K9s: The Ultimate Kubernetes Terminal UI for Cluster Management

K9s is a terminal-based UI for managing Kubernetes clusters that provides a fast, efficient alternative to kubectl commands and web dashboards. This comprehensive guide covers everything you need to master K9s for daily Kubernetes operations.

What is K9s?

K9s is an open-source, terminal-based UI that provides:

Key Features

Real-time Monitoring: Live view of cluster resources with auto-refresh
Resource Navigation: Quick navigation between namespaces and resources
Vim-style Shortcuts: Keyboard-driven workflow for power users
Context Switching: Easy switching between clusters and contexts
Built-in Shell: Execute commands directly in pods
Log Streaming: Tail logs from multiple containers
Resource Editing: Edit resources with your preferred editor
Custom Views: Create custom resource views and aliases
Plugins: Extend functionality with custom plugins
Port Forwarding: Manage port forwards from the UI

K9s vs. Other Tools

Feature	K9s	kubectl	Lens	Octant
Interface	Terminal	CLI	Desktop GUI	Web UI
Speed	Very Fast	Fast	Medium	Medium
Resource Usage	Minimal	Minimal	High	Medium
Offline Mode	✅ Yes	✅ Yes	⚠️ Limited	❌ No
Multi-cluster	✅ Easy	✅ Manual	✅ Easy	✅ Easy
Customization	✅ Extensive	⚠️ Limited	⚠️ Limited	⚠️ Limited
Learning Curve	Moderate	Low	Low	Low

Installation

macOS

# Using Homebrew
brew install derailed/k9s/k9s

# Upgrade
brew upgrade k9s

# Using MacPorts
sudo port install k9s

Linux

# Download latest release
wget https://github.com/derailed/k9s/releases/download/v0.31.9/k9s_Linux_amd64.tar.gz

# Extract
tar -xzf k9s_Linux_amd64.tar.gz

# Move to PATH
sudo mv k9s /usr/local/bin/

# Verify installation
k9s version

# Using package managers
# Arch Linux
yay -S k9s

# Ubuntu/Debian (via snap)
sudo snap install k9s

# Using go
go install github.com/derailed/k9s@latest

Windows

# Using Scoop
scoop install k9s

# Using Chocolatey
choco install k9s

# Manual installation
# Download from GitHub releases
# Extract and add to PATH

Docker

# Run K9s in Docker
docker run --rm -it \
  -v ~/.kube/config:/root/.kube/config \
  derailed/k9s

Getting Started

First Launch

# Launch K9s
k9s

# Launch with specific context
k9s --context production-cluster

# Launch with specific namespace
k9s -n production

# Launch in read-only mode
k9s --readonly

# Launch with specific kubeconfig
k9s --kubeconfig ~/.kube/custom-config

Navigation Keys:
- :pods          → View pods
- :deployments   → View deployments
- :services      → View services
- :namespaces    → View namespaces
- :nodes         → View nodes
- Ctrl-a         → Show all available resources
- /              → Filter resources
- Esc            → Clear filter/go back
- Ctrl-d         → Delete resource
- Enter          → View resource details
- l              → View logs
- s              → Shell into container
- Ctrl-c         → Exit K9s

Resource Navigation:
- j/k or ↓/↑     → Move down/up
- g              → Go to top
- Shift-g        → Go to bottom
- Ctrl-f         → Page down
- Ctrl-b         → Page up
- Space          → Mark resource
- Ctrl-space     → Mark all resources

Essential Commands

Viewing Resources

# In K9s command mode (press :)
:pods              # List pods
:po                # Short form
:deploy            # Deployments
:svc               # Services
:ing               # Ingresses
:cm                # ConfigMaps
:secret            # Secrets
:pvc               # PersistentVolumeClaims
:nodes             # Nodes
:ns                # Namespaces
:events            # Events
:jobs              # Jobs
:cronjobs          # CronJobs
:hpa               # Horizontal Pod Autoscalers
:psp               # Pod Security Policies

# View all resources
Ctrl-a

# View custom resources
:certificates      # cert-manager
:issuers          # cert-manager
:virtualservices  # Istio

Working with Pods

# In pod view
Enter         # Describe pod
l             # View logs
Ctrl-l        # Toggle log auto-scroll
p             # Previous logs (from restarted containers)
s             # Shell into container
a             # Attach to container
f             # Port-forward
Ctrl-k        # Kill pod (delete)
y             # View YAML
e             # Edit pod (not common, usually edit deployment)
Shift-f       # Toggle show/hide failed pods

