↓Skip to main content

Kubernetes Commands

2927 words·
Kubernetes Kubectl Commands
Table of Contents

Kubectl
#

Kubectl Installation
# 

# Install Kubectl
curl -LO https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/linux/amd64/kubectl &&
chmod +x kubectl &&
sudo mv kubectl /usr/local/bin/

# List kubectl commands
kubectl -h

# List help for specific command: "get" command
kubectl get -h

Kubectl Autocomplete
#

Insall Bash-Completion & Verify Installation
# 

# Search installed packages on Deb based system
dpkg -l | grep bash-completion

# Shell output:
ii  bash-completion                      1:2.11-8                                all          programmable completion for the bash shell

Install the bash-completion package if it’s not yet installed:

# Install bash-completion package
sudo apt update &&
sudo apt install bash-completion -y

Enable Kubectl Autocomplete
#

Current Shell:

# Enable kubectl autocomplete: Current shell
source <(kubectl completion bash)

Permanent:

# Append the command to the .bashrc file
echo "source <(kubectl completion bash)" >> ~/.bashrc

# Apply changes
source ~/.bashrc



Kubeconfig Context
# 

# Switch Kubeconfig context
kubectl config use-context <context-name>

# List current Kubeconfig context
kubectl auth whoami

# List Kubeconfig permissions in specific namespace
kubectl auth can-i --list -n <namespace-name>



Useful Pods
#

Maintenance Pods
# 

# Create netshoot pod
kubectl run netshoot \
  --image=nicolaka/netshoot \
  --restart=Never \
  -it --rm -- /bin/bash

# Utils available inside the pod
ping google.com
nslookup google.com
dig google.com
curl google.com

# Run pod for network troubleshooting 
kubectl run busybox --image=busybox --restart=Never --stdin --tty

# Utils available inside the pod
ping google.com
nslookup google.com

# Delete pod
kubectl delete pod busybox

# Run curl pod
kubectl run curltest --image=curlimages/curl:latest --rm -it --restart=Never -- sh

# Delete pod
kubectl delete pod curltest

Postgres Client Pod
# 

# Create Postgres client pod
kubectl run postgres-client \
  --rm -it \
  --restart=Never \
  --image=postgres:17 \
  -- bash

List Time on Kubernetes Node
# 

# List system time and date
kubectl run time-check \
  --rm \
  --attach \
  --restart=Never \
  --image=busybox:1.36 \
  --command -- date



Access Pod / Container Shell
#

Access Shell
# 

# Access an container shell: Single container pod
kubectl exec -it <pod-name> -- /bin/bash

# Access an container shell: Multi container pod
kubectl exec -it <pod-name> -c <container-name> -- /bin/bash

List Environment Variables
# 

# List environment variables: Single container pod
kubectl exec <pod-name> -- printenv

# List environment variables: Multi container pod
kubectl exec <pod-name> -c <container-name> -- printenv



Pod / Container Logs
# 

# Container logs: Single container pod
kubectl logs <pod-name>

# Container logs: Multi container pod
kubectl logs <pod-name> -c <container-name>

# Container logs: Follow the logs
kubectl logs -f <pod-name>

# Container logs: Define timeframe
kubectl logs <pod-name> --since=30m

# Container logs, previous pod
kubectl logs <pod-name> --previous



Port Forwarding
#

Kubernetes will forward to the correct container based on the destination port:

# Forward container port: Access from other hosts
kubectl -n <namespace-name> port-forward --address 0.0.0.0 <pod-name>  8080:80

# Curl the local port
curl localhost:8080

# Stop port forwarding
Strg + C



List Pods
#

List Pods via Label
# 

# List pods via label
kubectl -n kube-system get pods -l k8s-app=kube-dns

List Pods On Specific Node
# 

# List all pods that run on specific node
kubectl get pods -A -o wide --field-selector spec.nodeName=k8s-master-1

List Pods That Mount A PVC
# 

# List pods that mount a specific PVC
kubectl get pods -n dagster -o jsonpath='{range .items[*]}{.metadata.name}{"\t"}{range .spec.volumes[*]}{.persistentVolumeClaim.claimName}{" "}{end}{"\n"}{end}' | grep <pvc-name>

