In this section of the Longhorn Basics course, we will cover using Longhorn volumes in Kubernetes, including their types, usage, and backup/restore functionalities.

Using Longhorn Volumes in Kubernetes

Course Agenda

This section is divided into four parts:

  1. Longhorn Volume Types
  2. Using Longhorn Volumes
  3. Accessing the Longhorn UI
  4. Longhorn Backup and Restore

Longhorn Volume Types

Longhorn supports two types of volumes:

  1. RWO (Read Write Once)

    • Default volume type.
    • Can only be mounted by a single pod at a time.
    • Based on iSCSI block storage.

  2. RWX (Read Write Many)

    • Allows multiple pods to mount the volume simultaneously.
    • Based on NFS.

Using Longhorn Volumes

1. Creating a Longhorn Volume

To create a Longhorn volume, use the StorageClass created during Longhorn installation. Here’s an example YAML manifest for creating a PVC:

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: longhorn-pvc
spec:
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 5Gi
  storageClassName: longhorn

2. Attaching a Longhorn Volume to a Pod

Attach the created PVC to a pod by specifying it as a volume. Here’s an example YAML manifest:

apiVersion: v1
kind: Pod
metadata:
  name: longhorn-test-pod
spec:
  containers:
    - name: test-container
      image: busybox
      volumeMounts:
        - mountPath: "/data"
          name: longhorn-volume
  volumes:
    - name: longhorn-volume
      persistentVolumeClaim:
        claimName: longhorn-pvc

3. Mounting and Writing Data to a Longhorn Volume

Once the pod is running, access the pod and write data to the mounted volume path (e.g., /data).

Accessing the Longhorn UI

Longhorn provides a web interface to manage its components. Access the UI using one of these methods:

  1. Rancher UI
  2. kubectl port-forward
  3. Ingress

Important: The Longhorn UI is not exposed outside the cluster by default. If you expose it, ensure it is protected with authentication since the Longhorn UI lacks built-in authentication.

Longhorn Backup and Restore

1. Difference Between Backup and Snapshot

  • Snapshot:

    • A point-in-time copy of a volume.
    • Not a backup; stored on the same disk.
    • Used for rollbacks (e.g., testing and reverting).

  • Backup:

    • A copy of a volume stored externally.
    • Protects against disk/cluster failure.
    • Can restore entire volumes but not individual files.

2. Creating a Backup Target

Longhorn supports the following backup targets:

  • S3: Compatible with most S3 providers (e.g., AWS S3).
  • NFS
  • SMB/CIFS

3. Creating Backup/Snapshot Schedules

Define schedules using crontab syntax. Note:

  • Maximum of 100 snapshots per volume.
  • Snapshots are incremental, affecting performance if overused.

4. Restoring from Backup/Snapshot

  • Snapshots can be restored in-place (overwriting the existing volume).
  • Backups can be restored to new volumes.
  • Longhorn retains the original volume’s name and namespace, allowing easy restoration of deleted PVCs.

Thank You

This concludes the section on using Longhorn volumes in Kubernetes. In the next section, we will delve deeper into advanced Longhorn functionalities.