# Deploying Local Applications on Kubernetes: kubectl vs Helm In this guide, we’ll walk through deploying a basic web service to a Kubernetes cluster using two approaches: manually with `kubectl` and declaratively with Helm. Understanding both paths gives insight into Kubernetes application lifecycles and toolchain evolution. --- ## 1. Prerequisites * A running Kubernetes cluster (e.g., K3s, Minikube) * `kubectl` configured and tested * `helm` installed For simplicity, we'll deploy a containerized HTTP server using the `hashicorp/http-echo` image. --- ## 2. Deploying via kubectl (Raw Manifests) ### 2.1 Create YAML Files **deployment.yaml** ```yaml apiVersion: apps/v1 kind: Deployment metadata: name: echo-app spec: replicas: 1 selector: matchLabels: app: echo template: metadata: labels: app: echo spec: containers: - name: echo image: hashicorp/http-echo args: ["-text=Hello from kubectl"] ports: - containerPort: 5678 ``` **service.yaml** ```yaml apiVersion: v1 kind: Service metadata: name: echo-service spec: selector: app: echo ports: - protocol: TCP port: 80 targetPort: 5678 type: ClusterIP ``` ### 2.2 Apply the Manifests ```bash kubectl apply -f deployment.yaml kubectl apply -f service.yaml ``` ### 2.3 Verify ```bash kubectl get pods kubectl get svc ``` --- ## 3. Deploying via Helm ### 3.1 Create a Helm Chart ```bash helm create echo-chart ``` Replace the default template files: **echo-chart/values.yaml** ```yaml replicaCount: 1 image: repository: hashicorp/http-echo pullPolicy: IfNotPresent tag: latest args: - "-text=Hello from Helm" service: type: ClusterIP port: 80 containerPort: 5678 ``` **echo-chart/templates/deployment.yaml** ```yaml apiVersion: apps/v1 kind: Deployment metadata: name: {{ include "echo-chart.fullname" . }} spec: replicas: {{ .Values.replicaCount }} selector: matchLabels: app: {{ include "echo-chart.name" . }} template: metadata: labels: app: {{ include "echo-chart.name" . }} spec: containers: - name: {{ .Chart.Name }} image: "{{ .Values.image.repository }}:{{ .Values.image.tag }}" args: {{ .Values.args }} ports: - containerPort: {{ .Values.containerPort }} ``` **echo-chart/templates/service.yaml** ```yaml apiVersion: v1 kind: Service metadata: name: {{ include "echo-chart.fullname" . }} spec: type: {{ .Values.service.type }} selector: app: {{ include "echo-chart.name" . }} ports: - port: {{ .Values.service.port }} targetPort: {{ .Values.containerPort }} ``` ### 3.2 Install the Chart ```bash helm install echo-release ./echo-chart ``` ### 3.3 Verify ```bash helm list kubectl get all ``` --- ## 4. Summary: kubectl vs Helm | Feature | kubectl | Helm | | --------------------- | --------------------------- | -------------------------------------- | | File structure | Multiple flat YAMLs | Chart directory with templates | | Reusability | Limited | High (templated values, overrides) | | Rollback support | Manual | Built-in rollback/upgrade capability | | Application packaging | None | Yes (charts can be versioned & shared) | | Best for | Simple, one-off deployments | Structured, repeatable deployments | --- ## Common kubectl Commands by Use Case ### Cluster Management ```bash # Get cluster information kubectl cluster-info kubectl version kubectl get nodes kubectl describe node # Switch contexts and namespaces kubectl config get-contexts kubectl config use-context kubectl config set-context --current --namespace= ``` ### Resource Management ```bash # Create resources kubectl create namespace dev kubectl create deployment nginx --image=nginx kubectl create service clusterip nginx --tcp=80:80 # Apply manifests kubectl apply -f deployment.yaml kubectl apply -f ./manifests kubectl apply -k ./kustomize-dir # Delete resources kubectl delete pod kubectl delete -f deployment.yaml kubectl delete namespace dev --grace-period=0 --force ``` ### Monitoring and Debugging ```bash # Get resource status kubectl get pods -A --watch kubectl get deployments -n production kubectl get events --sort-by='.metadata.creationTimestamp' # Describe resources kubectl describe pod kubectl describe deployment kubectl describe service # View logs kubectl logs kubectl logs -f -c kubectl logs --previous # Previous container