# Multi-container pods
l → Tab       # Switch between container logs
s → Select    # Choose container for shell

Working with Logs

# Log viewing
l              # View logs
Ctrl-l         # Toggle auto-scroll
w              # Toggle log wrap
t              # Toggle timestamp
0-9            # Container selection (multi-container pods)
f              # Toggle fullscreen
c              # Clear logs
/              # Search in logs (press n for next, N for previous)
s              # Save logs to file

# Advanced log options
Shift-l        # View logs from previous container instance
Ctrl-r         # Toggle raw logs (no formatting)

Filtering and Searching

# Filter current view
/pattern       # Filter by pattern
Esc            # Clear filter

# Label filters
/app=nginx     # Filter by label
/!app=nginx    # Inverse filter

# Status filters
/Running       # Show only running pods
/Pending       # Show pending pods
/Error         # Show pods with errors

# Regex filters
/^prod-.*      # Pods starting with "prod-"
/.*-db$        # Pods ending with "-db"

# Multiple filters
/label1=value1,label2=value2

Advanced Features

Context and Namespace Management

# Switch context
:ctx           # List contexts
Enter          # Switch to selected context

# Switch namespace
:ns            # List namespaces
Enter          # Switch to namespace
0              # Switch to all namespaces

# Quick namespace switch
:ns/production  # Jump directly to namespace

Resource Operations

# Describe resource
d              # Describe selected resource
Enter          # Alternative to describe

# Edit resource
e              # Edit with default editor
v              # Edit with vim explicitly

# Delete resource
Ctrl-d         # Delete selected resource
Ctrl-k         # Alternative delete

# Scale deployment
:deploy → Select → s → Enter new count

# Restart deployment
:deploy → Select → Ctrl-r

# Cordon/Drain nodes
:nodes → Select → Ctrl-c  # Cordon
:nodes → Select → Ctrl-o  # Uncordon
:nodes → Select → Ctrl-d  # Drain

Port Forwarding

# From pod view
f              # Port forward
# Enter: local_port:container_port
# Example: 8080:80

# List active port forwards
:pf

# Stop port forward
:pf → Select → Ctrl-d

# Port forward to service
:svc → Select → f

Shell Access

# Shell into pod
s              # Shell access
# For multi-container pods, select container

# Common shell scenarios
:deploy → Select → s → Select pod → Select container

# Execute command
Ctrl-s         # Run command in container

# Attach to container
a              # Attach (for debugging)

Configuration

Config File Location

# K9s config location
~/.config/k9s/config.yml       # Linux/macOS
%APPDATA%\k9s\config.yml       # Windows

# View current config
cat ~/.config/k9s/config.yml

Custom Configuration

k9s:
  # Refresh rate (milliseconds)
  refreshRate: 2000

  # Max log buffer lines
  maxConnRetry: 5

  # Enable mouse support
  enableMouse: false

  # Headless mode
  headless: false

  # Logo
  logoless: false

  # Crumbsless
  crumbsless: false

  # Read-only mode
  readOnly: false

  # No exit on error
  noExitOnError: false

  # UI settings
  ui:
    enableMouse: false
    headless: false
    logoless: false
    crumbsless: false
    noIcons: false

  # Skip latest version check
  skipLatestRevCheck: false

  # Disable pod counts
  disablePodCounting: false

  # Shell pod settings
  shellPod:
    image: busybox:1.35.0
    command: []
    args: []
    namespace: default
    limits:
      cpu: 100m
      memory: 100Mi

  # Image scan settings
  imageScans:
    enable: true
    exclusions:
      namespaces: []
      labels: {}

  # Logger settings
  logger:
    tail: 100
    buffer: 5000
    sinceSeconds: -1
    fullScreen: false
    textWrap: false
    showTime: false

  # Thresholds
  thresholds:
    cpu:
      critical: 90
      warn: 70
    memory:
      critical: 90
      warn: 70

Skin/Theme Configuration

k9s:
  # General
  body:
    fgColor: dodgerblue
    bgColor: '#1e1e1e'
    logoColor: blue

  # Frames
  frame:
    border:
      fgColor: dodgerblue
      focusColor: lightskyblue
    menu:
      fgColor: white
      keyColor: pink
      numKeyColor: fuchsia
    crumbs:
      fgColor: white
      bgColor: steelblue
      activeColor: skyblue
    status:
      newColor: lightskyblue
      modifyColor: greenyellow
      addColor: lightskyblue
      errorColor: red
      highlightColor: orange
      killColor: mediumpurple
      completedColor: gray
    title:
      fgColor: white
      bgColor: steelblue
      highlightColor: orange
      counterColor: slateblue
      filterColor: slategray

