In Rancher v2.2 - v2.4 monitoring can be enabled at both the cluster and project level, this article explains the differences between the two.
- A Rancher v2.2 - v2.4 instance
Per the documentation on monitoring in Rancher v2.2 - v2.4, "cluster monitoring allows you to view the health of your Kubernetes cluster. Prometheus collects metrics from the cluster components below, which you can view in graphs and charts". This includes metrics for control plane components, the etcd database, and cluster nodes.
With the metrics provided by cluster monitoring you can then configure cluster alerts, for example for high node CPU or memory usage, or etcd member unavailability, etc.
Project monitoring can only be enabled with cluster monitoring enabled first.
Per the documentation, "project monitoring allows you to view the state of pods running in a given project. Prometheus collects metrics from the project’s deployed HTTP and TCP/UDP workloads". This enables you to configure scraping of custom metrics endpoints on workloads you deploy into the cluster.
With project monitoring you can then configure project alerts based on the custom metrics scraped, or workloads metrics, such as a pod's memory usage exceeding its quota.