Kubernetes - Commands

Kubectl?

"Kubectl is a command-line interface that lets you run commands on Kubernetes clusters." This is how we will be able to perform various operations on our cluster.
Kubectl depends on a kubeconfig. This is a configuration file for access to one or more clusters, to know which cluster is configured our Kubectl command, we can use:

Kubectl config current-context
Or if we want to change clusters in our config, we can use:

Kubectl config use-context

CLIENT CONFIGURATION

Setup autocomplete in bash; bash-completion package should be installed first

source <(kubectl completion bash)

View Kubernetes config

kubectl config view

View specific config items by json path

kubectl config view -o jsonpath='{.users[?(@.name == "k8s")].user.password}'

Set credentials for foo.kuberntes.com

kubectl config set-credentials kubeuser/foo.kubernetes.com --username=kubeuser --password=kubepassword

VIEWING, FINDING RESOURCES

List all services in the namespace

kubectl get services

List all pods in all namespaces in wide format

kubectl get pods -o wide --all-namespaces

List all pods in json (or yaml) format

kubectl get pods -o json

Describe resource details (node, pod, svc)

kubectl describe nodes my-node

List services sorted by name

kubectl get services --sort-by=.metadata.name

List pods sorted by restart count

kubectl get pods --sort-by='.status.containerStatuses[0].restartCount'

Rolling update pods for frontend-v1

kubectl rolling-update frontend-v1 -f frontend-v2.json

Scale a replicaset named 'foo' to 3

kubectl scale --replicas=3 rs/foo

Scale a resource specified in "foo.yaml" to 3

kubectl scale --replicas=3 -f foo.yaml

Execute a command in every pod / replica

for i in 0 1; do kubectl exec foo-$i -- sh -c 'echo $(hostname) > /usr/share/nginx/html/index.html'; done

Explain a resource

kubectl explain hpa

kubectl explain svc

Get nodes region and zone

kubectl get nodes

--label-columns failure-domain.beta.kubernetes.io/region,failure-domain.beta.kubernetes.io/zone

Get Arch, OS, Instance type and node type if kops

kubectl get nodes -o wide -L beta.kubernetes.io/arch -L beta.kubernetes.io/os -L beta.kubernetes.io/instance-type -L kops.k8s.io/instancegroupkubectl get nodes -L beta.kubernetes.io/arch -L beta.kubernetes.io/os -L beta.kubernetes.io/instance-type -L kops.k8s.io/instancegroup

Get node version and name only

kubectl get nodes -o custom-columns=NAME:.metadata.name,VER:.status.nodeInfo.kubeletVersion

Get scheduleable nodes

kubectl get nodes --output 'jsonpath={range $.items[*]}{.metadata.name} {.spec.taints[*].effect}{"\n"}{end}' | awk '!/NoSchedule/{print $1}'

Get all deployments nameonly

kubectl get deployment -o=jsonpath={.items[*].metadata.name}

Get one deployment only (first one)

kubectl get deployment -o=jsonpath={.items[0].metadata.name}

Get all pods statuses only

kubectl get pods -o=jsonpath=‘{.items[*].status.phase}’ --all-namespaces

Get pods qos

kubectl get pods --all-namespaces -o custom-columns=NAME:.metadata.name,NAMESPACE:.metadata.namespace,QOS-CLASS:.status.qosClass

Get images running

kubectl get pod -o=jsonpath='{.spec.containers[*].image}' --all-namespaces

kubectl get pod -o=custom-columns=NAME:.metadata.name,IMAGE:.spec.containers[*].image --all-namespaces

Where is my pod running

kubectl get pods -n sock-shop -l name=carts -o wide

Check node/pod usage memory and cpu

kubectl top nodes kubectl top pods

Check health of etcd

kubectl get --raw=/healthz/etcd

Check status of node autoscaler

kubectl describe configmap cluster-autoscaler-status -n kube-system

Get where pods are running from nodenames

kubectl get pod -o=custom-columns=NAME:.metadata.name,STATUS:.status.phase,NODE:.spec.nodeName --all-namespaceskubectl get pod -o=custom-columns=NODE:.spec.nodeName,NAME:.metadata.name --all-namespaces

Example sorting pods by nodeName:

kubectl get pods -o wide --sort-by="{.spec.nodeName}"

Example of getting pods on nodes using label filter:

for n in $(kubectl get nodes -l your_label_key=your_label_value --no-headers | cut -d " " -f1); do 
        kubectl get pods --all-namespaces  --no-headers --field-selector spec.nodeName=${n} 
  done

or by number of restarts

kubectl get pods --sort-by="{.status.containerStatuses[:1].restartCount}"

get containers name inside pod

kubectl get pods prometheus-prometheus-operator-prometheus-0 -o jsonpath='{.spec.containers[*].name}'

get disk space

kubectl exec -i -t prometheus-prometheus-operator-prometheus-0 -c thanos-sidecar -n monitoring -- df -h

Example filtering by nodeName using — template flag:

$ kubectl get nodes  NAME                         STATUS                     AGE
  ip-254-0-90-30.ec2.internal   Ready                      2d
  ip-254-0-90-35.ec2.internal   Ready                      2d
  ip-254-0-90-50.ec2.internal   Ready,SchedulingDisabled   2d
  ip-254-0-91-60.ec2.internal   Ready                      2d
  ip-254-0-91-65.ec2.internal   Ready                      2d
  $ kubectl get pods --template '{{range .items}}{{if eq .spec.nodeName "ip-254-0-90-30.ec2.internal"}}{{.metadata.name}}{{"\n"}}{{end}}}{{end}}'  filebeat-000
  app-0000
  node-exporter-0000
  prometheus-000

