Restore

Introduction

Restore page allows you to recover applications, namespaces, or an entire cluster to a previous state. You can create a restore from the Restores tab, view all restore jobs, with their current status, target cluster, and backup source.

---

Who Can Perform This Action?

Users need to have super-admin permission to create a restore.

Creating Restore

To restore a backup, you can create a restore:

1. Go to Data Protection Management → Restores.

2. Click + New Restore. A new modal window will appear.

3. Enter the required details for each section

   1. Basic Details

   Field	Required/Optional	Description
   Restore Name	Required	A unique name for identifying the restore job
   Description	Optional	A short description explaining the purpose of the restore
   Labels	Optional	Add key-value labels to organize or filter restore jobs
   Restore From Backup	Required	Choose how to restore: Specific Backup – Select a specific backup manually. Latest Backup Created by a Schedule – Automatically uses the latest backup generated by a defined schedule
   Backup to be Restored	Required	For: Specific Backup - Select the backup you want to restore from the available list Latest Backup Created by a Schedule - Select the backup schedule from which you want the latest backup to be restored

   2. Restore Target

   Field	Required/Optional	Description
   Restore To (Cluster)	Required	Select the destination cluster for the restore.
   All Namespaces	Optional	Restores all namespaces from the selected backup.
   Specific Namespaces	Optional	Select specific namespaces to restore. You can use regex to match multiple namespaces.
   Included Namespaces	Optional (visible only when Specific Namespaces is selected)	Select one or more namespaces to include in the restore process. Only these namespaces will be restored from the backup. You can also provide regex expressions to include namespaces that follow a specific naming pattern.
   Excluded Namespaces	Optional (visible only when Specific Namespaces is selected)	Select namespaces that should be excluded from the restore process, even if they exist in the backup. You can also provide regex expressions to exclude namespaces that follow a specific naming pattern.
   Filter resources using label selectors	Optional	Filters resources included in the restore using Kubernetes label selectors. Labels within the same group are evaluated using AND logic, while separate groups are evaluated using OR logic. Operators supported are Equals, In, NotIn,Exists,DoesNotExist

   3. Restore Settings

   Field	Required/Optional	Description
   Resource Status	Optional	Specify which Kubernetes resources should be included or excluded in the restore status. Resources must be provided in the format resourceKind.group, such as storageclasses.storage.k8s.io
   Namespace Mapping	Optional	Map source namespaces to new target namespaces. For example, restore prod → stage-prod
   Include Cluster Resources	Optional	Defines whether cluster-scoped resources should be restored. You can choose: Auto (Recommended): Automatically determines which cluster resources to include, based on context Include all: Restores all cluster-scoped resources from the backup Exclude all: Skips all cluster-scoped resources, restoring only namespace-scoped resources
   Restore Volumes	Optional	Enables restoring persistent volumes from snapshots taken during the backup. Toggle this on if you need your application data to be restored along with configurations
   Preserve Node Ports	Optional	Retains existing node ports for restored services to avoid port conflicts or reassignments during restore
   Existing Resource Policy	Optional	Specifies how to handle already existing resources: Skip Resources – Ignore existing ones Update Resources – Replace or modify existing resources
   Modify Resources During Restore	Optional	Allows you to modify resources dynamically during restore using: ConfigMap: Reference a ConfigMap containing JSON patch rules to make specific changes (for example, updating storage classes or image tags) JSON Patch: Apply JSON patches inline for fine-grained control

   ::: info Modify resources during restore

   A resource modifier ConfigMap lets you customize Kubernetes resources during a restore by applying JSON patch rules before they are recreated on the target cluster. This helps make backups portable and ensures smooth restores across different environments.

   Example use cases: Cluster migrations: Change storageClassNames in PVCs to match storage classes available in the new cluster. How to use: Create a ConfigMap with JSON patch rules defining the changes you want. Velero will automatically apply these rules to your resources during the restore, allowing you to tailor them for your target cluster's setup.

   :::

   4. Timeout & Execution Controls

   Field	Required/Optional	Description
   Parallel Files Download	Optional	Number of restore threads to run in parallel. Set to 0 to match the node CPU count automatically
   Write Sparse Files	Optional	Enables writing of sparse files during volume restoration to optimize disk space usage
   Item Operation Timeout	Optional	Duration (in seconds/minutes) to wait for individual restore operations before timing out

4. Click Save. A restore will be initiated.

---

Viewing Restores

After restoration of a backup is initiated, you can view your restore on the Restores Page under Data Protection Management. This page shows all your restore jobs, along with their current status, target cluster, and backup source.

Column	Description
Restore Name	Displays the unique name of the restore job. Click the name to view detailed information such as restored namespaces, logs, and resource summary.
Status	Shows the current state of the restore: Progressing – The restore process is currently in progress Completed – Restore has successfully finished, and all resources have been recovered Failed – The restore could not be completed due to configuration or resource-level errors Partially Completed – Restore finished, but some resources were skipped or encountered issues during restoration
Target Cluster	Displays the cluster where the backup was restored
Backup Source	Shows the name of the backup used for this restore
Schedule	Indicates the backup schedule from which the backup originated
Created By	Displays the user or service account that initiated the restore
Date	Shows the date and time when the restore was created or triggered

You can use the Search bar or apply filters such as Cluster, Status, and Created By to quickly find a specific restore job or narrow down the list based on recent restore activity.

Filter	Description
Cluster	Filter restores by the target cluster where they were applied
Status	Filter restores by their current state (for example, Progressing, Completed, Failed, Partially Failed)
Created By	Filter restores by the user or schedule that initiated them

These filters make it easier to locate specific restore jobs, track recovery operations across clusters, and verify restore outcomes — ensuring that your environments are safely and accurately recovered.

Inspecting Restore

When you click on a restore job from the Restores tab, a detailed view opens showing its configuration, progress, and results.
This page helps you verify whether the restore completed successfully and understand what resources were included or skipped.

The left-side of the page displays the following information:

Field	Description
Status	Shows the current state of the restore (Completed, Progressing, Failed, or Partially Completed)
Warnings & Errors	Indicates the number of warnings or errors encountered during the restore process
Resources	Displays the number of successfully restored resources against the total resources processed
Target Cluster	The cluster where the restore operation was performed
Created By	The user or service account that initiated the restore
Restored From Backup	Displays the backup name and timestamp used for this restore. Clicking the backup name navigates to its details view
Duration	Shows how long the restore took to complete
Included Resources	Lists all resource types included in the restore operation
Excluded Resources	Lists resources that were excluded from the restore, such as nodes, events, or cluster-scoped resources

On the right-side of the page, you can switch between two tabs: Namespace Mapping and Manifest, to review the details of what was restored.

Tab	Description
Namespace Mapping	Displays the mapping between source and target namespaces. This helps confirm that workloads were restored to the correct destination (for example, production-space → prod-space)
Manifest	Shows the YAML manifest of the restore job, including specifications, restore options, and applied configurations. This helps advanced users validate cluster resources and understand how restore parameters were executed