Skip to main content

Deployments

A Deployment provides declarative updates for Pods and ReplicaSets.

  • Manages the creation and scaling of ReplicaSets
  • Supports rolling updates and rollbacks
  • Ensures the desired number of Pods are running at all times
More Information

Example​

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:1.25
        ports:
        - containerPort: 80

Rolling Updates​

  • Default update strategy
  • Gradually replaces old Pods with new ones
  • Configurable via maxUnavailable (max Pods that can be unavailable during update) and maxSurge (max Pods created above the desired count)
  • The update is triggered when the Pod template (.spec.template) is changed
strategy:
  type: RollingUpdate
  rollingUpdate:
    maxUnavailable: 1
    maxSurge: 1

Rollbacks​

  • Deployment keeps a revision history of ReplicaSets
  • You can roll back to a previous revision:
# Check rollout history
kubectl rollout history deployment/nginx-deployment

# Roll back to the previous revision
kubectl rollout undo deployment/nginx-deployment

# Roll back to a specific revision
kubectl rollout undo deployment/nginx-deployment --to-revision=2
  • The number of revisions kept is controlled by .spec.revisionHistoryLimit (default: 10)
More Information

Scaling​

# Scale manually
kubectl scale deployment/nginx-deployment --replicas=5
  • Can also be scaled automatically with a HorizontalPodAutoscaler (HPA)

Recreate Strategy​

  • Terminates all existing Pods before creating new ones
  • Results in downtime during the update
  • Useful when the application does not support running multiple versions simultaneously
strategy:
  type: Recreate
More Information