k3s

Lightweight Kubernetes distribution for edge, IoT, and development. Full K8s API in a single binary under 100MB.

66 commands

Browse by Topic

Getting Started

Quick setup and installation

3commands

Installation

Install k3s

12commands

Cluster

Cluster management

13commands

kubectl

kubectl integration

16commands

Configuration

k3s config

12commands

Backup

Backup & restore

10commands

Install k3s

$ curl -sfL https://get.k3s.io | sh -

Install k3s lightweight Kubernetes on Linux

Check node status

$ sudo k3s kubectl get nodes

List cluster nodes using k3s kubectl

Copy kubeconfig

$ sudo cp /etc/rancher/k3s/k3s.yaml ~/.kube/config

Copy k3s kubeconfig for standard kubectl access

Install k3s server

$ curl -sfL https://get.k3s.io | sh -

Install k3s server with defaults.

Install without traefik

$ curl -sfL https://get.k3s.io | sh -s - --disable traefik

Install without built-in ingress.

Install with kubeconfig mode

$ curl -sfL https://get.k3s.io | sh -s - --write-kubeconfig-mode 644

Install with readable kubeconfig.

Uninstall k3s

$ /usr/local/bin/k3s-uninstall.sh

Completely remove k3s server.

Install specific version

$ curl -sfL https://get.k3s.io | INSTALL_K3S_VERSION=v1.28.0+k3s1 sh -

Install specific k3s version.

Install agent node

$ curl -sfL https://get.k3s.io | K3S_URL=https://server:6443 K3S_TOKEN=<token> sh -

Install k3s in agent-only mode joining an existing server.

Install with INSTALL_K3S_EXEC

$ curl -sfL https://get.k3s.io | INSTALL_K3S_EXEC="--disable servicelb --disable traefik" sh -

Pass server flags via INSTALL_K3S_EXEC env var.

Install with embedded etcd

$ curl -sfL https://get.k3s.io | sh -s - server --cluster-init

Initialize HA cluster with embedded etcd datastore.

Install with external database

$ curl -sfL https://get.k3s.io | sh -s - server --datastore-endpoint="postgres://user:pass@host:5432/k3s"

Install using an external PostgreSQL or MySQL datastore.

Air-gap install

$ sudo mkdir -p /var/lib/rancher/k3s/agent/images/ && sudo cp k3s-airgap-images.tar.gz /var/lib/rancher/k3s/agent/images/ && INSTALL_K3S_SKIP_DOWNLOAD=true ./install.sh

Install k3s in an air-gapped environment with pre-downloaded images.

Install with TLS SAN

$ curl -sfL https://get.k3s.io | sh -s - --tls-san my-load-balancer.example.com

Add additional Subject Alternative Name for the TLS certificate.

Join server to HA cluster

$ curl -sfL https://get.k3s.io | sh -s - server --server https://first-server:6443 --token <token>

Join additional server node to an existing HA etcd cluster.

Get node token

$ sudo cat /var/lib/rancher/k3s/server/node-token

Get token to join agents.

Join agent to cluster

$ curl -sfL https://get.k3s.io | K3S_URL=https://server:6443 K3S_TOKEN=xxx sh -

Join worker node to cluster.

Check k3s status

$ sudo systemctl status k3s

Check k3s service status.

Restart k3s

$ sudo systemctl restart k3s

Restart k3s service.

Uninstall agent

$ /usr/local/bin/k3s-agent-uninstall.sh

Remove k3s agent node.

Label a node

$ k3s kubectl label node <node-name> role=worker

Add a label to a cluster node.

Drain a node

$ k3s kubectl drain <node-name> --ignore-daemonsets --delete-emptydir-data

Safely evict all pods from a node before maintenance.

Cordon a node

$ k3s kubectl cordon <node-name>

Mark node as unschedulable to prevent new pods.

Uncordon a node

$ k3s kubectl uncordon <node-name>

Mark node as schedulable again after maintenance.

Reset cluster

$ k3s server --cluster-reset

Reset the cluster to a single-member etcd cluster.

