Setting up a monitoring stack in Kubernetes can be overwhelming if you're doing it for the first time. When I started working with Azure Kubernetes Service (AKS), I realized I needed better visibility into pod and node metrics — that’s when I turned to Prometheus and Grafana.
In this post, I’ll walk you through the exact steps I took to get Prometheus and Grafana running on AKS, along with some lessons I learned along the way.
## 🔧 Why Prometheus + Grafana?
Prometheus scrapes and stores time-series data from Kubernetes.
Grafana gives you beautiful, customizable dashboards.
Together, they provide powerful insights into your cluster’s performance.
## 🛠️ Prerequisites
- Azure CLI installed and authenticated
kubectlconfigured to your AKS cluster- Helm 3 installed
- A working AKS cluster
🚀 Step-by-Step Setup
1. Add the Helm Repos
bash
helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
helm repo add grafana https://grafana.github.io/helm-charts
helm repo update
**Install Prometheus:**
helm install prometheus prometheus-community/prometheus \
--namespace monitoring --create-namespace
Install Grafana:
helm install grafana grafana/grafana \
--namespace monitoring
Access the Grafana UI:
You can port-forward and access Grafana at http://localhost:3000:
kubectl port-forward svc/grafana 3000:80 -n monitoring
**Login using:**
Username: admin
Password: (from step 4)
**📊 Importing Dashboards**
Once inside Grafana, go to "Dashboards > Import" and use IDs from Grafana.com like:
Kubernetes Cluster Monitoring: 315
Node Exporter: 1860
**💡 Lessons Learned**
Always install Prometheus before Grafana if you want dashboards to auto-discover data sources.
Use persistent volumes for Prometheus and Grafana in production.
Set up alerts for critical metrics like pod restarts or high CPU usage.
I’m planning to extend this by:
Setting up AlertManager
Integrating with Slack for alerts
Enabling SSL and Ingress for Grafana
Stay tuned! 👀