Basic Usage¶

This guide covers the essential commands for managing your KubeAuto cluster on a daily basis.

Starting the Cluster¶

vagrant up

If the cluster was previously suspended, this resumes all VMs instantly. If VMs were halted, this performs a cold boot.

Stopping the Cluster¶

Suspend (Recommended for Daily Use)¶

vagrant suspend

• Saves the full VM state (RAM + CPU) to disk
• Fast resume — typically under 10 seconds
• Cluster remains exactly as you left it
• Uses additional disk space for the saved state

Halt (Clean Shutdown)¶

vagrant halt

• Gracefully shuts down all VMs
• Slower restart compared to suspend
• No additional disk usage

Accessing Nodes¶

SSH into the Control Plane¶

vagrant ssh controlplane

SSH into a Specific Worker¶

vagrant ssh node01
vagrant ssh node02

Exit SSH¶

exit

Using kubectl¶

After SSH-ing into the control plane, kubectl is available immediately:

# List all nodes
kubectl get nodes

# Show detailed node info
kubectl get nodes -o wide

# List all pods across all namespaces
kubectl get pods -A

# List system pods
kubectl get pods -n kube-system

# Watch Calico networking status
kubectl get tigerastatus

Managing Individual Nodes¶

You can target specific nodes with any Vagrant command:

# Suspend a specific node
vagrant suspend node01

# Halt a specific node
vagrant halt node02

# Resume a specific node
vagrant up node01

# Re-provision a node (re-runs setup scripts)
vagrant provision node01

Checking Cluster Status¶

After resuming or restarting the cluster:

vagrant ssh controlplane
kubectl get nodes

Deploying the Test Workload¶

KubeAuto includes a built-in test script:

vagrant ssh controlplane
sudo bash /vagrant/scripts/05-test.sh

This deploys:

• An Apache httpd deployment with 2 replicas
• A NodePort service exposing port 80

The script prints the URLs you can access from your host browser.

Quick Reference¶