscube.com/setup-kube-state-metrics
• https://devopscube.com/setup-grafana-kubernetes/