All Products
Search

This Product

    PolarDB:Backup and recovery

    Document Center

    PolarDB:Backup and recovery

    Last Updated:Aug 26, 2026

    This topic describes two backup and recovery methods and explains how to use them.

    Overview

    PolarDB-X 1.0 provides instance-level and database-level backup and recovery. You can back up an instance automatically or manually. Two backup methods are available: fast backup and consistent backup. To recover an instance, you can restore data to a new instance from an existing backup set.

    Backup methods

    PolarDB-X 1.0 provides two backup methods, fast backup and consistent backup, both of which support data recovery. The following table compares the two methods.

    Backup method

    Benefits

    Limitations

    Use cases

    fast backup

    • Fast backup and recovery.

    • Supports recovery from backup sets and point-in-time recovery.

    • Supports all instance versions.

    In database and table sharding scenarios, this method ensures data consistency for each RDS instance but does not guarantee it globally.

    Ideal for routine backup and recovery.

    consistent backup

    In database and table sharding scenarios, this method guarantees global data consistency.

    • Relatively slow backup and recovery.

    • Supports recovery from backup sets only; point-in-time recovery is not available.

    • Requires version 5.3.8 or later.

    • During a consistent backup, distributed transactions on the PolarDB-X 1.0 instance are locked for a few seconds. This lock does not affect non-transactional SQL or single-node transactions, but it blocks distributed transaction commits and may cause sub-second jitter in SQL execution response time (RT). We recommend that you perform consistent backups during off-peak hours.

    Suitable for backup and recovery scenarios that require strict data consistency, such as in the financial industry and for core online transaction processing (OLTP) businesses.

    Limitations

    • The PolarDB-X 1.0 automatic backup policy is disabled by default. You must enable it manually. For more information, see Configure an automatic backup policy.

    • The PolarDB-X 1.0 log backup feature depends on the underlying RDS instances. The log backup policy that you configure in the PolarDB-X 1.0 console is automatically synchronized to all underlying RDS instances. After configuring the policy, do not modify it in the RDS console to avoid invalidating your backup sets.

    • PolarDB-X 1.0 backup and recovery requires log backups. You must enable the log backup policy before you perform a backup or recovery to ensure your backup sets are valid.

    • Do not perform DDL operations during a backup to prevent backup and recovery failures.

    • During the backup process, ensure that the underlying RDS instances of your PolarDB-X 1.0 instance are running to avoid backup failures.

    • Ensure that all data tables have a primary key to maintain data accuracy during the recovery from a consistent backup.

    • Depending on resource availability for PolarDB-X 1.0 and RDS, the recovery process may automatically change the instance type and availability zone of the instance. After the instance is restored, verify and adjust the configuration as needed to avoid disrupting your business.

    • You are billed for the restored instance based on its actual billing method and instance type. For more information, see PolarDB-X 1.0 pricing.

    Configure an automatic backup policy

    1. Log on to the PolarDB for Xscale console, and select the region where your instance is located.

    2. On the Instances page, click the name of your instance.

    3. In the left-side navigation pane, click Backup and Restoration.

    4. Click the Backup Policy tab to view the current automatic backup policy. The PolarDB-X 1.0 automatic backup policy is disabled by default. If this is your first time on this page, you must configure the policy before it can take effect.

    5. Click Edit to configure the automatic backup policy.

    6. Set the parameters for the automatic backup policy and click OK.

    Configure local log settings

    1. Log on to the PolarDB for Xscale console, and select the region where your instance is located.

    2. On the Instances page, click the name of your instance.

    3. In the left-side navigation pane, click Backup and Restoration.

    4. Click Local Log Settings.

    5. Click Edit to modify the local log configuration.

    6. Click OK to save the configuration.

      Note

      A PolarDB-X 1.0 instance does not have a built-in log backup capability; instead, the policy you configure is applied to all underlying RDS instances.

    Back up data

    1. Log on to the PolarDB for Xscale console, and select the region where your instance is located.

    2. On the Instances page, click the name of your instance.

    3. In the left-side navigation pane, click Backup and Restoration.

    4. Click Data Backup.

    5. Select a Backup Method and a Backup Level.

      Note

      PolarDB-X 1.0 supports only RDS-level backups. If all your databases reside on a single RDS instance, the entire RDS instance is backed up even if you select only one database. The size of the resulting backup set is the same as if you had selected all databases.

    6. Click OK to submit the backup task.

    7. Click the button in the upper-right corner to view the progress of the backup task. You can view successfully created backup sets on the Data Backup Set tab.

    Recover data

    1. Log on to the PolarDB for Xscale console, and select the region where your instance is located.

    2. On the Instances page, click the name of your instance.

    3. In the left-side navigation pane, click Backup and Restoration.

    4. Click the Data Backup Set tab.

    5. Choose a data recovery method:

      • To recover data from a specific backup set, find the backup set in the list and click Restore in the Actions column.

      • To perform a point-in-time recovery, click the Data Recovery button in the upper-right corner of the page.

    6. Click Precheck to verify whether a valid backup set is available for recovery. If the precheck fails, you cannot recover the data.

    7. After the precheck succeeds, click Activate Now. You are then directed to the order confirmation page.

    8. After you confirm the order, click Activate Now to start the data recovery. You can view the progress of the recovery task in the task list in the upper-right corner of the page.

    9. After the recovery is complete, the restored instance appears on the Instances page.