Cluster monitoring

The following tools are available for monitoring the status of a Kubernetes cluster:

• Tools built into the Kubernetes Dashboard.

  They provide basic monitoring capabilities, allowing you to view information about Kubernetes resources.

• Monitoring addon.

  It provides advanced cluster monitoring capabilities, allowing you to view metrics for multiple Kubernetes resources, down to individual pods. It is also possible to configure alerts when specified thresholds for metrics are exceeded. For example, you can monitor if the cluster resources are not available or if the computing power of the worker nodes is insufficient.

  To find out if the monitoring addon is installed in the cluster, see the list of installed addons.

1. Connect to the cluster using Kubernetes Dashboard.

2. Get data about the cluster resources:

   1. Select the desired namespace from the drop-down list at the top of the Kubernetes Dashboard interface. The default namespace selected is default.
   2. From the side menu, select the Kubernetes resource of interest.

1. Make sure that the monitoring addon (kube-prometheus-stack) is installed in a cluster.

2. Make sure that you can connect to the cluster using kubectl.

3. Access the Grafana web interface:

   1. In a separate terminal session, run the command:

      Copy
      kubectl -n prometheus-monitoring port-forward service/kube-prometheus-stack-grafana 8001:80

   2. From the output of the command identify the port opened by kubectl to access Grafana.

      The port number is specified before the → symbol. For example, in this output the port 6637 is used to connect:

      Copy
      Forwarding from 127.0.0.1:6637 -> 3000Forwarding from [::1]:6637 -> 3000

   3. Open in your browser the URL for accessing the Grafana web interface:

      Copy
      http://localhost:<port>/

      The Grafana login page will appear.

4. Log in to Grafana. Depending on the parameters with which the addon was installed, use:

   • Login admin and temporary password. The password will need to be changed after the first login.
   • Login admin and a permanent password from the Kubernetes secret.

Now you can work with Grafana. For example, it is possible to create and view dashboards. Preconfigured dashboards with information about various Kubernetes resources are available in the Dashboards → Browse side menu.

Read more about working with Grafana in official documentation.