Restoring a DB Instance to a Point in Time

Scenarios

You can restore from automated backups to a specified point in time. The backup data can be restored to new or existing DB instances.

If you delete a database or modify some records in a database at a specified time, you only need to restore the database instead of restoring the whole DB instance. You can also restore databases to a point in time as required.

When you enter the time point that you want to restore the DB instance to, RDS downloads the most recent full backup file from OBS to the DB instance. Then, incremental backups are also restored to the specified point in time on the DB instance. Data is restored at an average speed of 30 MB/s.

Constraints

Constraints on restoring data to a new DB instance:
- The time zone of the new DB instance must be the same as that of the original DB instance. Otherwise, data inconsistency may occur.
Constraints on restoring data to an existing DB instance (other than the original instance):
- Restoring to an existing DB instance will overwrite data on it and cause the existing DB instance to be unavailable.
- To restore backup data to an existing DB instance, the selected DB instance must be in the same VPC as the original DB instance and must have the same DB engine and the same or later version than the original DB instance.
- The storage space of the selected DB instance must be no less than that of the original DB instance. Otherwise, data will not be restored.
- The time zone of the selected DB instance must be the same as that of the original DB instance. Otherwise, data inconsistency may occur.
- DB instances with the TDE function enabled cannot be restored from backups to existing DB instances.

Procedure

Log in to the management console.
Click in the upper left corner and select a region.
Click in the upper left corner of the page and choose Databases > Relational Database Service.
On the Instances page, click the target DB instance.
In the navigation pane on the left, choose Backups & Restorations. On the displayed page, click Restore to Point in Time.
In the displayed dialog box, specify required information and click OK.
1. Select the time range, select or enter a time point within the acceptable range.
  
  If your instance has been restored before by overwriting its data, the period from the time when the restoration started to the time when the first backup was created after the restoration will not be shown in the restorable time range.
2. Select a restoration method.
  - Restore to Existing
    If you have enabled operation protection, click Send Code in the displayed Identity Verification dialog box and enter the obtained verification code. Then, click OK.
    
    Two-factor authentication improves the security of your account and cloud product. For details about how to enable operation protection, see Identity and Access Management User Guide.
    
    Select an existing DB instance and click Next.
3. Select the databases to be restored. You can rename these databases as required. If you do not enter a new name, the original database name will be used.
  - The new database names must be different from each other and must be different from the original database names.
  - The new database names cannot contain the following fields (case-insensitive): rdsadmin, master, msdb, tempdb, model, and resource.
  - Each database name consists of 1 to 64 characters. Only letters, digits, hyphens (-), underscores (_), and periods (.) are allowed.
View the restoration result. The result depends on which restoration method was selected:
- Restore to Existing
  On the Instances page, the status of the DB instance changes from Restoring to Available.