View k3s logs

$ sudo journalctl -u k3s -f

Follow k3s service logs in real time.

Stop k3s service

$ sudo systemctl stop k3s

Stop the k3s server process.

Delete a node

$ k3s kubectl delete node <node-name>

Remove a node from the cluster.

Create config file

$ sudo mkdir -p /etc/rancher/k3s && sudo touch /etc/rancher/k3s/config.yaml

Create k3s configuration file.

View k3s check-config

$ k3s check-config

Check system requirements.

Enable secrets encryption

$ curl -sfL https://get.k3s.io | sh -s - --secrets-encryption

Install with secrets encryption enabled.

Configure private registry

$ sudo nano /etc/rancher/k3s/registries.yaml

Edit registries.yaml to configure private container registries.

Disable multiple components

$ curl -sfL https://get.k3s.io | sh -s - --disable traefik --disable servicelb --disable metrics-server

Install k3s with multiple built-in components disabled.

Set custom data directory

$ curl -sfL https://get.k3s.io | sh -s - --data-dir /opt/k3s

Store k3s data in a custom directory instead of /var/lib/rancher/k3s.

Configure containerd

$ sudo cp /var/lib/rancher/k3s/agent/etc/containerd/config.toml /var/lib/rancher/k3s/agent/etc/containerd/config.toml.tmpl

Create custom containerd config template for k3s.

Set node name

$ curl -sfL https://get.k3s.io | sh -s - --node-name my-node

Install with a custom node name instead of hostname.

Set cluster CIDR

$ curl -sfL https://get.k3s.io | sh -s - --cluster-cidr 10.42.0.0/16 --service-cidr 10.43.0.0/16

Configure custom pod and service CIDR ranges.

Use flannel backend

$ curl -sfL https://get.k3s.io | sh -s - --flannel-backend wireguard-native

Install with WireGuard-native as the flannel backend for encrypted networking.

Set kubelet args

$ curl -sfL https://get.k3s.io | sh -s - --kubelet-arg max-pods=150

Pass custom arguments to the kubelet process.

Disable cloud controller

$ curl -sfL https://get.k3s.io | sh -s - --disable-cloud-controller

Disable the built-in cloud controller manager.

Create etcd snapshot

$ k3s etcd-snapshot save --name backup-$(date +%Y%m%d)

Create manual etcd snapshot.

List snapshots

$ k3s etcd-snapshot ls

List available etcd snapshots.

Restore from snapshot

$ k3s server --cluster-reset --cluster-reset-restore-path=/path/to/snapshot

Restore cluster from snapshot.

Delete snapshot

$ k3s etcd-snapshot delete backup-20240101

Delete an etcd snapshot.

Configure auto snapshots

$ curl -sfL https://get.k3s.io | sh -s - --etcd-snapshot-schedule-cron '0 */6 * * *'

Enable automatic snapshots every 6 hours.

Set snapshot retention

$ curl -sfL https://get.k3s.io | sh -s - --etcd-snapshot-schedule-cron '0 */12 * * *' --etcd-snapshot-retention 10

Configure scheduled snapshots with a retention count of 10.

Snapshot to S3

$ k3s etcd-snapshot save --name s3-backup --s3 --s3-bucket my-bucket --s3-region us-east-1 --s3-access-key <key> --s3-secret-key <secret>

Save an etcd snapshot directly to an S3-compatible bucket.

Restore from S3 snapshot

$ k3s server --cluster-reset --cluster-reset-restore-path=s3-backup --etcd-s3 --etcd-s3-bucket my-bucket --etcd-s3-region us-east-1

Restore cluster from an S3-stored etcd snapshot.

Set snapshot directory

$ curl -sfL https://get.k3s.io | sh -s - --etcd-snapshot-dir /mnt/backups/k3s

Configure a custom directory for storing etcd snapshots.

Backup kubeconfig

$ sudo cp /etc/rancher/k3s/k3s.yaml ~/k3s-kubeconfig-backup.yaml

Create a backup copy of the kubeconfig file.

