monitoring-stack

monitoring-stack

A lightweight, self-contained Prometheus + Grafana monitoring stack for Kubernetes — demonstrating hands-on observability chart design (Kubernetes-native Prometheus scrape config, auto-provisioned Grafana datasource, persistent TSDB storage).

What's Included

• Prometheus Deployment with a Kubernetes pod-discovery scrape config and persistent storage
• Grafana Deployment auto-provisioned with Prometheus as its default datasource
• Admin credentials pulled from a Kubernetes Secret (never hardcoded in values)

Before You Install

Create the Grafana admin secret first:

kubectl create secret generic grafana-admin-credentials \
  --from-literal=admin-password='<your-strong-password>' \
  -n monitoring

Install

helm install monitoring ./monitoring-stack -n monitoring --create-namespace

Access Grafana

kubectl port-forward svc/monitoring-grafana 3000:3000 -n monitoring
# Visit http://localhost:3000

Note

This chart is intentionally minimal for demonstration purposes. For production clusters, prefer the official kube-prometheus-stack chart, which adds Alertmanager, pre-built dashboards, and Kubernetes-native CRDs (ServiceMonitor/PodMonitor). This chart shows the underlying mechanics those larger charts automate.