k3s v1.34.11

k3s is Kubernetes with the bulk removed: one binary, SQLite instead of etcd by default, legacy and cloud-provider components stripped out, and sensible defaults bundled. It is conformant, so manifests written for any Kubernetes run on it.

Installing

curl -sfL https://get.k3s.io | sh -
sudo k3s kubectl get nodes

That is a working cluster, and it runs on a Raspberry Pi.

mkdir -p ~/.kube
sudo cp /etc/rancher/k3s/k3s.yaml ~/.kube/config
sudo chown "$USER" ~/.kube/config
kubectl get nodes

Adding a node:

sudo cat /var/lib/rancher/k3s/server/node-token      # on the server
curl -sfL https://get.k3s.io | K3S_URL=https://server:6443 K3S_TOKEN=<token> sh -

What is bundled

Traefik as an ingress controller, so an Ingress resource works immediately.

ServiceLB, so a LoadBalancer service binds a port on the node rather than sitting Pending forever, which is what happens on bare-metal Kubernetes without extra components.

local-path-provisioner, so PersistentVolumeClaims get local disk automatically. Fine on one node, and note that node-local storage undermines rescheduling on a cluster.

Do you actually need it

The honest answer for most homelabs is no. Docker Compose runs the same services with a fraction of the concepts, and Podman Quadlet gives you systemd supervision.

Kubernetes earns its complexity when you have several machines and want workloads rescheduled automatically when one fails, or when learning it is itself the goal.

Our k3s guide covers the tradeoff and the minimum set of concepts.

Uninstalling

/usr/local/bin/k3s-uninstall.sh         # server
/usr/local/bin/k3s-agent-uninstall.sh   # agent

Worth knowing before you install, because it removes everything cleanly, which is not true of every Kubernetes distribution.

Verify Your Download

Release binaries on GitHub are published with SHA-256 checksums.