Kuberblue#

Kuberblue is the Kubernetes-ready variant of Immutablue, designed for running container orchestration workloads.

Overview#

Kuberblue includes everything needed to run a Kubernetes cluster:

Pre-installed Kubernetes components (kubeadm, kubelet, kubectl)
Container runtime (CRI-O)
Networking components
Automated cluster initialization

Installation#

sudo bootc switch ghcr.io/immutablue/kuberblue:latest
sudo reboot

Configuration#

Kuberblue settings are configured via settings.yaml:

kuberblue:
  install_dir: "/usr/immutablue-build-kuberblue"
  config_dir: "/etc/kuberblue"
  uid: 970

Setting	Default	Description
`.kuberblue.install_dir`	`/usr/immutablue-build-kuberblue`	Build artifacts directory
`.kuberblue.config_dir`	`/etc/kuberblue`	Configuration directory
`.kuberblue.uid`	`970`	System user UID

Cluster Setup#

Initialize a New Cluster#

# Initialize the cluster
kuberblue init

# Get kubeconfig
kuberblue get-config

Join an Existing Cluster#

# Get join command from master node
kuberblue get-join-command

# On worker node
kuberblue join <token> <master-ip>

Commands#

The kuberblue command provides cluster management:

Command	Description
`kuberblue init`	Initialize a new cluster
`kuberblue reset`	Reset/destroy the cluster
`kuberblue get-config`	Copy kubeconfig to user
`kuberblue put-config`	Deploy kubeconfig to cluster
`kuberblue untaint-master`	Allow pods on master node
`kuberblue deploy`	Deploy applications

Justfile Commands#

# Via immutablue command
immutablue kuberblue_init
immutablue kuberblue_reset
immutablue kuberblue_get_config

Systemd Services#

Service	Description
`kuberblue-onboot.service`	Cluster initialization on boot

Directory Structure#

/usr/immutablue-build-kuberblue/    # Build artifacts
/etc/kuberblue/                      # Configuration
/usr/libexec/kuberblue/             # Scripts
    ├── kube_setup/                  # Setup scripts
    │   ├── kube_init.sh
    │   ├── kube_reset.sh
    │   ├── kube_get_config.sh
    │   └── ...
    ├── setup/                       # Boot scripts
    └── variables.sh                 # Configuration variables

Networking#

Kuberblue supports common CNI plugins:

Flannel (default)
Calico
Cilium

Storage#

For persistent storage, consider:

Local path provisioner
NFS provisioner
TrueBlue variant for ZFS-backed storage

Best Practices#

Single-Node Cluster#

For development or small deployments:

# Initialize
kuberblue init

# Allow workloads on master
kuberblue untaint-master

Multi-Node Cluster#

Initialize master node
Get join command from master
Join worker nodes

Resource Limits#

Recommended minimums:

Master: 2 CPU, 4GB RAM
Worker: 2 CPU, 2GB RAM

Troubleshooting#

Cluster won’t initialize#

# Check kubelet status
systemctl status kubelet

# View kubelet logs
journalctl -u kubelet -f

# Reset and retry
kuberblue reset
kuberblue init

Pods stuck in Pending#

# Check node status
kubectl get nodes

# Check pod events
kubectl describe pod <pod-name>