Mastering kubectl: Essential Commands for Kubernetes Operations

Kubectl Common Commands Guide

Kubectl is the most direct way to operate a Kubernetes cluster, especially for operations personnel who need detailed mastery of these commands.

Kubectl Autocomplete

# setup autocomplete in bash, bash-completion package should be installed first.</code>
<code>$ source <(kubectl completion bash)</code>
<code># setup autocomplete in zsh</code>
<code>$ source <(kubectl completion zsh)

Kubectl Context and Configuration

Set the Kubernetes cluster for kubectl interactions and modify configuration information. Refer to the kubeconfig file documentation for detailed information.

# Show merged kubeconfig</code>
<code>$ kubectl config view</code>
<code># Use multiple kubeconfig files and view merged config</code>
<code>$ KUBECONFIG=~/.kube/config:~/.kube/kubconfig2 kubectl config view</code>
<code># Get password for e2e user</code>
<code>$ kubectl config view -o jsonpath='{.users[?(@.name == "e2e")].user.password}'</code>
<code># Show current context</code>
<code>$ kubectl config current-context</code>
<code># Set default context</code>
<code>$ kubectl config use-context my-cluster-name</code>
<code># Add a new cluster with basic authentication</code>
<code>$ kubectl config set-credentials kubeuser/foo.kubernetes.com --username=kubeuser --password=kubepassword</code>
<code># Set context with specific user and namespace</code>
<code>$ kubectl config set-context gce --user=cluster-admin --namespace=foo && kubectl config use-context gce

Create Objects

Kubernetes manifests can be defined in JSON or YAML format, using the extensions .yaml, .yml, or .json.

# Create resources</code>
<code>$ kubectl create -f ./my-manifest.yaml</code>
<code># Create resources from multiple files</code>
<code>$ kubectl create -f ./my1.yaml -f ./my2.yaml</code>
<code># Create resources from all manifests in a directory</code>
<code>$ kubectl create -f ./dir</code>
<code># Create resources from a URL</code>
<code>$ kubectl create -f https://git.io/vPieo</code>
<code># Run an nginx instance</code>
<code>$ kubectl run nginx --image=nginx</code>
<code># Get documentation for pods and services</code>
<code>$ kubectl explain pods,svc</code>
<code># Create multiple YAML objects from stdin</code>
<code>$ cat <<EOF | kubectl create -f -
apiVersion: v1
kind: Pod
metadata:
  name: busybox-sleep
spec:
  containers:
  - name: busybox
    image: busybox
    args:
    - sleep
    - "1000000"
---
apiVersion: v1
kind: Pod
metadata:
  name: busybox-sleep-less
spec:
  containers:
  - name: busybox
    image: busybox
    args:
    - sleep
    - "1000"
EOF</code>
<code># Create a Secret with several keys</code>
<code>$ cat <<EOF | kubectl create -f -
apiVersion: v1
kind: Secret
metadata:
  name: mysecret
type: Opaque
data:
  password: $(echo "s33msi4" | base64)
  username: $(echo "jane" | base64)
EOF

Show and Find Resources

# List all services in all namespaces</code>
<code>$ kubectl get services</code>
<code># List all pods in all namespaces</code>
<code>$ kubectl get pods --all-namespaces</code>
<code># List pods with wide output</code>
<code>$ kubectl get pods -o wide</code>
<code># Get a specific deployment</code>
<code>$ kubectl get deployment my-dep</code>
<code># Include uninitialized pods</code>
<code>$ kubectl get pods --include-uninitialized</code>
<code># Describe nodes and pods</code>
<code>$ kubectl describe nodes my-node</code>
<code>$ kubectl describe pods my-pod</code>
<code># List services sorted by name</code>
<code>$ kubectl get services --sort-by=.metadata.name</code>
<code># Sort pods by restart count</code>
<code>$ kubectl get pods --sort-by='.status.containerStatuses[0].restartCount'</code>
<code># Get version label of pods with app=cassandra</code>
<code>$ kubectl get pods --selector=app=cassandra -o jsonpath='{.items[*].metadata.labels.version}'</code>
<code># Get ExternalIP of all nodes</code>
<code>$ kubectl get nodes -o jsonpath='{.items[*].status.addresses[?(@.type=="ExternalIP")].address}'</code>
<code># List pod names belonging to a specific replication controller (example uses jq)</code>
<code>$ sel=${$(kubectl get rc my-rc --output=json | jq -j '.spec.selector | to_entries | .[] | "\(.key)=\(.value),"')%?}
$ echo $(kubectl get pods --selector=$sel --output=jsonpath={.items..metadata.name})</code>
<code># Show which nodes are ready</code>
<code>$ JSONPATH='{range .items[*]}{@.metadata.name}:{range @.status.conditions[*]}{@.type}={@.status};{end}{end}' && kubectl get nodes -o jsonpath="$JSONPATH" | grep "Ready=True"</code>
<code># List Secrets used by current pods</code>
<code>$ kubectl get pods -o json | jq '.items[].spec.containers[].env[]?.valueFrom.secretKeyRef.name' | grep -v null | sort | uniq

