Kubernetes Commands¶

Kubectl¶

Installation¶

'x86/64' architecture

curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"

Version and Info¶

kubectl version
kubectl cluster-info
kubectl cluster-info dump > cluster-info.dump

Top¶

Check metric server exists

kubectl top nodes

If returned any values means 'metric server' is active and the other not.

Nodes¶

Get All Nodes¶

kubectl get nodes
kubectl get nodes -o wide
kubectl get nodes -o wide -w/--watch

Describe Nodes¶

kubectl describe node <node-name> #describe node details including limits

Set label for a node¶

kubectl label node <node-name> kubernetes.io/role=node
kubectl label node <node-name> label_key=label_value
kubectl label node <node-name> label_key- # remove added label from node"

Taint¶

add/del taint

kubectl taint node <node-name> taint_label_key=taint_label_value:TaintEffect # add taint label to node"
kubectl taint node <node-name> taint_label_key- # remove added taint label to node"

TaintEffect: [NoSchedule, PreferNoSchedule or NoExecute]

get node taint

kubectl describe node <node-name> | grep -i "Taints"

Pods¶

kubectl -n kube-system get pods -o wide
kubectl -n kube-system describe pods <podname>
kubectl delete <pod_name>
kubectl get pods --show-labels
kubectl get pods -o yaml > pods.yaml
kubectl describe pod
kubectl explain pod
kubectl explain pod.spec
kubectl explain pod.spec.ports
kubectl explain pod.spec.containers.ports
kubectl explain pod.spec.affinity
kubectl exec -it <pod-name> bash # get an interactive shell from a running pod
kubectl exec -it <pod-name> -- bash
kubectl get pods --selector key=value

Resource Modification¶

kubectl create -f <pod_file_name.yml>
kubectl replace -f <pod_file_name.yml>
kubectl get rs -o wide
kubectl apply -f <file-name.yml>
kubectl delete -f <manifest-filename.yml>

Explain¶

kubectl explain <config-kind>
kubectl explain resourcequota
kubectl explain resourcequota.spec
kubectl explain resourcequota.spec.hard

Update a resource¶

kubectl edit <unit-type> <unit-name>
kubectl scale --replicas=6 deployment <deployment-name>

Namespace¶

kubectl get namespace
kubectl get ns
kubectl create namespace <ns-name>
kubectl get -n <ns-name> pods
kubectl describe <ns-name>
kubectl -n <ns-name> events
kubectl get pods --all-namespaces
kubectl config view --minify --output 'jsonpath={..namespace}' # get current ns

ConfigMap¶

# imperative method
# from cli
kubectl create configmap <config-name> --from-literal=<key=value>
# from file
kubectl create configmap <config-name> --from-file=<path-to-file>
# declarative method
kubectl create/apply -f <configmapfile.yml>
# describe
kubectl describe cm <configmapfile.yml>
kubectl describe cm <configmapfile.yml> --output yaml

Secret¶

# imperative method
# from cli
kubectl create secret <config-name> --from-literal=<key=value>
# from file
kubectl create secret <config-name> --from-=<path-to-file>
# declarative method
kubectl create/apply -f <secret-configmapfile.yml>
# describe
kubectl describe secret <configmapfile.yml>
kubectl describe secret <configmapfile.yml> --output yaml

Drain/Cordon/Uncordon¶

kubectl drain <node-name>; drain execute cordon command
kubectl cordon <node-name> ;set unscheduled label to node
kubectl uncordon <node-name> ;Remove unscheduled label from node

Application Lifecycle Commands¶

kubectl create -f <deployment-filename.yml>
kubectl replace -f <deployment-filename.yml>
kubectl apply -f <deployment-filename.yml> # both create & replace
kubectl get <deployment-name>
kubectl set image <deployment-name> <image=imagename:imagetag>

Rollout¶

kubectl rollout status <deployments/deployment-name>
kubectl rollout history <deployments/deployment-name>
kubectl rollout undo <deployments/deployment-name> # go to pervious version
kubectl rollout undo <deployments/deployment-name> --to-revision <rv-no> # go to pervious version
watch -d kubectl get replicasets

watch and debug command¶

kubectl describe <resource-type> <resource-name>
kubectl logs -f <resource-name>
kubectl events --for <resource-type>/<resource-name> --watch
kubectl events --types=Warning,Normal

Kubernetes Dashboard¶

Create serviceaccount first

kubectl create serviceaccount dashboard-admin -n kube-system

Bind SA to Cluster via clusterrolebinding

kubectl create clusterrolebinding dashboard-admin \
  --clusterrole=cluster-admin \
  --serviceaccount=kube-system:dashboard-admin

Get the admin token and paste it to your dashboard page

kubectl -n kube-system create token dashboard-admin --duration=87600h