instance # Execute commands in containers kubectl exec -it -- /bin/bash kubectl exec -- curl -s localhost:8080 ``` ### Application Scaling and Updates ```bash # Scale resources kubectl scale deployment nginx --replicas=3 kubectl autoscale deployment nginx --min=2 --max=5 --cpu-percent=80 # Update applications kubectl set image deployment/nginx nginx=nginx:1.19 kubectl rollout status deployment/nginx kubectl rollout history deployment/nginx kubectl rollout undo deployment/nginx ``` ### Resource Inspection ```bash # Get detailed information kubectl get pod -o yaml kubectl get service -o json kubectl get deployments -o wide # Custom columns and sorting kubectl get pods --sort-by='.metadata.creationTimestamp' kubectl get pods -o custom-columns=NAME:.metadata.name,STATUS:.status.phase ``` ### Troubleshooting ```bash # Network debugging kubectl port-forward pod/ 8080:80 kubectl proxy --port=8080 kubectl top pods kubectl top nodes # Resource validation kubectl auth can-i create deployments kubectl api-resources kubectl explain deployment.spec ``` ### Configuration and Security ```bash # Config and secrets kubectl create configmap app-config --from-file=./config.properties kubectl create secret generic db-creds \ --from-literal=username=admin \ --from-literal=password=secret # RBAC management kubectl create serviceaccount kubectl create role pod-reader --verb=get --verb=list --resource=pods kubectl create rolebinding pod-reader-binding --role=pod-reader --serviceaccount=default:default ``` These commands cover the essential kubectl operations for managing Kubernetes clusters and applications effectively. --- ## Common Helm Commands by Use Case ### Repository Management ```bash # Add and update repositories helm repo add bitnami https://charts.bitnami.com/bitnami helm repo update helm repo list helm repo remove bitnami # Search for charts helm search repo nginx helm search hub postgresql ``` ### Chart Installation and Upgrades ```bash # Install a chart with specific release name and namespace helm install nginx-prod bitnami/nginx --namespace prod helm install redis-cache bitnami/redis --set password=secretpass # Upgrade existing releases helm upgrade nginx-prod bitnami/nginx --reuse-values helm upgrade --install nginx-prod bitnami/nginx # Install if not exists # Roll back to previous release helm rollback nginx-prod 1 # Roll back to revision 1 ``` ### Release Management ```bash # List and inspect releases helm list --all-namespaces helm status nginx-prod helm history nginx-prod # Uninstall releases helm uninstall nginx-prod helm uninstall nginx-prod --keep-history # Preserve release history ``` ### Development and Debugging ```bash # Test template rendering helm template nginx-prod bitnami/nginx helm template . --debug # For local chart development # Validate chart helm lint ./my-chart helm install nginx-prod . --dry-run # Show differences before upgrade helm diff upgrade nginx-prod bitnami/nginx # Requires helm-diff plugin ``` ### Value Management ```bash # Work with values helm get values nginx-prod helm get values nginx-prod --revision 1 helm show values bitnami/nginx # Install/upgrade with multiple value files helm install nginx-prod bitnami/nginx \ -f values.yaml \ -f values.prod.yaml \ --set service.type=LoadBalancer ``` ### Package Management ```bash # Create and package charts helm create my-chart helm package ./my-chart helm push my-chart-0.1.0.tgz chartmuseum # Requires push plugin # Download chart dependencies helm dependency update ./my-chart helm dependency build ./my-chart ``` ### Plugin Management ```bash # Work with plugins helm plugin install https://github.com/databus23/helm-diff helm plugin list helm plugin uninstall diff ``` These commands cover most day-to-day Helm operations, from basic chart installation to advanced development workflows. --- ## Next Steps * Integrate Helm charts into your CI/CD pipelines * Add Ingress for external exposure * Expand charts with `ConfigMap`, `Secrets`, `Probes`, and resource limits Mastering both `kubectl` and Helm ensures you can operate at all abstraction levels of Kubernetes—from imperative one-off tasks to fully automated rollouts.

Design a Min Stack