List Pod Resource Usage
# 

# List pods regarding RAM usage
kubectl -n <namespace> top pod

List Pods / Sort by Resource Usage
# 

# List pods regarding RAM usage
kubectl top pods -A --sort-by=memory

# Shell output:
NAMESPACE                NAME                                                        CPU(cores)   MEMORY(bytes)   
loki                     loki-chunks-cache-0                                         4m           4141Mi          
loki                     loki-ingester-0                                             56m          3229Mi          
loki                     loki-ingester-1                                             64m          2686Mi          
vmks                     vmstorage-vmks-victoria-metrics-k8s-stack-1                 94m          2420Mi          
vmks                     vmstorage-vmks-victoria-metrics-k8s-stack-0                 103m         2311Mi          
grafana                  grafana-8548c57ccc-rssmv                                    37m          1062Mi          

# List pods regarding RAM usage, namespace specific, containers
kubectl top pods -n loki --containers --sort-by=memory

# Shell output:
POD                                     NAME              CPU(cores)   MEMORY(bytes)   
loki-chunks-cache-0                     memcached         28m          4124Mi          
loki-chunks-cache-0                     exporter          3m           17Mi            
loki-ingester-0                         ingester          95m          3293Mi          
loki-ingester-1                         ingester          79m          2890Mi          
loki-ingester-2                         ingester          68m          738Mi           
loki-distributor-54f8484489-hqmcv       distributor       59m          491Mi           
loki-query-frontend-889cdfc8b-lc66q     query-frontend    40m          392Mi           
loki-distributor-54f8484489-xgxlc       distributor       56m          361Mi           
loki-querier-698dc6fb99-qspkw           querier           119m         320Mi           
loki-querier-698dc6fb99-7625j           querier           11m          276Mi           
loki-querier-698dc6fb99-hslgs           querier           125m         148Mi           
loki-distributor-54f8484489-6s877       distributor       52m          99Mi            
loki-query-frontend-889cdfc8b-tqr67     query-frontend    3m           87Mi            
loki-index-gateway-0                    index-gateway     3m           80Mi            
loki-compactor-0                        compactor         10m          71Mi            
loki-index-gateway-1                    index-gateway     2m           53Mi            
loki-query-scheduler-757fd976d9-m2v46   query-scheduler   2m           34Mi            
loki-query-scheduler-757fd976d9-zsh4j   query-scheduler   2m           32Mi            
loki-results-cache-0                    exporter          2m           13Mi            
loki-results-cache-0                    memcached         1m           11Mi



Pod Troubleshooting
#

Delete Pod with Force
# 

# Delete pod with force
kubectl -n <namespace> delete pod <pod-name>  --grace-period=0 --force

Patch Finalizer
# 

# Patch finalizer when resource is stuck in deleting
kubectl patch chi -n <namespace> <pod-name> --type=merge -p '{"metadata":{"finalizers":[]}}'



Scale Deployment
# 

# Set the number of replicas for a deployment 
kubectl scale deployment <deployment-name> --replicas=2

Restart Rollout
# 

# Set the number of replicas for a deployment 
kubectl rollout restart <deployment-name>



Certificates
#

List Certificates
# 

# List certificates
kubectl get certificate -n <namespace>

# List certificate details
kubectl describe certificate <certificate-name>

List Certificate Request
# 

# Check certificate request
kubectl get certificaterequest -n <namespace>
kubectl get certificaterequest -A

# List Cert request details
kubectl describe certificaterequest <certificate-request-name>

# Delete certificate request
kubectl -n <namespace> delete certificaterequest <certificate-request-name>

List Orders
# 

# List Orders
kubectl get orders.acme.cert-manager.io -n <namespace>

# List Orders details
kubectl -n <namespace> describe orders.acme.cert-manager.io <ordername>

# Delete Order
kubectl -n <namespace> delete order <ordername>

List Challenges
# 

# List Challenges
kubectl get challenges.acme.cert-manager.io -n <namespace>

# List Challgenge details
kubectl -n <namespace> describe challenges.acme.cert-manager.io <challenge-name>

Manually Renew Certificate
# 

# Install cmctl
curl -LO https://github.com/cert-manager/cmctl/releases/latest/download/cmctl_linux_amd64
chmod +x cmctl_linux_amd64
sudo mv cmctl_linux_amd64 /usr/local/bin/cmctl

