Restoring a Cluster Instance to a Point in Time

Precautions

• To use this function, contact customer service to apply for the corresponding permission.
• Only cluster instances of versions 4.0, 4.2, and 4.4 can be restored to a specified point in time.
• Data can be restored to a specific point in time only after the automated and incremental backup policies are enabled.
• Data can be restored to a new instance or the original instance.
• To ensure data security, the dropDatabase operation is blocked when the incremental backup is restored to a point in time. Empty databases or views may exist after the restoration. You can delete them.
• Data cannot be restored to a point in time in any of the following scenarios: rename operation, collmod operation, creating a user, deleting a user, creating a role, deleting a role, enabling shard IP addresses of a cluster instance, changing the password of the shard node user, enabling config IP addresses of a cluster instance, changing the password of the config node user, and changing the password of the rwuser user. When a restricted scenario occurs, the incremental backup stops. After the next automated full backup, the incremental backup resumes.
• If the time window of the full backup overlaps with that of the incremental backup, the full backup prefers. The incremental backup is restricted so that a few time ranges are not within the recovery time window.

Procedure

1. Log in to the management console.
2. Click in the upper left corner and select a region and a project.

   If you want compute and network resources dedicated to your exclusive use, enable a DeC and apply for DCC resources. After enabling a DeC, you can select the DeC region and project.

3. Click in the upper left corner of the page and choose Databases > Document Database Service.
4. On the Instances page, click the cluster instance name.
5. In the navigation pane on the left, choose Backups & Restorations.
6. On the Backups & Restorations page, click Restore to Point in Time.
   Figure 1 Restoring a cluster instance to a point in time
   

7. Select the date and time range, select or enter a time point within the acceptable range, and select Create New Instance or Restore to Original.
   Figure 2 Restoring to a point in time
   

8. On the displayed page, the instance is restored based on the restoration method you selected in 7.
   • Create New Instance

     The Create New Instance page is displayed for you to create an instance using the backup data. The new instance is independent from the original one.

     • You are recommended to deploy the restored instance in a different AZ to ensure that applications will not be adversely affected by the failure in any single AZ.
     • The database type, DB instance type, compatible MongoDB version, storage engine, and storage type must be the same as those of the original and cannot be changed.
     • The storage space is the same as that of the original instance by default. You can increase the storage space, but you cannot reduce it.
     • Other settings can be modified. For details, see Buying a Cluster Instance.
   • Restore to Original
     Check that the status of the instance on the Instances page is Restoring.

     

     • Restoring backup data to the original instance will overwrite existing data on the instance and cause the instance to be unavailable during the restoration. Exercise caution when performing this operation.
     • The administrator password of the instance remains unchanged after the restoration.