Scenario

A Deployment named backend in the dev namespace is failing to start. Pods are stuck in ImagePullBackOff due to configuration issues.

Task

Fix the Deployment so pods successfully pull the image and enter the Running state.

Image Information

Step 1: Verify Pod status and identify the error

kubectl describe pod -n dev -l app=backend

Observe status is ImagePullBackOff or ErrImagePull. Events show "Failed to pull image" due to incorrect image name or missing credentials.

Step 2: Fix the image name

kubectl set image deployment/backend alpine=ghcr.io/prepare-sh/alpine:3.23.2 -n dev

Updates the Deployment with the correct image ghcr.io/prepare-sh/alpine:3.23.2.

Step 3: Create the registry secret

kubectl create secret docker-registry regcred \
  --docker-server=ghcr.io \
  --docker-username=preparesh-bot \
  --docker-password=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 \
  -n dev

Creates a docker-registry secret for GitHub Container Registry authentication.

Step 4: Configure ServiceAccount with the secret

kubectl patch serviceaccount default -n dev -p '{"imagePullSecrets": [{"name": "regcred"}]}'

Patches the default ServiceAccount to include regcred, applying credentials to all Pods using that account.

Step 5: Restart Pods to apply changes

kubectl delete pod -n dev -l app=backend

Deletes existing Pods. The Deployment recreates them with the new ServiceAccount configuration.

Step 6: Verify Pods are Running

kubectl get pods -n dev -w

Watch as Pods transition from Pending → ContainerCreating → Running. Press Ctrl+C to exit.