# Verify installation
cmctl version

# Trigger renewal of certificate
cmctl -n <namespace> renew >certificate-name>

# Shell output:
Manually triggered issuance of Certificate <namespace>/<cert-name>

Traces public DNS resolution for the TXT record
# 

# Run test pod
kubectl run -it --rm dns-debug \
  --image=ghcr.io/nicolaka/netshoot \
  --restart=Never -- bash

# Test DNS resolution
dig +trace _acme-challenge.jklug.work TXT

CertManager Logs
# 

# List CertManager logs
kubectl logs -n cert-manager deploy/cert-manager

# List CertManager logs
kubectl logs -n cert-manager deploy/cert-manager --since=24h | grep -iE \
'authoritative|nameserver|self check|propagation|error'

Update CertManager Nameserver
# 

# Update CertManager DNS servers
helm upgrade cert-manager jetstack/cert-manager \
  -n cert-manager \
  --reuse-values \
  --set "extraArgs={--dns01-recursive-nameservers=1.1.1.1:53\,8.8.8.8:53,--dns01-recursive-nameservers-only}"



Nodes
#

List Nodes & Check Cluster Status
# 

# List nodes
kubectl get nodes

# List nodes: List node IP and Kubernetes version
kubectl get nodes -o wide

# List full Node details
kubectl describe node <node-name>

List Node Resource Usage
# 

# List Node resource usage
kubectl top node
  • Divides the usage by the node’s allocatable memory. That is the part Kubernetes lets pods use.
# List Node resource usage
kubectl top node --show-capacity
  • Divides the usage by the node’s capacity, which is the total RAM of the VM.

Label Worker Nodes
# 

# Label a worker node
kubectl label nodes <node-name> kubernetes.io/role=worker

List Node Taints
# 

# List node taints
kubectl describe node <node> | grep -A5 Taints

Kubelet Logs
# 

# List Kubelet logs of a Kubernetes Node: Run on desired node
sudo journalctl -u kubelet

Kubernetes Version
# 

# List Kubernetes version: Client (Kubectl) & Server (Kubernetes Cluster)
kubectl version

Shutdown Node / Maintenance Mode
# 

# drain each node, which safely evicts all pods from the node in preparation for maintenance
kubectl drain node3 --ignore-daemonsets --delete-emptydir-data

# Verify the status
kubectl get nodes

# Shell output
NAME    STATUS                     ROLES           AGE    VERSION
node1   Ready                      control-plane   6d1h   v1.29.5
node2   Ready                      control-plane   6d1h   v1.29.5
node3   Ready,SchedulingDisabled   <none>          6d1h   v1.29.5
node4   Ready,SchedulingDisabled   <none>          6d1h   v1.29.5

# Uncord node / mark it as schedulable
kubectl uncordon node3

# Verify the status
kubectl get nodes

# Shell output
NAME    STATUS   ROLES           AGE    VERSION
node1   Ready    control-plane   6d1h   v1.29.5
node2   Ready    control-plane   6d1h   v1.29.5
node3   Ready    <none>          6d1h   v1.29.5
node4   Ready    <none>          6d1h   v1.29.5

Switch the Default Namespace
#

List Current Cluster
# 

# List the current kubectl context: Cluster & user
kubectl config view | grep current-context

# Shell output:
current-context: kubernetes-admin@k8s.jkw.local

Set new Default Namespace
# 

# Set the namespace "new-namespace" as the default namespace
kubectl config set-context kubernetes-admin@k8s.jkw.local --namespace new-namespace

# Set the namespace "default" as the default namespace
kubectl config set-context kubernetes-admin@k8s.jkw.local --namespace default

Verify the default Namespace
# 

# Verify the configuration
kubectl config view

# Shell output:
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: DATA+OMITTED
    server: https://127.0.0.1:6443
  name: k8s.jkw.local
contexts:
- context:
    cluster: k8s.jkw.local
    namespace: new-namespace # Default namespace
    user: kubernetes-admin
  name: kubernetes-admin@k8s.jkw.local
current-context: kubernetes-admin@k8s.jkw.local
kind: Config
preferences: {}
users:
- name: kubernetes-admin
  user:
    client-certificate-data: DATA+OMITTED
    client-key-data: DATA+OMITTED



