Harvester "croned/automatic" backup


does anybody knows how I can automate the process of creating backups? in ui there is an click & forget but how can this process done automatically?

Thanks for the support

It’s important to understand that these are really Longhorn snapshots/backups IE storage only backups, no quiesce or memory snapshots.

So with it’s being Longhorn, you just to add the backup target (NFS or S3) and define your schedule.

Go to Advanced → Settings and fill out the target details.

For the schedule part you need to access the longhorn UI (See [FEATURE] Scheduling of VM Backups · Issue #2756 · harvester/harvester · GitHub)

Go Support at the bottom left of the page → Access Embedded Longhorn UI → Recurring Job. Then follow the Longhorn doc Longhorn | Documentation to setup your schedule.

NOTE: Snapshots are local only (great for rolling back an OS upgrade or application issue) with backups being a snapshot that had been offloaded to a external target which is great for DR IE You deleted the wrong VM or lost your primary datacenter and need to restore.

P.S. Backing up a k8s nodes is allowed but it is not recommend as the only restore option for your cluster as it would require a good amount of work to recover a cluster from a VM snapshot. Tho no one ever got fired for too many backups.