Below is a list of the most common kubectl commands you'll frequently use to manage your setup:
Check cluster components (control plane availability):
kubectl get componentstatusesGet general cluster information:
kubectl cluster-infoList all nodes in the cluster:
kubectl get nodesGet detailed information about a node:
kubectl describe node <node-name>View all pods across all namespaces:
kubectl get pods --all-namespacesList the pods in a specific namespace (e.g., default, longhorn-system):
kubectl get pods -n <namespace>Get detailed information for a specific pod:
kubectl describe pod <pod-name> -n <namespace>Delete a pod (useful for forcing a restart):
kubectl delete pod <pod-name> -n <namespace>Create or apply resources from a YAML file (deploy manifests, services, etc.):
kubectl apply -f <filename>.yamlGet logs from a pod (basic logs for debugging):
kubectl logs <pod-name> -n <namespace>Stream continuous logs from a pod (for dynamic updates):
kubectl logs -f <pod-name> -n <namespace>Get logs for a specific container in a multi-container pod:
kubectl logs <pod-name> -c <container-name> -n <namespace>Service & Endpoint Management 13. List all services in a namespace:
kubectl get svc -n <namespace>Get detailed information about a service:
kubectl describe svc <service-name> -n <namespace>Forward a local port to a port within a pod (e.g., for accessing a specific pod's service locally like a database):
kubectl port-forward <pod-name> <local-port>:<remote-port> -n <namespace>Storage Management (Longhorn) 16. List Longhorn volumes:
kubectl get volumes -n longhorn-systemDescribe a Longhorn volume:
kubectl describe <longhorn-volume-name> -n longhorn-systemCheck the status of Longhorn-csi or any stateful sets:
kubectl get statefulsets -n longhorn-systemList all namespaces:
kubectl get namespacesSwitch context to a different namespace:
Change the default namespace to avoid having to specify -n <namespace> each time:
kubectl config set-context --current --namespace=<namespace>List PostgreSQL-related resources (assuming you have some CRDs or kubedb/k8s-postgres-operator):
kubectl get postgresql -n <namespace>Describe a PostgreSQL instance, if any:
kubectl describe postgresql <pg-instance-name> -n <namespace>Connect to the PostgreSQL pod for database debugging:
kubectl exec -it <pg-pod-name> -n <namespace> -- psql -U postgresView resource usage (CPU/Memory) for nodes and pods (if metrics-server is installed):
kubectl top nodes
kubectl top pods -n <namespace>Check events for troubleshooting issues within a namespace:
kubectl get events -n <namespace>Get details about a Deployment:
kubectl describe deployment <deployment-name> -n <namespace>Scale up/down the number of replicas in a Deployment:
kubectl scale deployment <deployment-name> --replicas=<number-of-replicas> -n <namespace>Check recent events to diagnose issues:
kubectl get events --sort-by='.metadata.creationTimestamp' -n <namespace>Open a shell session inside a running container:
kubectl exec -it <pod-name> -n <namespace> -- /bin/bashRun one-off commands in a container (e.g., to run a curl command):
kubectl exec -it <pod-name> -n <namespace> -- curl <url>List all service accounts:
kubectl get serviceaccounts -n <namespace>Get details about a specific service account:
kubectl describe serviceaccount <service-account-name> -n <namespace>View all ConfigMaps (commonly used for storing configuration data):
kubectl get configmap -n <namespace>Describe a specific ConfigMap for details:
kubectl describe configmap <configmap-name> -n <namespace>List Secrets (API keys, credentials, etc.):
kubectl get secrets -n <namespace>Decode a base64-encoded Secret to reveal its content:
kubectl get secret <secret-name> -n <namespace> -o jsonpath="{.data.<secret-key>}" | base64 --decode- Always be careful when using commands like
delete, especially if you're managing persistent services like Longhorn, PostgreSQL databases, etc., since it could lead to unintended data loss. - You may also install tools like
kubectl krewto extend the functionality ofkubectl.