LimitRange Resource Limits
#

Create LimitRange Resource
# 

# Create a LimitRange configuration
vi namespace-limits.yaml

apiVersion: v1
kind: LimitRange
metadata:
  name: example-namespace-limits
  namespace: example-namespace
spec:
  limits:
  - type: Pod
    max:
      cpu: "2"
      memory: "1Gi"
    min:
      cpu: "200m"
      memory: "100Mi"
  - type: Container
    default:
      memory: "300Mi"
      cpu: "200m"
    defaultRequest:
      memory: "200Mi"
      cpu: "100m"
    max:
      memory: "600Mi"
      cpu: "400m"
    min:
      memory: "100Mi"
      cpu: "100m"

# Deploy the LimitRange resource
kubectl create -f namespace-limits.yaml

List LimitRange Resources
# 

# List the LimitRange resource for the namespace
kubectl get LimitRange --namespace example-namespace

# Shell output:
NAME                       CREATED AT
example-namespace-limits   2024-06-01T11:33:15Z

List Namespace Resource Limits
# 

# List the namespace details
kubectl describe namespace example-namespace

# Shell output:
Name:         example-namespace
Labels:       kubernetes.io/metadata.name=example-namespace
Annotations:  <none>
Status:       Active

No resource quota.

Resource Limits
 Type       Resource  Min    Max    Default Request  Default Limit  Max Limit/Request Ratio
 ----       --------  ---    ---    ---------------  -------------  -----------------------
 Pod        cpu       200m   2      -                -              -
 Pod        memory    100Mi  1Gi    -                -              -
 Container  memory    100Mi  600Mi  200Mi            300Mi          -
 Container  cpu       100m   400m   100m             200m           -

Delete LimitRange Resource
# 

# Delete the LimitRange resource
kubectl delete LimitRange example-namespace-limits --namespace example-namespace

YAML Resources
#

Apply / Update Resources from YAML Config
# 

# Deploy resources updates from YAML configuration: Current namespace
kubectl apply -f filename.yaml

# Deploy resources updates from YAML configuration: Specific namespace
kubectl apply -f filename.yaml -n <namespace-name>

Delete Resources from YAML Config
# 

# Delete resources from YAML configuration
kubectl delete -f filename.yaml

Create YAML Manifest from Deployment
# 

# Save YAML Manifest from existing deployment
kubectl -n <namespace> get deployment <deployment-name>  -o yaml > <filename>.yaml

Example: Backup & Restore Deployment
#

This is an example for CoreDNS deployment backup and restore:

# Create a backup of a Deployment
kubectl get deployment coredns -n kube-system -o yaml > coredns-deployment.yaml

# Delete the Deployment
kubectl delete deployment coredns -n kube-system

# Restore the Deployment
kubectl apply -f coredns-deployment.yaml



Services
#

Service Types
#

  • ClusterIP Default service type when no service type is defined

  • --type=NodePort Creates a NodePort service

  • --type=LoadBalancer Creates a LoadBalancer service

Access NodePort
# 

NAME          TYPE        CLUSTER-IP     EXTERNAL-IP      PORT(S)          AGE
testservice   NodePort    10.233.13.17   192.168.30.100   8000:32007/TCP   2s

A NodePort service exposes a static port (NodePort) randomly assigned, in this example port “32007” on all nodes external IPs in the cluster, and internal on the cluster IP, in this example on port “8000”. Optional an external IP can be defined.

Create Service
#

Service for Deployment
#

ClusterIP Service:

# Create service type "ClusterIP" for a deployment
kubectl expose deployment <deployment-name> --port=8000 --target-port=80 --name=<service-name>

# Create service type "ClusterIP" for a deployment: Add external IP to access the service
kubectl expose deployment <deployment-name> --port=8000 --target-port=80 --name=<service-name> --external-ip="192.168.30.100"

NodePort Service:

# Create service type "NodePort" for a deployment
kubectl expose deployment <deployment-name> --type=NodePort --port=8000 --target-port=80 --name=<service-name>

# Create service type "NodePort" for a deployment: Add external IP to access the service
kubectl expose deployment <deployment-name> --type=NodePort --port=8000 --target-port=80 --name=<service-name> --external-ip="192.168.30.100"

LoadBalancer Service:

# Create service type "LoadBalancer" for a deployment: Random LoadBalancer IP
kubectl expose deployment <deployment-name> --type=LoadBalancer --port=8000 --target-port=80 --name=<service-name>

# Create service type "LoadBalancer" for a deployment: Define a specific LoadBalancer IP
kubectl expose deployment <deployment-name> --type=LoadBalancer --port=8000 --target-port=80 --name=<service-name> --external-ip="192.168.30.111"