Discussion

Loading comments...

k3s

Lightweight Kubernetes distribution for edge, IoT, and development. Full K8s API in a single binary under 100MB.

66 commands

Browse by Topic

Getting Started

Quick setup and installation

3commands

Installation

Install k3s

12commands

Cluster

Cluster management

13commands

kubectl

kubectl integration

16commands

Configuration

k3s config

12commands

Backup

Backup & restore

10commands

Install k3s

$ curl -sfL https://get.k3s.io | sh -

Install k3s lightweight Kubernetes on Linux

Check node status

$ sudo k3s kubectl get nodes

List cluster nodes using k3s kubectl

Copy kubeconfig

$ sudo cp /etc/rancher/k3s/k3s.yaml ~/.kube/config

Copy k3s kubeconfig for standard kubectl access

Install k3s server

$ curl -sfL https://get.k3s.io | sh -

Install k3s server with defaults.

Install without traefik

$ curl -sfL https://get.k3s.io | sh -s - --disable traefik

Install without built-in ingress.

Install with kubeconfig mode

$ curl -sfL https://get.k3s.io | sh -s - --write-kubeconfig-mode 644

Install with readable kubeconfig.

Uninstall k3s

$ /usr/local/bin/k3s-uninstall.sh

Completely remove k3s server.

Install specific version

$ curl -sfL https://get.k3s.io | INSTALL_K3S_VERSION=v1.28.0+k3s1 sh -

Install specific k3s version.

Install agent node

$ curl -sfL https://get.k3s.io | K3S_URL=https://server:6443 K3S_TOKEN=<token> sh -

Install k3s in agent-only mode joining an existing server.

Install with INSTALL_K3S_EXEC

$ curl -sfL https://get.k3s.io | INSTALL_K3S_EXEC="--disable servicelb --disable traefik" sh -

Pass server flags via INSTALL_K3S_EXEC env var.

Install with embedded etcd

$ curl -sfL https://get.k3s.io | sh -s - server --cluster-init

Initialize HA cluster with embedded etcd datastore.

Install with external database

$ curl -sfL https://get.k3s.io | sh -s - server --datastore-endpoint="postgres://user:pass@host:5432/k3s"

Install using an external PostgreSQL or MySQL datastore.

Air-gap install

$ sudo mkdir -p /var/lib/rancher/k3s/agent/images/ && sudo cp k3s-airgap-images.tar.gz /var/lib/rancher/k3s/agent/images/ && INSTALL_K3S_SKIP_DOWNLOAD=true ./install.sh

Install k3s in an air-gapped environment with pre-downloaded images.

Install with TLS SAN

$ curl -sfL https://get.k3s.io | sh -s - --tls-san my-load-balancer.example.com

Add additional Subject Alternative Name for the TLS certificate.

Join server to HA cluster

$ curl -sfL https://get.k3s.io | sh -s - server --server https://first-server:6443 --token <token>

Join additional server node to an existing HA etcd cluster.

Get node token

$ sudo cat /var/lib/rancher/k3s/server/node-token

Get token to join agents.

Join agent to cluster

$ curl -sfL https://get.k3s.io | K3S_URL=https://server:6443 K3S_TOKEN=xxx sh -

Join worker node to cluster.

Check k3s status

$ sudo systemctl status k3s

Check k3s service status.

Restart k3s

$ sudo systemctl restart k3s

Restart k3s service.

Uninstall agent

$ /usr/local/bin/k3s-agent-uninstall.sh

Remove k3s agent node.

Label a node

$ k3s kubectl label node <node-name> role=worker

Add a label to a cluster node.

Drain a node

$ k3s kubectl drain <node-name> --ignore-daemonsets --delete-emptydir-data

Safely evict all pods from a node before maintenance.

Cordon a node

$ k3s kubectl cordon <node-name>

Mark node as unschedulable to prevent new pods.

Uncordon a node

$ k3s kubectl uncordon <node-name>

Mark node as schedulable again after maintenance.

Reset cluster

$ k3s server --cluster-reset

Reset the cluster to a single-member etcd cluster.

