CNCF Atlas

English

日本語

English

日本語

Appearance

Getting Started ​

Based on the install instructions in the longhorn/longhorn README.md and the official docs (source 9). The latest release at the time of writing is v1.12.0. Commands assume a running Kubernetes cluster and kubectl configured against it.

Prerequisites ​

Per the manager README.md requirements section:

  • A Kubernetes cluster with mount propagation enabled.
  • On every host: iscsiadm/open-iscsi and an NFS client (nfs-common/nfs-utils/nfs-client) installed.
  • A filesystem (ext4 or XFS) that supports the file extents feature for data storage.

Run the official environment check before installing:

bash
curl -sSfL https://raw.githubusercontent.com/longhorn/longhorn/master/scripts/environment_check.sh | bash

Install ​

Apply the bundled deployment manifest. It installs the manager, instance manager, CSI driver, and UI into the longhorn-system namespace.

bash
kubectl create -f https://raw.githubusercontent.com/longhorn/longhorn/master/deploy/longhorn.yaml

A first working setup ​

  1. Wait for the control plane to come up. Every component runs in longhorn-system.

    bash
    kubectl -n longhorn-system get pods --watch

  2. Confirm the default StorageClass was created. The manifest registers longhorn.

    bash
    kubectl get storageclass longhorn

  3. Request a volume with a PersistentVolumeClaim that uses the longhorn class.

    yaml
    apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: longhorn-demo
spec:
  accessModes:
    - ReadWriteOnce
  storageClassName: longhorn
  resources:
    requests:
      storage: 1Gi

  4. Apply it and confirm it binds.

    bash
    kubectl apply -f longhorn-demo.yaml
kubectl get pvc longhorn-demo

Verify it works ​

The claim should reach Bound, and Longhorn should report a healthy Volume CR with its replicas scheduled.

bash
kubectl get pvc longhorn-demo
kubectl -n longhorn-system get volumes.longhorn.io

A healthy volume shows State: attached (once a pod mounts it) or detached with Robustness: healthy before use. You can also open the Longhorn UI service in longhorn-system to see the volume, its engine, and its replicas.

Where to go next ​

For production concerns Longhorn documents separately, see the official docs (source 9): dedicated disks and node configuration, replica count and data-locality tuning (source 12), backups to S3 or NFS, disaster recovery volumes, and the v2 (SPDK) data engine.