Service for Pod
# 

# Create service type "ClusterIP" for a pod
kubectl expose pod <pod-name> --port=8000 --target-port=80 --name=<service-name>

# Create service type "ClusterIP" for a pod: Add external IP to access the service
kubectl expose pod <pod-name> --port=8000 --target-port=80 --name=<service-name> --external-ip="192.168.30.100"

Service for Replication Controller
# 

# Create service type "ClusterIP" for a Replication Controller
kubectl expose rc <rc-name> --port=8000 --target-port=80 --name=<service-name>

# Create service type "ClusterIP" for a Replication Controller: Add external IP to access the service
kubectl expose rc <rc-name> --port=8000 --target-port=80 --name=<service-name> --external-ip="192.168.30.100"

List Services
# 

# List all services / service details
kubectl get svc

List Service Details
# 

# List service details
kubectl get svc <service-name>

# List more service details
kubectl describe svc <service-name>

Delete Service
# 

# Delete service
kubectl delete service <service-name>

StorageClass, PV, PVC
#

StorageClasses
# 

# List storage clases
kubectl get storageclasses

# Delete storage class
kubectl delete storageclass storageclass-name

Persistent Volume Claims
# 

# List PVC
kubectl get pvc

# Delete PVC
kubectl delete pvc pvc-name

Persistent Volumes
# 

# List PV
kubectl get pv

# Delete PV

kubectl delete pv pv-name



Horizontal Pod Autoscaler (HPA)
#

Deploy HPA
# 

# Deploy a pod autoscaler: Maintains between 1 and 10 pod replicas
kubectl autoscale deployment deployment-name --cpu-percent=50 --min=1 --max=10 --name=hpa-name

List & Watch HPA / Check Status
# 

# List the HPA status: Wait till the HPA gets a target output from the metrics server
kubectl get hpa

# Wacht the load of the HPA
kubectl get hpa hpa-name --watch

List HPA Details
# 

# List HPA details
kubectl describe hpa hpa-name

Delete HPA
# 

# Wacht the HPA
kubectl delete hpa hpa-name

ConfigMaps
# 

# List ConfigMaps
kubectl get cm

# Edit ConfigMap
kubectl edit cm <configmap-name>

# List ConfigMap details
kubectl describe cm <configmap-name>

# Delete ConfigMap
kubectl delete cm <configmap-name>

Secrets
# 

# List secrets
kubectl get secrets

# List secret details
kubectl describe secret <secret-name>

# Delete secret
kubectl delete secret <secret-name>

Cronjobs
#

YAML Configuration Example
# 

# Create cronjob configuration
vi example-conjob.yaml

apiVersion: batch/v1
kind: CronJob
metadata:
  name: example-cronjob
spec:
  schedule: "*/1 * * * *"
  jobTemplate: # Defines the job to be created on each schedule occurrence
    spec:
      template: # The Pod template
        spec:
          containers:
          - name: cronjob-container
            image: busybox
            args:
            - /bin/sh
            - -c
            - date; echo Hello from the Kubernetes cluster
          restartPolicy: OnFailure

# Deploy cronjob
kubectl apply -f example-conjob.yaml

List Cronjobs
#

Example Details:

# List cronjobs
kubectl get cronjobs

# Shell output:
NAME               SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
example-cronjob    */1 * * * *   False     0        16s             40s

  • SUSPEND = False CronJob is active and will trigger executions according to its defined schedule

  • SUSPEND = True CronJob is currently suspended, and no new jobs will be started until it is unsuspended

Command Overview:

# List cronjobs: In current namespace
kubectl get cronjobs

# List cronjobs: In a specific namespace
kubectl get cronjobs -n namespace-name

# List cronjobs: In all namespaces
kubectl get cronjobs --all-namespaces

Suspend & Resume Conjob
#

Example Details:

# Suspend cronjob: Example
kubectl patch cronjob example-cronjob -p '{"spec":{"suspend": true}}'

# Verify the cronjob status
kubectl get cronjobs

# Shell output:
NAME               SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
example-cronjob    */1 * * * *   True      0        43s             67s

# Resume cronjob: Example
kubectl patch cronjob example-cronjob -p '{"spec":{"suspend": false}}'

Command Overview:

