grafana-prometheus-kubenetes

Grafana + Prometheus (Kubernetes Metrics)

This stack provides monitoring and visualization for Kubernetes-related metrics.
Prometheus collects and stores metrics, while Grafana reads from Prometheus and renders dashboards.

How it works

flowchart LR
    K8s([Kubernetes]) -->|metrics| Prometheus[Prometheus :9090]
    Prometheus --> Data[(Time-Series DB)]
    Grafana[Grafana :3000] --> Prometheus
    User([User]) --> Grafana

Loading

1. Prometheus scrapes metrics targets configured in prometheus/prometheus.yml.
2. Metrics are stored in Prometheus time-series storage.
3. Grafana connects to Prometheus as a datasource.
4. You explore metrics through Grafana dashboards and queries.

Stack details in this repo

• Prometheus image: prom/prometheus:latest
• Grafana image: grafana/grafana:latest
• Prometheus UI: http://<host-ip>:9090
• Grafana UI: http://<host-ip>:3000
• Persistent data:
  • prometheus_data:/prometheus
  • grafana_data:/var/lib/grafana
• Mounted config/files:
  • ./prometheus/prometheus.yml:/etc/prometheus/prometheus.yml
  • ./token.txt:/etc/prometheus/token.txt:ro

Environment variables

Set via .env (copy from .env.example):

• GF_SECURITY_ADMIN_USER (default: admin)
• GF_SECURITY_ADMIN_PASSWORD (default: admin)
• GF_USERS_ALLOW_SIGN_UP (default: false)

How to run

From the repository root:

cd grafana-prometheus-kubenetes
cp .env.example .env
docker compose up -d

Open:

• Prometheus: http://localhost:9090
• Grafana: http://localhost:3000

Use the Grafana admin credentials from your .env.

Useful commands:

docker compose ps
docker compose logs -f
docker compose restart
docker compose down

Use it effectively

• Validate scrape targets in Prometheus (Status -> Targets).
• Add/import Grafana dashboards for Kubernetes components.
• Keep token.txt secure because it may contain cluster access credentials.

Notes

• Ensure prometheus/prometheus.yml has reachable targets from this Docker host.
• Change default Grafana password before exposing the stack externally.