Here’s a step-by-step tutorial for deploying Prometheus on Kubernetes using Rancher Desktop on macOS, with a focus on collecting Kubernetes metrics.

Prerequisites

1. Rancher Desktop: Make sure Rancher Desktop is installed and configured to use Kubernetes.
2. kubectl CLI: You need kubectl installed and configured to interact with your Rancher Desktop cluster.
3. Helm CLI: You also need Helm installed, as we’ll use it to install Prometheus.

Set up Kubernetes in Rancher Desktop

1. Start Rancher Desktop and ensure Kubernetes is running.
2. Verify Cluster Access by running:

   kubectl get nodes
   

   You should see the nodes in your Rancher Desktop Kubernetes cluster.

Add the Prometheus Helm Repository

Prometheus can be installed easily using Helm, which allows you to customize configurations.

1. Add the Prometheus community Helm chart repository:

   helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
   

2. Update your Helm repositories:

   helm repo update
   

Install Prometheus Using Helm

To deploy Prometheus on your cluster, run the following Helm command. This installation will include both Prometheus and the necessary tools for scraping Kubernetes metrics, such as the kube-state-metrics and node-exporter.

helm install prometheus prometheus-community/kube-prometheus-stack --namespace monitoring --create-namespace

This command does the following:

• Deploys the kube-prometheus-stack, which includes Prometheus, Grafana, and other components for Kubernetes monitoring.
• Creates a namespace called monitoring (if it doesn’t exist).

Notes

• You can customize the installation by adding flags or by creating a custom values file (more on this in next step).

Customize Prometheus Configuration (Optional)

If you want to customize your Prometheus configuration, you can create a values.yaml file. Below is an example of common customizations, such as adjusting retention period or configuring Prometheus to scrape custom metrics.

1. Create a file named values.yaml:

   prometheus:
     prometheusSpec:
       retention: 7d                 # Retain metrics for 7 days
       serviceMonitorSelector: {}     # Scrape all ServiceMonitors in the namespace
       podMonitorSelector: {}         # Scrape all PodMonitors in the namespace
   
   grafana:
     enabled: true
     adminPassword: 'admin'          # Set a password for Grafana
   
   kube-state-metrics:
     enabled: true                   # Enable Kube State Metrics for Kubernetes data
   

2. Install or upgrade the kube-prometheus-stack with your custom values:

   helm upgrade --install prometheus prometheus-community/kube-prometheus-stack --namespace monitoring -f values.yaml
   

This command applies the settings from values.yaml during the Helm deployment.

Verify the Prometheus Installation

1. Check the Status of Prometheus:

   kubectl get pods -n monitoring
   

   You should see multiple pods, including Prometheus, Grafana, and exporters like kube-state-metrics and node-exporter.

2. Access Prometheus Dashboard (Port-forwarding):

   kubectl port-forward svc/prometheus-kube-prometheus-prometheus -n monitoring 9090:9090
   

   Now, visit http://localhost:9090 on your browser to access the Prometheus dashboard.

Set up Grafana to View Kubernetes Metrics (Optional)

Grafana is included in the kube-prometheus-stack, making it easy to visualize the metrics collected by Prometheus.

1. Port-forward to access Grafana:

   kubectl port-forward svc/prometheus-grafana -n monitoring 3000:80
   

   Access Grafana at http://localhost:3000.

2. Log in using the default credentials:

   • Username: admin
   • Password: admin (or whatever you set in values.yaml)

3. Explore Dashboards:

   • In Grafana, go to Dashboards > Manage. You should see several pre-configured dashboards for Kubernetes and Prometheus metrics.

Verify Prometheus is Collecting Kubernetes Metrics

1. In the Prometheus UI (accessed at localhost:9090), go to Status > Targets.
2. You should see several targets, including:
   • kubelet endpoints (for node metrics)
   • kube-state-metrics (for cluster state data)
   • node-exporter (for node metrics)

Each target should show an UP status, indicating that Prometheus is successfully scraping metrics.

(Optional) Expose Prometheus and Grafana Services

If you want to access Prometheus or Grafana without using port-forwarding, you can expose these services with a LoadBalancer or Ingress.

Example with Ingress:

1. Apply an Ingress resource to expose the Grafana service:

   apiVersion: networking.k8s.io/v1
   kind: Ingress
   metadata:
     name: grafana-ingress
     namespace: monitoring
   spec:
     rules:
     - host: grafana.local
       http:
         paths:
         - path: /
           pathType: Prefix
           backend:
             service:
               name: prometheus-grafana
               port:
                 number: 80
   

2. Update your /etc/hosts file to point grafana.local to your cluster’s IP or localhost, depending on your setup.

3. Access Grafana at http://grafana.local.

Conclusion

You’ve successfully deployed Prometheus on Kubernetes using Rancher Desktop on macOS! This setup includes:

• Prometheus for collecting Kubernetes metrics.
• Grafana for visualizing metrics.
• Various exporters (like node-exporter and kube-state-metrics) for enhanced monitoring.