View k3s logs

$ sudo journalctl -u k3s -f

Follow k3s service logs in real time.

Stop k3s service

$ sudo systemctl stop k3s

Stop the k3s server process.

Delete a node

$ k3s kubectl delete node <node-name>

Remove a node from the cluster.

Create config file

$ sudo mkdir -p /etc/rancher/k3s && sudo touch /etc/rancher/k3s/config.yaml

Create k3s configuration file.

View k3s check-config

$ k3s check-config

Check system requirements.

Enable secrets encryption

$ curl -sfL https://get.k3s.io | sh -s - --secrets-encryption

Install with secrets encryption enabled.

Configure private registry

$ sudo nano /etc/rancher/k3s/registries.yaml

Edit registries.yaml to configure private container registries.

Disable multiple components

$ curl -sfL https://get.k3s.io | sh -s - --disable traefik --disable servicelb --disable metrics-server

Install k3s with multiple built-in components disabled.

Set custom data directory

$ curl -sfL https://get.k3s.io | sh -s - --data-dir /opt/k3s

Store k3s data in a custom directory instead of /var/lib/rancher/k3s.

Configure containerd

$ sudo cp /var/lib/rancher/k3s/agent/etc/containerd/config.toml /var/lib/rancher/k3s/agent/etc/containerd/config.toml.tmpl

Create custom containerd config template for k3s.

Set node name

$ curl -sfL https://get.k3s.io | sh -s - --node-name my-node

Install with a custom node name instead of hostname.

Set cluster CIDR

$ curl -sfL https://get.k3s.io | sh -s - --cluster-cidr 10.42.0.0/16 --service-cidr 10.43.0.0/16

Configure custom pod and service CIDR ranges.

Use flannel backend

$ curl -sfL https://get.k3s.io | sh -s - --flannel-backend wireguard-native

Install with WireGuard-native as the flannel backend for encrypted networking.

Set kubelet args

$ curl -sfL https://get.k3s.io | sh -s - --kubelet-arg max-pods=150

Pass custom arguments to the kubelet process.

Disable cloud controller

$ curl -sfL https://get.k3s.io | sh -s - --disable-cloud-controller

Disable the built-in cloud controller manager.

Create etcd snapshot

$ k3s etcd-snapshot save --name backup-$(date +%Y%m%d)

Create manual etcd snapshot.

List snapshots

$ k3s etcd-snapshot ls

List available etcd snapshots.

Restore from snapshot

$ k3s server --cluster-reset --cluster-reset-restore-path=/path/to/snapshot

Restore cluster from snapshot.

Delete snapshot

$ k3s etcd-snapshot delete backup-20240101

Delete an etcd snapshot.

Configure auto snapshots

$ curl -sfL https://get.k3s.io | sh -s - --etcd-snapshot-schedule-cron '0 */6 * * *'

Enable automatic snapshots every 6 hours.

Set snapshot retention

$ curl -sfL https://get.k3s.io | sh -s - --etcd-snapshot-schedule-cron '0 */12 * * *' --etcd-snapshot-retention 10

Configure scheduled snapshots with a retention count of 10.

Snapshot to S3

$ k3s etcd-snapshot save --name s3-backup --s3 --s3-bucket my-bucket --s3-region us-east-1 --s3-access-key <key> --s3-secret-key <secret>

Save an etcd snapshot directly to an S3-compatible bucket.

Restore from S3 snapshot

$ k3s server --cluster-reset --cluster-reset-restore-path=s3-backup --etcd-s3 --etcd-s3-bucket my-bucket --etcd-s3-region us-east-1

Restore cluster from an S3-stored etcd snapshot.

Set snapshot directory

$ curl -sfL https://get.k3s.io | sh -s - --etcd-snapshot-dir /mnt/backups/k3s

Configure a custom directory for storing etcd snapshots.

Backup kubeconfig

$ sudo cp /etc/rancher/k3s/k3s.yaml ~/k3s-kubeconfig-backup.yaml

Create a backup copy of the kubeconfig file.

Discussion

Loading comments...