Skip to content

Instantly share code, notes, and snippets.

@pydevops
Last active August 27, 2024 01:26
Show Gist options
  • Save pydevops/0efd399befd960b5eb18d40adb68ef83 to your computer and use it in GitHub Desktop.
Save pydevops/0efd399befd960b5eb18d40adb68ef83 to your computer and use it in GitHub Desktop.
k8s kubectl cheat sheet

references

cheatsheet & tips

cool gear to have

JSONAPTH

# loop with range
# list pod's name
k get po -o jsonpath='{range .items[*]}{.metadata.name}{"\n"}{end}'
k get po -o jsonpath={.items..metadata.name}
# list node names and cpu capacity
k get nodes -o jsonpath='{range .items[*]}{.metadata.name}{"\t"}{.status.capacity.cpu}{"\n"}{end}'

# list image
kubectl get pods -o jsonpath="{..image}" | tr -s '[[:space:]]' '\n'  | sort | uniq
kubectl get pods -o jsonpath="{.items[*].spec.containers[*].image}"
kubectl get pods -o jsonpath='{.items[*].status.podIP}' 
kubectl get pods -o jsonpath='{range .items[*]}{"\n"}{.metadata.name}{":\t"}{range .spec.containers[*]}{.image}{", "}{end}{end}'

kubectl get svc -l component=elasticsearch,role=client -o jsonpath='{..ip}'
grace=$(kubectl get po cassandra-0 -o=jsonpath=‘{.spec.terminationGracePeriodSeconds}’) 
grace=$(kubectl get sts -l component=elasticsearch,role=data -o jsonpath='{..terminationGracePeriodSeconds}'

# list node instance type, zone, ami
echo "***list karpenter nodes ami"
kubectl get nodes -L karpenter.k8s.aws/instance-ami-id -l karpenter.sh/provisioner-name=default -o custom-columns="Name:.metadata.name,InstanceType:.metadata.labels.node\.kubernetes\.io/instance-type,\
Zone:.metadata.labels.topology\.kubernetes\.io/zone,Ami:.metadata.labels.karpenter\.k8s\.aws/instance-ami-id"

custom-columns

# same query  in jsonpath
k get po -o custom-columns='POD_NAME:metadata.name'
# same query as in jsonpath for node and cpu counts
k get nodes -o custom-columns="NODE:.metadata.name,CPU:.status.capacity.cpu"

# node name
k get nodes -o custom-columns=NAME:.metadata.name
# pod name
k get po -o custom-columns=NAME:.metadata.name
# image name
k get po -o custom-columns='IMAGE:spec.containers[*].image'

# list container image and k8s-app lable value in kube-system
k get deployment -o custom-columns='IMAGE:.spec.template.spec.containers[*].image,LABEL:.spec.template.metadata.labels.k8s-app' -n kube-system

sort-by

# implict range or items[*]

k get nodes --sort-by=".metadata.name"
k get nodes --sort-by=".status.capacity.cpu"

k get po --sort-by=.spec.nodeName -o wide
k get po --sort-by=".metadata.creationTimestamp"
k get pv --sort-by=.spec.capacity.storage -o custom-columns="NAME:.metadata.name,CAPACITY:.spec.capacity.storage"

clean up pods

Did this for cleaning up pods with not in Running state such as Terminated

k get po --field-selector=status.phase!=Running -o custom-columns=":metadata.name" --no-headers | xargs kubectl delete po

mysql and psql

kubectl run -n default mysql-client-${USER} --image=mysql:5.7 -it --restart=Never -- /bin/bash
kubectl run -n default psql-cli-${USER} --image=postgres -it  --restart=Never -- bash

wait for

kubectl -n istio-system wait --for=jsonpath='{.data.ca\.crt}' secrets/cacerts

debug

kubectl run -it --rm debug --image=busybox -- sh

cert compare

kubectl get validatingwebhookconfigurations.admissionregistration.k8s.io aws-load-balancer-webhook -ojsonpath={.webhooks[0].clientConfig.caBundle}  | base64 -d  | openssl x509 -noout -text
kubectl get secret -n kube-system  aws-load-balancer-tls -ojsonpath="{.data.ca\.crt}" |base64 -d   |openssl x509 -noout -text

context, namespace

 get current context: kubectl config view -o=jsonpath='{.current-context}'
 get all contexts:  kubectl config get-contexts -o=name | sort -n
 get namesapce:  kubectl get namespaces -o=jsonpath='{range .items[*].metadata.name}{@}{"\n"}{end}'
 
kubectl config use-context <cluster_name_in_kubeconfig>
kubectl --context <context>

## set the namespace for the current context
kubectl config set-context gke_sandbox-co_us-west1-a_cka --namespace=kube-system
kubectl config set-context --current --namespace=kube-system