Check pods which are not Runnning

kubectl get pods --field-selector=status.phase!=Running --all-namespaces

Sort Nodes by Role, Age and kubelet version

kubectl get nodes --sort-by={.metadata.labels."kubernetes\.io\/role"}kubectl get node --sort-by={.status.nodeInfo.kubeletVersion}watch kubectl get node --sort-by={.status.nodeInfo.kubeletVersion} watch "kubectl get nodes --sort-by={.metadata.labels.\"kubernetes\.io\/role\"}"kubectl get nodes --sort-by=".status.conditions[?(@.reason == 'KubeletReady' )].lastTransitionTime"

Query apiservers

kubectl get --raw=/apis kubectl get --raw=/logs/kube-apiserver.log

Setup a deployment with limits and requests

kubectl run ken-test --image=kenichishibata/docker-curl -i --tty --limits='cpu=50m,memory=128Mi' --requests='cpu=50m,memory=128Mi' kubectl delete deployment ken-test

Get events for an individual resource

kubectl get event --field-selector=involvedObject.name =foo -w

Get apiresources

Check for an api resources available, this should show your crd api endpoints as well

kubectl api-resources kubectl api-versions

Check apiservices added (registered)

kubectl get apiservices.apiregistration.k8s.io kubectl get apiservices.apiregistration.k8s.io v1beta1.metrics.k8s.io -o yaml

Check hpa (maybe because you have custom metrics enabled in prometheus)?

kubectl get hpa kubectl get hpa --all-namespaces kubectl get --raw /apis/metrics.k8s.io

Kube Diff

kubectl alpha diff -h

Run pod shell

apiVersion: v1
kind: Pod
metadata:
  name: testing
  namespace: monitoring
spec:
  serviceAccountName: build-robot
  automountServiceAccountToken: false
  containers:
  - image: alpine:3.2
    command:
      - /bin/sh
      - "-c"
      - "sleep 60m"
    imagePullPolicy: IfNotPresent
    name: alpine
  restartPolicy: Always

kubectl apply -f testing.yaml --namespace=monitoring

kubectl exec -ti testing --namespace=monitoring -- sh

or

kubectl run -it testing --image testing --namespace monitoring /bin/sh

MANAGE RESOURCES

Create one or more resources from your file or folder.

kubectl create -f FILE

Applies a configuration change to a resource from your file.

kubectl apply -f FILE

Deletes one or more Kubernetes resources from a configuration file or directly from resource names.
e.g. Kubectl delete my_pod (destroy the pod on the cluster named my_pod)

kubectl delete (-f FILE | TYPE [PREFIX_NAME | NAME])

Lets you expose a local port to the port of a POD that is running on the Kubernetes cluster. Useful to debug.
- e.g. Kubectl port-forward my_pod 80:3000 (exposes the port 3000 of the pod my_pod on our local port 80)

kubectl port-forward POD [LOCAL_PORT:]REMOTE_PORT

Run a resource in the Kubernetes cluster.
- e.g.kKubectl run -i --tty busybox --image=busybox -- sh
- -> Run a pod as an interactive shell

kubectl run NAME --image=image [--env=”key=value”] [--port=port] [--replicas=replicas]

Get documentation for pod or service

kubectl explain pods,svc

Create resource(s) like pods, services or daemonsets

kubectl create -f ./my-manifest.yaml

Apply a configuration to a resource

kubectl apply -f ./my-manifest.yaml

Start a single instance of Nginx

kubectl run nginx --image=nginx

Create a secret with several keys

cat <<EOF | kubectl create -f -
apiVersion: v1
kind: Secret
metadata:
 name: mysecret
type: Opaque
data:
 password: $(echo "s33msi4" | base64)
 username: $(echo "jane"| base64)
EOF

Delete a resource

kubectl delete -f ./my-manifest.yaml

MONITORING & LOGGING

Kubectl monitoring commands

Kubectl get pods

This command lists pods on the Kubernetes cluster. This command works for all types of Kubernetes resources: pods, services, deployments, cronjobs, events, ingresses, etc. We can also add parameters:

--all-namespaces: List all resources of all namespaces.
-o wide: List all resources with more details.

Kubectl describe pod

The describe command gives a verbose display of the pod unlike the get and basic display. This allows having the events, useful when a pod does not start.

e.g. Kubectl describe pods my-pod.

kubectl logs [-f] POD [-c CONTAINER]

This command displays the logs of your POD. We can add the -c container option when we want to display the logs of a multi-container pod. The -f command displays the output of the logs continuously (stream).

Example: Kubectl logs -f my_pod -c my_app

-> Stream the logs of the container my_app on the my_pod pod.

kubectl top pod POD_NAME --containers

Displays the metrics for a given pod and its containers within a Kubernetes cluster.

Deploy Heapster from Github repository

kubectl create -f deploy/kube-config/standalone/

Show metrics for nodes

kubectl top node

Show metrics for pods

kubectl top pod

Show metrics for a given pod and its containers

kubectl top pod pod_name --containers

Dump pod logs (stdout)

kubectl logs pod_name

Stream pod container logs (stdout, multi-container case)

kubectl logs -f pod_name -c my-container

Files

kubernetes-commands.md

Latest commit

History

kubernetes-commands.md

File metadata and controls