Update Resources

$ kubectl rolling-update frontend-v1 -f frontend-v2.json           # Rolling update pod frontend-v1</code>
<code>$ kubectl rolling-update frontend-v1 frontend-v2 --image=image:v2  # Update resource name and image</code>
<code>$ kubectl rolling-update frontend --image=image:v2                 # Update image in frontend pod</code>
<code>$ kubectl rolling-update frontend-v1 frontend-v2 --rollback        # Roll back ongoing update</code>
<code>$ cat pod.json | kubectl replace -f -                              # Replace pod from stdin JSON</code>
<code># Force replace (deletes then recreates, causing service interruption)</code>
<code>$ kubectl replace --force -f ./pod.json</code>
<code># Expose RC nginx on port 80 mapping to container port 8000</code>
<code>$ kubectl expose rc nginx --port=80 --target-port=8000</code>
<code># Update image tag of a single‑container pod to v4</code>
<code>$ kubectl get pod mypod -o yaml | sed 's/\(image: myimage\):.*$/\1:v4/' | kubectl replace -f -</code>
<code># Add a label</code>
<code>$ kubectl label pods my-pod new-label=awesome</code>
<code># Add an annotation</code>
<code>$ kubectl annotate pods my-pod icon-url=http://goo.gl/XXBTWq</code>
<code># Autoscale deployment "foo"</code>
<code>$ kubectl autoscale deployment foo --min=2 --max=10

Patch Resources

Use strategic merge patches to modify resources.

# Partially update a node</code>
<code>kubectl patch node k8s-node-1 -p '{"spec":{"unschedulable":true}}'</code>
<code># Update container image in a pod (name is required for merge key)</code>
<code>$ kubectl patch pod valid-pod -p '{"spec":{"containers":[{"name":"kubernetes-serve-hostname","image":"new image"}]}}'</code>
<code># JSON patch to update container image</code>
<code>$ kubectl patch pod valid-pod --type='json' -p='[{"op": "replace", "path": "/spec/containers/0/image", "value":"new image"}]'</code>
<code># JSON patch to remove livenessProbe from a deployment</code>
<code>$ kubectl patch deployment valid-deployment --type json -p='[{"op": "remove", "path": "/spec/template/spec/containers/0/livenessProbe"}]'

Edit Resources

Edit any API resource in your preferred editor.

# Edit service named docker-registry</code>
<code>$ kubectl edit svc/docker-registry</code>
<code># Use a different editor</code>
<code>$ KUBE_EDITOR="nano" kubectl edit svc/docker-registry

Scale Resources

# Scale a replicaset named 'foo' to 3</code>
<code>$ kubectl scale --replicas=3 rs/foo</code>
<code># Scale a resource defined in foo.yaml to 3</code>
<code>$ kubectl scale --replicas=3 -f foo.yaml</code>
<code># Scale deployment mysql from 2 to 3 replicas</code>
<code>$ kubectl scale --current-replicas=2 --replicas=3 deployment/mysql</code>
<code># Scale multiple replication controllers</code>
<code>$ kubectl scale --replicas=5 rc/foo rc/bar rc/baz

