Do not skip PVC when building VMIs resaources graph during restore item action.

[amastbau]

In order to restore one or more Virtual Machines out of many in the SAME namespace, velero offers restoring by resource label.
When using a label selector, any resource not including the restore label are excluded by velero.
That might include other resources that are quieted for the VM, such as datavolumes and PVCs.
In order to resolve this, we already include any additional required resource in a graph which is return to velero for backup/restore even if those resources do not include matching label.

With this PR we are removing a condition results in excluding PVC for datavolume type VM volume from the graph, and a the datavolume pending for ever for the existence of the associated PVC after restore.

fixes issue: https://issues.redhat.com/browse/OADP-5608

How to test?
Deploy the following VM, with DataVolume type volume and a label.

$ kubectl create namespace test-vm

apiVersion: kubevirt.io/v1
kind: VirtualMachine
metadata:
  labels:
    app: test-vm-1
  name: test-vm-1
  namespace: test-vm
spec:
  dataVolumeTemplates:
  - metadata:
      annotations:
        cdi.kubevirt.io/storage.deleteAfterCompletion: "false"
      creationTimestamp: null
      name: test-vm-1-dv
    spec:
      pvc:
        accessModes:
        - ReadWriteOnce
        resources:
          requests:
            storage: 150Mi
        volumeMode: Block
      source:
        registry:
          pullMethod: node
          url: docker://quay.io/kubevirt/cirros-container-disk-demo
  running: true
  template:
    metadata:
      creationTimestamp: null
      name: test-vm-1
    spec:
      architecture: amd64
      domain:
        devices:
          disks:
          - disk:
              bus: virtio
            name: volume0
          interfaces:
          - masquerade: {}
            name: default
          rng: {}
        machine:
          type: q35
        resources:
          requests:
            memory: 256M
      networks:
      - name: default
        pod: {}
      terminationGracePeriodSeconds: 0
      volumes:
      - dataVolume:
          name: test-vm-1-dv
        name: volume0

$ velero backup create test-vm-backup --include-namespaces=test-vm

$ kubectl delete namespace/test-vm

$ velero restore create restore --from-backup=test-vm-backup --selector=app=test-vm-1

# The DataVolume and the PVC are included in the restore:

$ velero restore describe restore --details
Name:         restore
Namespace:    openshift-adp
Labels:       <none>
Annotations:  <none>

Phase:                       Completed
Total items to be restored:  7
Items restored:              7

Started:    2025-04-15 12:47:34 +0000 UTC
Completed:  2025-04-15 12:47:45 +0000 UTC

Backup:  test-vm-backup

Namespaces:
  Included:  all namespaces found in the backup
  Excluded:  <none>

Resources:
  Included:        *
  Excluded:        nodes, events, events.events.k8s.io, backups.velero.io, restores.velero.io, resticrepositories.velero.io, csinodes.storage.k8s.io, volumeattachments.storage.k8s.io, backuprepositories.velero.io
  Cluster-scoped:  auto

Namespace mappings:  <none>

Label selector:  app=test-vm-1

Or label selector:  <none>

Restore PVs:  auto

CSI Snapshot Restores:
  test-vm/test-vm-1-dv:
    Snapshot:
      Snapshot Content Name: 869720b2434ab00b3947a8811a1d62d3c427a389ae14d6b7edc5ebf4ada6c1dc
      Storage Snapshot ID: 0001-0011-openshift-storage-0000000000000002-6e885852-eacd-4673-8c04-ed000af1ff7c
      CSI Driver: openshift-storage.rbd.csi.ceph.com

Existing Resource Policy:   <none>
ItemOperationTimeout:       4h0m0s

Preserve Service NodePorts:  auto

Uploader config:


HooksAttempted:   0
HooksFailed:      0

Resource List:
  cdi.kubevirt.io/v1beta1/DataVolume:
    - test-vm/test-vm-1-dv(created)
  kubevirt.io/v1/VirtualMachine:
    - test-vm/test-vm-1(created)
  snapshot.storage.k8s.io/v1/VolumeSnapshot:
    - test-vm/velero-test-vm-1-dv-jp4jt(created)
  snapshot.storage.k8s.io/v1/VolumeSnapshotContent:
    - snapcontent-9765a149-df9b-4882-aa46-7783b012f5c2(created)
  v1/Namespace:
    - test-vm(created)
  v1/PersistentVolume:
    - pvc-19e9056b-c772-433b-8f46-5cc3d97f65f5(skipped)
  v1/PersistentVolumeClaim:
    - test-vm/test-vm-1-dv(created)

Which issue(s) this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close the issue(s) when PR gets merged):
Fixes #

Special notes for your reviewer:

Release note:

FixBug: Include PVC in restore VM graph so it will be included even when using label selector that would exclude it.

[kubevirt-bot]

Hi @amastbau. Thanks for your PR.

PRs from untrusted users cannot be marked as trusted with /ok-to-test in this repo meaning untrusted PR authors can never trigger tests themselves. Collaborators can still trigger tests on the PR using /test all.

I understand the commands that are listed here.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[alromeros]

/cc @alromeros

[ShellyKa13]

Looks good, I would say that it will be good to also explain the issue in the commit message.
Also missing a UT for this
And I believe this change justifies a release note since it fixes a bug

[weshayutin]

test container based vm's restore via selector

All passed

Setup

• also using: Enable selective PVC restore by moving size patch to backup action and updating CSI additional items openshift/velero#379
  USING:
  unsupportedOverrides:
  kubevirtPluginImageFqin: quay.io/amastbau/kubevirt-velero-plugin:latest
  veleroImageFqin: quay.io/amastbau/velero:latest-l

BACKUP 3 cirros vm's

e.g. from vm-1

apiVersion: kubevirt.io/v1
kind: VirtualMachine
metadata:
  annotations:
  generation: 1
  labels:
    app: cirros-test-cont-1
  name: cirros-test-cont-1
  namespace: cirros-test-cont
spec:
  dataVolumeTemplates:
  - metadata:
      annotations:
        cdi.kubevirt.io/storage.deleteAfterCompletion: "false"
      creationTimestamp: null
      name: cirros-test-cont-1-dv
    spec:
      pvc:
        accessModes:
        - ReadWriteOnce
        resources:
          requests:
            storage: 150Mi
        volumeMode: Block
      source:
        registry:
          pullMethod: node
          url: docker://quay.io/kubevirt/cirros-container-disk-demo
  running: true
  template:
    metadata:
      creationTimestamp: null
      name: cirros-test-cont-1
    spec:
      architecture: amd64
      domain:
        devices:
          disks:
          - disk:
              bus: virtio
            name: volume0
          interfaces:
          - masquerade: {}
            name: default
          rng: {}
        machine:
          type: q35
        resources:
          requests:
            memory: 256M
      networks:
      - name: default
        pod: {}
      terminationGracePeriodSeconds: 0
      volumes:
      - dataVolume:
          name: cirros-test-cont-1-dv
        name: volume0

backup results

backup describe https://termbin.com/ma7l

Nuke the vm's from orbit

oc delete -f cirros-test-1.yaml -f cirros-test-2.yaml -f cirros-test-3.yaml

Restore ALL

https://termbin.com/utox

Nuke all, restore 1 vm via selector

https://termbin.com/21vg

Nuke all, restore 1 vm ( second vm) via selector

https://termbin.com/shi2

[alromeros]

Looks good, just couple of comments

[ShellyKa13]

/approve
Thanks @amastbau looks good

[kubevirt-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: ShellyKa13

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• OWNERS [ShellyKa13]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[alromeros]

Nit but could you fix the typos in the commit name?
Do not skip PVC when building VMIs resaources grap -> Do not skip PVC when building VMIs resources graph
Looks good otherwise, thanks!

[ShellyKa13]

Thanks approving with the approval of @alromeros
/lgtm