[QUESTION] Delete snapshot after successful backup ?

[tcoupin]

Question

My volumes are backuped every night using velero and CSI volume snapshot with type: bak.
I do not need to keep snapshots in the cluster.

Is there a way to not keep snapshot for successful backup?

Currently, I'm thinking to add cronjob to execute:

kubectl get backups.longhorn.io -n longhorn-system -o custom-columns=SNAPSHOT:.spec.snapshotName | grep '^snapshot-'| xargs kubectl delete snapshots.longhorn.io -n longhorn-system

Environment

• Longhorn version: 1.4.0
• Kubernetes version: 1.24
• Node config
  • OS type and version ubuntu 22.04
  • CPU per node: 4
  • Memory per node: 8
  • Disk type hdd
  • Network bandwidth and latency between the nodes: 10G
• Underlying Infrastructure (e.g. on AWS/GCE, EKS/GKE, VMWare/KVM, Baremetal): vmware, rke2

Additional context

Add any other context about the problem here.

[weizhe0422]

I think there is an optional parameter: --snapshot-volumes=false to disable creating snapshots when doing Velero backup creation.

ref. https://velero.io/docs/v1.4/how-velero-works/

[R-Studio]

If you change this, then Velero don't trigger a backup of the persistent volumes.

[Pro]

@tcoupin did you already find a good solution for your question?

I have the exact same use-case. I.e., using type: bak with velero and storing the backups in S3 backend.
So I am guessing that the snapshots within Longhorn are actually not needed and could be deleted to save space.

The big question is: will restore actions with velero still work.

[tcoupin]

No, I have scheduled a cronjob to delete useless snapshots (longhorn.io/v1beta2). Velero works with volumesnapshots (snapshot.storage.k8s.io/v1) and the created volumesnapshotcontents is linked to the backups (longhorn.io/v1beta2) so there is no problem to restore pvc.