Restore to a new cluster

Back up the old cluster

  1. Deploy the 1st cluster then install Longhorn system and Velero.
  2. Deploy some workloads using Longhorn volumes then write some data:
    1. A simple pod using multiple volumes. And some volumes are using backing images.
    2. A StatefulSet.
    3. A Deployment with a RWX volume.
  3. Config some recurring policies for the volumes.
  4. Create backups for all volumes.
  5. Create a cluster backup via Velero. 
    velero backup create lh-cluster --exclude-resources persistentvolumes,persistentvolumeclaims,backuptargets.longhorn.io,backupvolumes.longhorn.io,backups.longhorn.io,nodes.longhorn.io,volumes.longhorn.io,engines.longhorn.io,replicas.longhorn.io,backingimagedatasources.longhorn.io,backingimagemanagers.longhorn.io,backingimages.longhorn.io,sharemanagers.longhorn.io,instancemanagers.longhorn.io,engineimages.longhorn.io

Restore to a new cluster

  1. Deploy the 2nd cluster then install Velero only. You can try with different cluster config (more nodes or disks) here.
  2. Restore the cluster backup. e.g., 
    velero restore create --from-backup lh-cluster
  3. Removing all old instance manager pods and backing image manager pods from namespace longhorn-system. Since there is no corresponding InstanceManager CR or BackingImageManager CR for these old pods.
  4. Re-config nodes and disks for the restored Longhorn system if necessary.
  5. Re-create backing images.
  6. Restore all Longhorn volumes from the remote backup target.
  7. Update the access mode to ReadWriteMany since all restored volumes are mode ReadWriteOnce by default.
  8. Create PVCs and PVs with previous names for the restored volumes.
  9. Verify all workloads work fine with correct data.

GitHub issue link: https://github.com/longhorn/longhorn/issues/3367

[Edit]