Kubernetes Operations

Deployment Manifest

apiVersion: apps/v1
kind: Deployment
metadata:
  name: api-server
  labels:
    app: api-server
    version: v1
spec:
  replicas: 3
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxSurge: 1
      maxUnavailable: 0
  selector:
    matchLabels:
      app: api-server
  template:
    metadata:
      labels:
        app: api-server
        version: v1
    spec:
      containers:
        - name: api
          image: registry.example.com/api:1.2.0
          ports:
            - containerPort: 8080
          resources:
            requests:
              cpu: 100m
              memory: 128Mi
            limits:
              cpu: 500m
              memory: 512Mi
          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
      topologySpreadConstraints:
        - maxSkew: 1
          topologyKey: kubernetes.io/hostname
          whenUnsatisfiable: DoNotSchedule
          labelSelector:
            matchLabels:
              app: api-server

Always set resource requests and limits. Use topology spread constraints for high availability.

Helm Chart Structure

chart/
  Chart.yaml
  values.yaml
  values-staging.yaml
  values-production.yaml
  templates/
    deployment.yaml
    service.yaml
    ingress.yaml
    hpa.yaml
    _helpers.tpl

# values.yaml
replicaCount: 2
image:
  repository: registry.example.com/api
  tag: "1.2.0"
  pullPolicy: IfNotPresent
resources:
  requests:
    cpu: 100m
    memory: 128Mi
  limits:
    cpu: 500m
    memory: 512Mi
autoscaling:
  enabled: true
  minReplicas: 2
  maxReplicas: 10
  targetCPUUtilization: 70

HorizontalPodAutoscaler

apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: api-server
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: api-server
  minReplicas: 2
  maxReplicas: 10
  metrics:
    - type: Resource
      resource:
        name: cpu
        target:
          type: Utilization
          averageUtilization: 70
    - type: Resource
      resource:
        name: memory
        target:
          type: Utilization
          averageUtilization: 80
  behavior:
    scaleDown:
      stabilizationWindowSeconds: 300

Troubleshooting Commands

# Pod diagnostics
kubectl describe pod <pod-name> -n <namespace>
kubectl logs <pod-name> -c <container> --previous
kubectl exec -it <pod-name> -- /bin/sh

# Resource usage
kubectl top pods -n <namespace> --sort-by=memory
kubectl top nodes

# Network debugging
kubectl run debug --image=nicolaka/netshoot --rm -it -- bash
nslookup <service-name>.<namespace>.svc.cluster.local

# Events sorted by time
kubectl get events -n <namespace> --sort-by='.lastTimestamp'

# Find pods not running
kubectl get pods -A --field-selector=status.phase!=Running

Anti-Patterns

• Running containers as root without securityContext.runAsNonRoot: true
• Missing resource requests/limits (causes scheduling issues and noisy neighbors)
• Using latest tag instead of pinned image versions
• Not setting PodDisruptionBudget for critical workloads
• Storing secrets in ConfigMaps instead of Secrets (or external secret managers)
• Ignoring pod anti-affinity for replicated deployments

Checklist

• All containers have resource requests and limits
• Liveness and readiness probes configured
• Images use specific version tags, not latest
• Secrets stored in Kubernetes Secrets or external vault
• PodDisruptionBudget set for production workloads
• NetworkPolicies restrict traffic between namespaces
• Topology spread constraints or anti-affinity for HA
• Helm values split per environment (staging, production)