API

# Print the supported API group and its versions on the server, in the form of "group/version"
k api-versions | sort 

# list api-resources with sorting
kubectl api-resources --sort-by=name 
kubectl api-resources --sort-by=kind

# find out what is under the api group

k api-resources --api-group=networking.k8s.io
NAME              SHORTNAMES   APIVERSION             NAMESPACED   KIND
ingressclasses                 networking.k8s.io/v1   false        IngressClass
ingresses         ing          networking.k8s.io/v1   true         Ingress
networkpolicies   netpol       networking.k8s.io/v1   true         NetworkPolicy

# then we can explain with $APIVERSION
k explain --api-version=$APIVERSION ingress --recursive
k explain --api-version=apps/v1 deployment --recursive

# for each "group/version" in the output above except for "api/v1"
kubectl get --raw /apis/${group/version} |  jq -r '.resources[].kind'

kubectl get --raw /apis/apps/v1 | jq . -C | less -R

list resources under a specific api version.

This is due to API deprecations

kubectl get deployments.v1.apps

secret

echo $(kubectl get secret/terraform -o jsonpath="{.data['terraform\.json']}" | base64 --decode)

Play with jid and jq

Get the TCP LB port and IP

  EXT_IP="$(kubectl get svc hello-server -o=jsonpath='{.status.loadBalancer.ingress[0].ip}')"
  EXT_PORT=$(kubectl --namespace default get service hello-server -o=jsonpath='{.spec.ports[0].port}')
  echo "$EXT_IP:$EXT_PORT"
  [ "$(curl -s -o /dev/null -w '%{http_code}' "$EXT_IP:$EXT_PORT"/)" -eq 200 ] || exit 1

deployment

rollout

kubectl rollout pause deployment/hello
kubectl rollout status deployment/hello
# check the versions on pods
kubectl get pods -o jsonpath --template='{range .items[*]}{.metadata.name}{"\t"}{"\t"}{.spec.containers[0].image}{"\n"}{end}'
kubectl rollout resume deployment/hello
# roll back
kubectl rollout undo deployment/hello

rbac

# list what a sa 's rbac
k auth can-i --list  --as system:serviceaccount:datadog:datadog 
k auth can-i get crd --as system:serviceaccount:velero:velero
k auth can-i '*' '*' --as system:serviceaccount:default:remote-admin-sa --all-namespaces
# list what I can do
k auth can-i get crd
k auth can-i '*' '*' --all-namespaces

# with krew plugins

## check out rbac roles for a given user/group,sa

## first find out what we have 
k rbac-lookup -k user
k rbac-lookup -k group
k rbac-lookup -k serviceaccount
# then find out what velero can do
k rbac-lookup velero -o wide

# from resource perspective
k who-can list '*'
k who-can create customresourcedefinition

## access matrix for user/group,sa
k access-matrix --sa default:deployer
k access-matrix --sa kube-system:kube-state-metrics

find top resource hungry pod

# pod sort by cpu
k top pods --sort-by=cpu --no-headers 
# container sort by memory
k top pods --containers --sort-by=memory
kubectl top pods -A --no-headers | sort -rn -k 3
# memory
kubectl top pods -A --no-headers | sort -rn -k 4
# top 1
kubectl top pod  --no-headers | grep -v NAME | sort -k 3 -nr | awk -F ' ' 'NR==1{print $1}'

metrics

# all nodes
kubectl get --raw /apis/metrics.k8s.io/v1beta1/nodes | jq -C . | less -R
# individual node 
kubectl get --raw /apis/metrics.k8s.io/v1beta1/nodes/$NODE_NAME

# all pods
kubectl get --raw /apis/metrics.k8s.io/v1beta1/pods | jq . -C | less -R
# individual pod
kubectl get --raw /apis/metrics.k8s.io/v1beta1/namespaces/$NS/pods/$POD

# jq
kubectl get --raw /apis/metrics.k8s.io/v1beta1/nodes \
| jq '[.items [] | {nodeName: .metadata.name, nodeCpu: .usage.cpu, nodeMemory: .usage.memory}]'

kubectl get --raw /apis/metrics.k8s.io/v1beta1/pods | jq . -C | less -R
@pydevops
Copy link
Author

Remove argocd apps finalizer script

#!/usr/bin/env bash
APPS=$(kubectl -n argocd get app -o jsonpath='{range .items[*]}{.metadata.name}{"\n"}{end}')
for app in $APPS
do
 echo "patch $app 's finalizer"
 kubectl patch app/$app \
    --type json \
    --patch='[ { "op": "remove", "path": "/metadata/finalizers" } ]'
done

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment