Prometheus + Grafana Training

• Prometheus
• Grafana
• kube-prometheus-stack Helm chart (Prometheus Operator-based installation)

---

Repo structure

.
├── 00-namespace.yaml
├── 1-stack/                 # kube-prometheus-stack Helm values
├── 2-nginx/                  # nginx + stub_status enabled
├── 3-exporters/              # nginx exporter deployment + service
├── 4-servicemonitors/        # ServiceMonitor objects
├── 5-access/                 # helper scripts (port-forward + credentials)
└── debug/                    # quick debugging helpers

---

1) Create namespace

kubectl apply -f 00-namespace.yaml

2) Install Prometheus + Grafana stack (minimal)

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

helm install sina-monitoring prometheus-community/kube-prometheus-stack \
  -n monitoring \
  -f 1-stack/values-minimal.yaml \
  --wait \
  --timeout 15m

3) Deploy nginx + exporter + ServiceMonitor

kubectl apply -n monitoring -f 2-nginx/
kubectl apply -n monitoring -f 3-exporters/
kubectl apply -n monitoring -f 4-servicemonitors/

---

Access

Start port-forwards

./5-access/start-port-forwards.sh

Stop port-forwards

./5-access/stop-port-forwards.sh

Grafana credentials

./5-access/credentials.sh

URLs

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

---

validation

1) targets

• http://localhost:9090/

You should see:

• nginx-exporter target UP
• node-exporter, kubelet, kube-state-metrics targets UP

2) NGINX metrics

PromQL :

• nginx_up → should be 1
• nginx_http_requests_total exists

3) Grafana dashboards

• Node Exporter Full: 1860
• NGINX Exporter: 12708

---