  # Views
  views:
    table:
      fgColor: lightskyblue
      bgColor: default
      cursorColor: steelblue
      markColor: darkgoldenrod
    xray:
      fgColor: lightskyblue
      bgColor: default
      cursorColor: steelblue
      cursorTextColor: black
      graphicColor: darkgoldenrod
    yaml:
      keyColor: steelblue
      colonColor: white
      valueColor: lightskyblue
    logs:
      fgColor: lightskyblue
      bgColor: default

Aliases

alias:
  # Resource aliases
  dp: v1/deployments
  sec: v1/secrets
  ing: extensions/v1beta1/ingresses
  svc: v1/services

  # Custom shortcuts
  pp: v1/pods
  pvc: v1/persistentvolumeclaims
  pv: v1/persistentvolumes

  # Helm
  hr: helm.toolkit.fluxcd.io/v2beta1/helmreleases

  # Cert-manager
  cert: cert-manager.io/v1/certificates
  issuer: cert-manager.io/v1/issuers

  # Istio
  vs: networking.istio.io/v1beta1/virtualservices
  dr: networking.istio.io/v1beta1/destinationrules
  gw: networking.istio.io/v1beta1/gateways

Plugins

Plugin Configuration

plugin:
  # Debug with ephemeral container
  debug:
    shortCut: Shift-D
    description: Debug
    scopes:
    - pods
    command: kubectl
    background: false
    args:
    - debug
    - -it
    - -n
    - $NAMESPACE
    - $NAME
    - --image=nicolaka/netshoot
    - --target=$NAME
    - -- bash

  # Dive into container image
  dive:
    shortCut: d
    description: Dive image
    scopes:
    - containers
    command: dive
    background: false
    args:
    - $COL-IMAGE

  # Stern for logs
  stern:
    shortCut: Ctrl-L
    description: Stern logs
    scopes:
    - pods
    command: stern
    background: false
    args:
    - --tail
    - 50
    - -n
    - $NAMESPACE
    - $NAME

  # Get pod logs since time
  logs-since:
    shortCut: Shift-L
    description: Logs 1h
    scopes:
    - pods
    command: kubectl
    background: false
    args:
    - logs
    - -n
    - $NAMESPACE
    - $NAME
    - --since=1h

  # Top pod
  top:
    shortCut: Shift-T
    description: Top pod
    scopes:
    - pods
    command: kubectl
    background: false
    args:
    - top
    - pod
    - -n
    - $NAMESPACE
    - $NAME

  # Restart deployment
  restart:
    shortCut: Shift-R
    description: Restart
    scopes:
    - deployments
    command: kubectl
    background: false
    confirm: true
    args:
    - rollout
    - restart
    - deployment
    - -n
    - $NAMESPACE
    - $NAME

  # Describe node
  node-describe:
    shortCut: Shift-D
    description: Describe
    scopes:
    - nodes
    command: kubectl
    background: false
    args:
    - describe
    - node
    - $NAME

  # Flux reconcile
  flux-reconcile:
    shortCut: Shift-R
    description: Flux reconcile
    scopes:
    - helmreleases
    command: flux
    background: false
    args:
    - reconcile
    - helmrelease
    - -n
    - $NAMESPACE
    - $NAME

Hotkeys and Shortcuts

Global Shortcuts

? or Ctrl-h      → Show help
Ctrl-a           → Show all resources
Ctrl-c           → Exit K9s
:                → Command mode
/                → Filter mode
Esc              → Back/Clear
Ctrl-r           → Refresh
Ctrl-d           → Delete resource
Ctrl-k           → Kill pod
u                → Show UID column

View-Specific Shortcuts

Pods:
l                → Logs
p                → Previous logs
s                → Shell
a                → Attach
f                → Port forward
Shift-f          → Show/hide failed pods
Ctrl-k           → Kill pod
y                → View YAML

Deployments:
s                → Scale
Ctrl-r           → Restart rollout
r                → Rollout status

Nodes:
Ctrl-c           → Cordon
Ctrl-o           → Uncordon
Ctrl-d           → Drain
y                → View YAML

Logs:
w                → Toggle wrap
t                → Toggle timestamp
f                → Toggle fullscreen
c                → Clear logs
s                → Save logs
0-9              → Container selection

Productivity Tips

Quick Resource Access

# Use command mode for fast access
:po                    # Pods
:po -n production     # Pods in namespace
:deploy/nginx         # Specific deployment
:svc/api              # Specific service

# Use context/namespace combo
:ctx/prod → :ns/api → :po

Monitoring Multiple Resources

# Open multiple K9s instances
# Terminal 1: Monitor pods
k9s -n production

# Terminal 2: Monitor nodes
k9s
:nodes

# Terminal 3: Monitor events
k9s
:events

# Use tmux/screen for split view
tmux
Ctrl-b %              # Split vertical
k9s                   # First pane
Ctrl-b o              # Switch pane
k9s -n production     # Second pane