Delete Resources

# Delete resources defined in pod.json</code>
<code>$ kubectl delete -f ./pod.json</code>
<code># Delete a pod and a service</code>
<code>$ kubectl delete pod,service baz foo</code>
<code># Delete pods and services with a label</code>
<code>$ kubectl delete pods,services -l name=myLabel</code>
<code># Include uninitialized resources</code>
<code>$ kubectl delete pods,services -l name=myLabel --include-uninitialized</code>
<code># Delete all pods and services in a namespace</code>
<code>$ kubectl -n my-ns delete po,svc --all

Interact with Running Pods

# Show pod logs</code>
<code>$ kubectl logs my-pod</code>
<code># Show logs of a specific container</code>
<code>$ kubectl logs my-pod -c my-container</code>
<code># Stream pod logs</code>
<code>$ kubectl logs -f my-pod</code>
<code># Run an interactive shell in a pod</code>
<code>$ kubectl run -i --tty busybox --image=busybox -- sh</code>
<code># Attach to a running container</code>
<code>$ kubectl attach my-pod -i</code>
<code># Port‑forward pod port 6000 to local 5000</code>
<code>$ kubectl port-forward my-pod 5000:6000</code>
<code># Execute a command in a pod (single‑container case)</code>
<code>$ kubectl exec my-pod -- ls /</code>
<code># Execute a command in a specific container</code>
<code>$ kubectl exec my-pod -c my-container -- ls /</code>
<code># Show metrics for a pod and its containers</code>
<code>$ kubectl top pod POD_NAME --containers

Interact with Nodes and Cluster

# Mark node unschedulable</code>
<code>$ kubectl cordon my-node</code>
<code># Drain node for maintenance</code>
<code>$ kubectl drain my-node</code>
<code># Mark node schedulable again</code>
<code>$ kubectl uncordon my-node</code>
<code># Show node metrics</code>
<code>$ kubectl top node my-node</code>
<code># Show cluster info</code>
<code>$ kubectl cluster-info</code>
<code># Dump cluster state to stdout</code>
<code>$ kubectl cluster-info dump</code>
<code># Dump cluster state to a directory</code>
<code>$ kubectl cluster-info dump --output-directory=/path/to/cluster-state</code>
<code># Update or add a taint</code>
<code>$ kubectl taint nodes foo dedicated=special-user:NoSchedule

kubectl set Command

The kubectl set family configures specific resources or modifies existing ones.

kubectl set resources

This command sets resource limits and requests for containers.

# Set CPU limit to 200m and memory to 512Mi for nginx container in a deployment</code>
<code>$ kubectl set resources deployment nginx -c=nginx --limits=cpu=200m,memory=512Mi</code>
<code># Set both limits and requests</code>
<code>$ kubectl set resources deployment nginx --limits=cpu=200m,memory=512Mi --requests=cpu=100m,memory=256Mi</code>
<code># Remove resource specifications</code>
<code>$ kubectl set resources deployment nginx --limits=cpu=0,memory=0 --requests=cpu=0,memory=0

kubectl set selector

Sets the selector for a resource; existing selectors are overwritten.

Syntax: selector (-f FILENAME | TYPE NAME) EXPRESSIONS [--resource-version=version]

kubectl set image

Updates the container image of existing resources. Supported resource types include pod, replicationcontroller, deployment, daemonset, job, and replicaset.

# Set nginx container image to nginx:1.9.1 in a deployment</code>
<code>$ kubectl set image deployment/nginx nginx=nginx:1.9.1</code>
<code># Update all deployments and RCs</code>
<code>$ kubectl set image deployments,rc nginx=nginx:1.9.1 --all</code>
<code># Update all containers in a daemonset</code>
<code>$ kubectl set image daemonset abc *=nginx:1.9.1</code>
<code># Update image from a local file</code>
<code>$ kubectl set image -f path/to/file.yaml nginx=nginx:1.9.1 --local -o yaml

Resource Types

The table below lists all supported Kubernetes resource types and their aliases.

Formatted Output

To output detailed information in a specific format, add the -o or --output flag to the kubectl command.

Kubectl Detailed Output and Debugging

Use the -v or --v flag followed by an integer to set the log verbosity level.