Snapshot Restore API

The snapshot.kubevirt.io API Group defines resources for snapshotting and restoring KubeVirt VirtualMachines

Prerequesites

VolumeSnapshotClass

KubeVirt leverages the VolumeSnapshot functionality of Kubernetes CSI drivers for capturing persistent VirtualMachine state. So, you should make sure that your VirtualMachine uses DataVolumes or PersistentVolumeClaims backed by a StorageClass that supports VolumeSnapshots and a VolumeSnapshotClass is properly configured for that StorageClass.

KubeVirt looks for Kubernetes Volume Snapshot related APIs/resources in the v1 version. To make sure that KubeVirt’s snapshot controller is able to snapshot the VirtualMachine and referenced volumes as expected, Kubernetes Volume Snapshot APIs must be served from v1 version.

To list VolumeSnapshotClasses:

  1. kubectl get volumesnapshotclass

Make sure the provisioner property of your StorageClass matches the driver property of the VolumeSnapshotClass

Even if you have no VolumeSnapshotClasses in your cluster, VirtualMachineSnapshots are not totally useless. They will still backup your VirtualMachine configuration.

Snapshot Feature Gate

Snapshot/Restore support must be enabled in the feature gates to be supported. The feature gates field in the KubeVirt CR must be expanded by adding the Snapshot to it.

Snapshot a VirtualMachine

Snapshotting a virtualMachine is supported for online and offline vms.

When snapshotting a running vm the controller will check for qemu guest agent in the vm. If the agent exists it will freeze the vm filesystems before taking the snapshot and unfreeze after the snapshot. It is recommended to take online snapshots with the guest agent for a better snapshot, if not present a best effort snapshot will be taken.

Note To check if your vm has a qemu-guest-agent check for ‘AgentConnected’ in the vm status.

There will be an indication in the vmSnapshot status if the snapshot was taken online and with or without guest agent participation.

Note Online snapshot with hotplugged disks is supported, only persistent hotplugged disks will be included in the snapshot.

To snapshot a VirtualMachine named larry, apply the following yaml.

  1. apiVersion: snapshot.kubevirt.io/v1alpha1
  2. kind: VirtualMachineSnapshot
  3. metadata:
  4.   name: snap-larry
  5. spec:
  6.   source:
  7.     apiGroup: kubevirt.io
  8.     kind: VirtualMachine
  9.     name: larry

To wait for a snapshot to complete, execute:

  1. kubectl wait vmsnapshot snap-larry --for condition=Ready

You can check the vmSnapshot phase in the vmSnapshot status. It can be one of the following:

  • InProgress
  • Succeeded
  • Failed.

The vmSnapshot has a default deadline of 5 minutes. If the vmSnapshot has not succeessfully completed before the deadline, it will be marked as Failed. The VM will be unfrozen and the created snapshot content will be cleaned up if necessary. The vmSnapshot object will remain in Failed state until deleted by the user. To change the default deadline add ‘FailureDeadline’ to the VirtualMachineSnapshot spec with a new value. The allowed format is a duration string which is a possibly signed sequence of decimal numbers, each with optional fraction and a unit suffix, such as “300ms”, “-1.5h” or “2h45m”

  1. apiVersion: snapshot.kubevirt.io/v1alpha1
  2. kind: VirtualMachineSnapshot
  3. metadata:
  4.   name: snap-larry
  5. spec:
  6.   source:
  7.     apiGroup: kubevirt.io
  8.     kind: VirtualMachine
  9.     name: larry
  10.   failureDeadline: 1m

In order to set an infinite deadline you can set it to 0 (not recommended).

Restoring a VirtualMachine

To restore the VirtualMachine larry from VirtualMachineSnapshot snap-larry, Stop the VM, wait for it to be stopped and then apply the following yaml.

  1. apiVersion: snapshot.kubevirt.io/v1alpha1
  2. kind: VirtualMachineRestore
  3. metadata:
  4.   name: restore-larry
  5. spec:
  6.   target:
  7.     apiGroup: kubevirt.io
  8.     kind: VirtualMachine
  9.     name: larry
  10.   virtualMachineSnapshotName: snap-larry

To wait for a restore to complete, execute:

  1. kubectl wait vmrestore restore-larry --for condition=Ready

Cleanup

Keep VirtualMachineSnapshots (and their corresponding VirtualMachineSnapshotContents) around as long as you may want to restore from them again.

Feel free to delete restore-larry as it is not needed once the restore is complete.