Common Workflows

# Debug failing pod
:po → /status=Error → Enter → l → s

# Scale deployment
:deploy → /app=api → s → Enter new count

# Check resource usage
:nodes → Shift-t
:po → Select → Shift-t

# Port forward to service
:svc → /app=database → f → 5432:5432

# View all resources in namespace
:ns/production → Ctrl-a

# Quick log search
:po → l → /ERROR → n (next) → N (previous)

Troubleshooting

Common Issues

# K9s won't start
k9s --version                    # Check version
k9s info                         # Show diagnostics
rm -rf ~/.config/k9s/            # Reset config
rm -rf ~/.k9s/                   # Remove cache

# Performance issues
# Edit config to increase refresh rate
refreshRate: 5000                # 5 seconds instead of 2

# Disable pod counting
disablePodCounting: true

# Connection issues
kubectl cluster-info             # Verify cluster access
k9s --context different-context  # Try different context

# RBAC issues
# Check permissions
kubectl auth can-i list pods --all-namespaces

# Use read-only mode if needed
k9s --readonly

Debug Mode

# Enable debug logging
k9s --write --logLevel debug

# View logs
tail -f /tmp/k9s-*.log

Production Best Practices

Security

# Use read-only mode in production
k9s --readonly

# Limit access with RBAC
# Create ServiceAccount with limited permissions
kubectl create sa k9s-viewer
kubectl create clusterrolebinding k9s-viewer \
  --clusterrole=view \
  --serviceaccount=default:k9s-viewer

# Use that context with K9s
k9s --context k9s-viewer-context

Configuration Management

# Version control K9s config
cd ~/.config/k9s
git init
git add config.yml skin.yml alias.yml plugin.yml
git commit -m "K9s configuration"

# Share config across team
git remote add origin git@github.com:company/k9s-config.git
git push -u origin main

Team Standards

# Shared team config
k9s:
  refreshRate: 2000
  readOnly: false  # Set to true for prod

  # Consistent shell pod
  shellPod:
    image: company/debug-tools:latest
    namespace: default

  # Standard thresholds
  thresholds:
    cpu:
      critical: 90
      warn: 70
    memory:
      critical: 90
      warn: 70

Comparison with kubectl

# kubectl commands → K9s equivalents

# List pods
kubectl get pods                → :po
kubectl get pods -n prod        → :ns/prod → :po
kubectl get pods -w             → :po (auto-refresh)

# Describe
kubectl describe pod nginx      → :po → /nginx → Enter

# Logs
kubectl logs nginx              → :po → /nginx → l
kubectl logs -f nginx           → :po → /nginx → l (auto-tail)

# Delete
kubectl delete pod nginx        → :po → /nginx → Ctrl-d

# Shell
kubectl exec -it nginx -- sh    → :po → /nginx → s

# Port forward
kubectl port-forward svc/api    → :svc → /api → f

# Edit
kubectl edit deploy nginx       → :deploy → /nginx → e

# Scale
kubectl scale deploy nginx --replicas=3  → :deploy → /nginx → s

Conclusion

K9s transforms Kubernetes cluster management with its efficient, keyboard-driven interface. By mastering K9s shortcuts, plugins, and configuration options, you can dramatically improve your productivity when working with Kubernetes clusters.

The tool’s real-time monitoring, intuitive navigation, and extensibility make it an essential tool for anyone managing Kubernetes clusters, from development to production environments.

Master Kubernetes tools and best practices with our comprehensive Kubernetes training programs. Contact us for customized training designed for your team’s needs.

K9s: The Ultimate Kubernetes Terminal UI for Cluster Management

K9s: The Ultimate Kubernetes Terminal UI for Cluster Management

What is K9s?

Key Features

K9s vs. Other Tools

Installation

macOS

Linux

Windows

Docker

Getting Started

First Launch

Basic Navigation

Essential Commands

Viewing Resources

Working with Pods

Working with Logs

Filtering and Searching

Advanced Features

Context and Namespace Management

Resource Operations

Port Forwarding

Shell Access

Configuration

Config File Location

Custom Configuration

Skin/Theme Configuration

Aliases

Plugins

Plugin Configuration

Hotkeys and Shortcuts

Global Shortcuts

View-Specific Shortcuts

Productivity Tips

Quick Resource Access

Monitoring Multiple Resources

Common Workflows

Troubleshooting

Common Issues

Debug Mode

Production Best Practices

Security

Configuration Management

Team Standards

Comparison with kubectl

Conclusion