Kubernetes

Kubernetes

Overview

Kubernetes orchestrates containerized applications across clusters of machines, handling deployment, scaling, networking, and self-healing. Resources are declared in YAML manifests and managed via kubectl or GitOps tools.

Deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-app
  labels:
    app: my-app
spec:
  replicas: 3
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
    spec:
      containers:
        - name: app
          image: my-app:1.0.0
          ports:
            - containerPort: 8080
          resources:
            requests:
              cpu: 100m
              memory: 128Mi
            limits:
              cpu: 500m
              memory: 256Mi
          livenessProbe:
            httpGet:
              path: /healthz
              port: 8080
            initialDelaySeconds: 10
            periodSeconds: 15
          readinessProbe:
            httpGet:
              path: /ready
              port: 8080
            initialDelaySeconds: 5
            periodSeconds: 5
          env:
            - name: DATABASE_URL
              valueFrom:
                secretKeyRef:
                  name: db-credentials
                  key: url

Service

apiVersion: v1
kind: Service
metadata:
  name: my-app
spec:
  selector:
    app: my-app
  ports:
    - port: 80
      targetPort: 8080
  type: ClusterIP

Ingress

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: my-app
  annotations:
    cert-manager.io/cluster-issuer: letsencrypt-prod
spec:
  tls:
    - hosts: [my-app.example.com]
      secretName: my-app-tls
  rules:
    - host: my-app.example.com
      http:
        paths:
          - path: /
            pathType: Prefix
            backend:
              service:
                name: my-app
                port:
                  number: 80

ConfigMap and Secret

apiVersion: v1
kind: ConfigMap
metadata:
  name: app-config
data:
  LOG_LEVEL: "info"
  FEATURE_FLAGS: "new-ui=true"
---
apiVersion: v1
kind: Secret
metadata:
  name: db-credentials
type: Opaque
stringData:
  url: "postgresql://user:pass@db:5432/myapp"

Key Commands

# Apply manifests
kubectl apply -f manifests/

# Check rollout status
kubectl rollout status deployment/my-app

# Scale
kubectl scale deployment/my-app --replicas=5

# View logs
kubectl logs -f deployment/my-app

# Port forward for debugging
kubectl port-forward svc/my-app 8080:80

# Diff before apply
kubectl diff -f manifests/

Best Practices

• Always set resource requests and limits to ensure fair scheduling and prevent noisy neighbors.
• Use liveness probes (restart on failure) and readiness probes (stop routing traffic) for self-healing.
• Use Secrets for sensitive data, ConfigMaps for configuration — never hardcode either.
• Use kubectl diff before kubectl apply to review changes.
• Pin container image tags to specific versions (not latest) for reproducible deployments.
• Use namespaces to isolate environments or teams.
• Use labels and selectors consistently for organizing and querying resources.
• Use rolling update strategy (default) with maxUnavailable and maxSurge for zero-downtime deploys.