# Suspend cronjob: Syntax
kubectl patch cronjob cronjob-name -p '{"spec":{"suspend": true}}'

# Resume cronjob: Syntax
kubectl patch cronjob cronjob-name -p '{"spec":{"suspend": false}}'

Check Job Execution
# 

# List jobs
kubectl get jobs

# Shell output:
NAME                        COMPLETIONS   DURATION   AGE
example-cronjob-28621260   1/1           5s         4m56s
example-cronjob-28621264   1/1           5s         49s

List Cronjob Details
# 

# List cronjobs: In current namespace
kubectl describe cronjob cronjob-name

# List cronjobs: In a specific namespace
kubectl describe cronjob cronjob-name -n namespace-name

Delete Cronjob
# 

# Delete cronjob: In current namespace
kubectl delete cronjob cronjob-name 

# Delete cronjob: In a specific namespace
kubectl delete cronjob cronjob-name -n namespace-name



Utilities
#

K9s
#

Install K9s
#

Find latest release:
https://github.com/derailed/k9s/tags

# Download package (Debian based)
wget https://github.com/derailed/k9s/releases/download/v0.32.5/k9s_linux_amd64.deb

# Install package
sudo dpkg -i k9s_linux_amd64.deb

# Install missing dependencies
sudo apt install -f

Verify K9s Installation
# 

# Verify installation / check version
k9s version

Start K9s
# 

# Start k9s
k9s

# Exit k9s
Strg + C

K9s Commands
# 

# List nodes
:node

# List namespaces
:ns

# List deployments
:deploy

# List pods
:pod

# List services
:svc

# Go back
ESC

# List logs
l



CRICTL ( Container Runtime Interface commandline)
#

List Pods
# 

# List pods
sudo crictl pods

# Shelloutput:
POD ID              CREATED             STATE               NAME                           NAMESPACE           ATTEMPT             RUNTIME
01af3ad77e53f       7 minutes ago       Ready               nginx                          default             0                   (default)
286b4244a3c6d       14 minutes ago      Ready               svclb-traefik-e17c3435-j5gl2   kube-system         2                   (default)
51e1265e2dfa5       15 minutes ago      Ready               loki-stack-promtail-dh7b5      default             2                   (default)
31e1a65143a40       15 minutes ago      Ready               loki-stack-0                   default             2                   (default)

List Containers
# 

# List containers
sudo crictl ps -a

# Shell output:
CONTAINER           IMAGE               CREATED             STATE               NAME                ATTEMPT             POD ID              POD
8fbd470d59191       1e5f3c5b981a9       8 minutes ago       Running             nginx               0                   01af3ad77e53f       nginx
80452d57546a6       edc812b8e25d0       15 minutes ago      Running             lb-tcp-443          2                   286b4244a3c6d       svclb-traefik-e17c3435-j5gl2
e55811e88e9ba       edc812b8e25d0       15 minutes ago      Running             lb-tcp-80           2                   286b4244a3c6d       svclb-traefik-e17c3435-j5gl2
d4e3ff82fb9b7       6860eccd97258       15 minutes ago      Running             promtail            2                   51e1265e2dfa5       loki-stack-promtail-dh7b5
9649420c8fb15       42dd246360519       15 minutes ago      Running             loki                2                   31e1a65143a40       loki-stack-0

List Images
# 

# List images
sudo crictl images
sudo k3s crictl images

# Shell output:
IMAGE                              TAG                 IMAGE ID            SIZE
docker.io/grafana/loki             2.6.1               42dd246360519       20.1MB
docker.io/grafana/promtail         2.9.3               6860eccd97258       75.1MB
docker.io/library/nginx            latest              1e5f3c5b981a9       72.4MB
docker.io/rancher/klipper-lb       v0.4.7              edc812b8e25d0       4.78MB
docker.io/rancher/mirrored-pause   3.6                 6270bb605e12e       301kB

Delete Unused Images
# 

# Delete all unused images
sudo crictl rmi --prune
sudo k3s crictl rmi --prune

# Cronjob
0 5 * * 1 /usr/local/bin/crictl rmi --prune

# If a timeout error like this appears:
E0224 14:13:54.431447 2408301 log.go:32] "RemoveImage from image service failed" err="rpc error: code = DeadlineExceeded desc = context deadline exceeded" image="sha256:4f2ddb5d593f740f63deb711c27181d968147e15315026184f710e346dfb4a0c"

# Just run the prune command several times
sudo k3s crictl rmi --prune

# If necessary restart K3s agent service first
sudo systemctl restart k3s-agent