Deploy containerized applications to your Kubernetes clusters with Clanker’s simplified deployment workflow.
Quick deployment Deploy a container image with automatic service creation:
# Deploy nginx with default settings clanker k8s deploy nginx # Deploy with custom name and port clanker k8s deploy nginx --name my-app --port 80 --replicas 3 # Deploy to specific namespace clanker k8s deploy myapp:v1.0 --namespace production --replicas 5
Clanker automatically creates both a Deployment and a LoadBalancer Service for the application.
Deployment options Basic configuration Container image to deploy (e.g., nginx, myapp:v1.0)
Deployment name (defaults to image name without tag)
Kubernetes namespace for deployment
Preview and apply Show deployment plan without applying changes
Apply deployment without confirmation prompt
Deployment plan View what will be created before deploying:
clanker k8s deploy nginx --name web --replicas 3 --plan
Example output: === Deployment Plan === Operation: Deploy application Name: web Image: nginx Replicas: 3 Namespace: default Resources to be created: ✓ Deployment: web ✓ Service: web (LoadBalancer) Steps: 1. Create Deployment manifest 2. Create Service manifest 3. Apply manifests to cluster 4. Wait for pods to be ready Connection: kubectl get pods -n default -l app=web kubectl get service web -n default
Generated manifests Clanker generates standard Kubernetes manifests for deployments:
apiVersion : apps/v1 kind : Deployment metadata : name : nginx namespace : default spec : replicas : 3 selector : matchLabels : app : nginx template : metadata : labels : app : nginx spec : containers : - name : nginx image : nginx ports : - containerPort : 80
Deployment workflow
Generate plan
Clanker analyzes the deployment parameters and generates an execution plan showing what resources will be created.
Confirm changes
Review the plan and confirm (or use --apply to skip confirmation).
Apply manifests
Clanker applies the Deployment and Service manifests to the cluster using kubectl apply.
Verify deployment
Monitor pod creation and readiness: kubectl get pods -l app=nginx kubectl get service nginx
Managing deployments After deployment, use standard kubectl commands:
# View pods kubectl get pods -l app=nginx # View service and get LoadBalancer IP kubectl get service nginx # Scale deployment kubectl scale deployment nginx --replicas=5 # Update image kubectl set image deployment/nginx nginx=nginx:1.21 # View logs kubectl logs -l app=nginx # Delete deployment kubectl delete deployment nginx kubectl delete service nginx
Advanced deployment scenarios Deploy to multiple namespaces Deploy the same application to different environments:
# Development clanker k8s deploy myapp:dev --namespace development --replicas 1 # Staging clanker k8s deploy myapp:staging --namespace staging --replicas 2 # Production clanker k8s deploy myapp:v1.0 --namespace production --replicas 5
Custom port mappings Deploy applications with non-standard ports:
# Node.js app on port 3000 clanker k8s deploy myapp:latest --port 3000 # Go API on port 8080 clanker k8s deploy api-server:v2.0 --port 8080 --replicas 3
High availability deployments Deploy with multiple replicas for redundancy:
clanker k8s deploy critical-app:v1.0 \ --replicas 5 \ --namespace production
Kubernetes will distribute replicas across nodes for high availability.
Implementation details The deployment process is implemented in cmd/k8s.go:939:
func runDeploy ( cmd * cobra . Command , args [] string ) error { image := args [ 0 ] ctx := context . Background () deployName := k8sDeployName if deployName == "" { // Extract name from image parts := strings . Split ( image , "/" ) deployName = parts [ len ( parts ) - 1 ] if idx := strings . Index ( deployName , ":" ); idx > 0 { deployName = deployName [: idx ] } } // Generate deploy plan deployPlan := plan . GenerateDeployPlan ( plan . DeployOptions { Name : deployName , Image : image , Port : k8sDeployPort , Replicas : k8sReplicas , Namespace : k8sNamespace , Type : "deployment" , }) // Build deployment manifest manifest := fmt . Sprintf ( `apiVersion: apps/v1 kind: Deployment metadata: name: %s namespace: %s spec: replicas: %d selector: matchLabels: app: %s template: metadata: labels: app: %s spec: containers: - name: %s image: %s ports: - containerPort: %d --- apiVersion: v1 kind: Service metadata: name: %s namespace: %s spec: selector: app: %s ports: - port: %d targetPort: %d type: LoadBalancer ` , deployName , k8sNamespace , k8sReplicas , deployName , deployName , deployName , image , k8sDeployPort , deployName , k8sNamespace , deployName , k8sDeployPort , k8sDeployPort ) // Apply using kubectl client := k8s . NewClient ( "" , "" , viper . GetBool ( "debug" )) output , err := client . Apply ( ctx , manifest , k8sNamespace ) if err != nil { return fmt . Errorf ( "failed to deploy: %w " , err ) } fmt . Println ( output ) return nil }
Troubleshooting Pods not starting Check pod status and events:
# View pod status kubectl get pods -l app=myapp # Describe pod for events kubectl describe pod < pod-nam e > # Check logs kubectl logs < pod-nam e >
Service not accessible Verify service configuration:
# Get service details kubectl get service myapp -o wide # Check endpoints kubectl get endpoints myapp # Verify pods are ready kubectl get pods -l app=myapp
Image pull errors Ensure the image exists and is accessible:
# Use fully qualified image name clanker k8s deploy docker.io/library/nginx:latest # For private registries, create image pull secret kubectl create secret docker-registry regcred \ --docker-server= < registry > \ --docker-username= < user > \ --docker-password= < password >
Next steps
Ask mode Query your deployments with natural language
Monitoring Monitor deployment health and metrics