Logical backup and recovery for ApsaraDB for Redis - Data Management

Limitations

Before you begin, check whether your instance is supported:

Supported version: Redis 4.0 only. Redis 5.0 and later are not supported.
Supported architecture: Standard master-replica instances only. Cluster instances are not supported.

Features

Feature	Description
SQL file backup	Backs up data in JSON format.
Geo-redundancy	Backs up to an Object Storage Service (OSS) bucket in a specified destination region. Restore to the source region, destination region, or another region.
Long-term archiving	Retains backup data for up to 10 years. After 10 years, Data Disaster Recovery transfers backup sets to Archive Storage.
Fine-grained backup	Backs up a single database, multiple databases, or an entire Tair (Redis OSS-compatible) instance.
Backup speed limit	Limits backup speed to reduce the impact on database performance.
Single-database restoration	Restores a single database without restoring the entire instance, shortening recovery time.

Prerequisites

Before you begin, make sure you have:

A Tair (Redis OSS-compatible) standard master-replica instance running Redis 4.0
A backup schedule. For instructions, see Create a backup schedule. When purchasing, set Data Source Type to Redis and Backup Method to Logical Backup

Configure a backup schedule

Log on to the Data Management (DMS) console V5.0.
In the top navigation bar, choose Security and Specifications (DBS) > Disaster Recovery for Data (DBS) > Backup Plan.

In simple mode, move the pointer over the icon in the upper-left corner and choose All Features > Security and Specifications (DBS) > Disaster Recovery for Data (DBS) > Backup Plan.
On the Backup Schedules page, find your backup schedule and click Configure Backup Schedule in the Actions column.

In the Configure Backup Source and Destination step, configure the following parameters and click Next.

Section	Parameter	Description
N/A	Schedule name	A name for the backup schedule. Data Disaster Recovery auto-generates a name. Use a descriptive name that makes the schedule easy to identify. Names do not need to be unique.
Backup source information	Backup mode	The backup method. Defaults to the method selected at purchase. This example uses Logical Backup.
	Database location	Set to ApsaraDB for Redis.
	Instance region	The region where the source instance resides. Displayed only when Database location is set to ECS-Hosted Database, ApsaraDB for Redis, or No public network IP: Port's self-built database (accessed through the database gateway).
	Database type	Defaults to Redis.
	Redis instance ID	The ID of the source Tair (Redis OSS-compatible) instance. Displayed only when Database location is set to ApsaraDB for Redis.
	Password	The password for the database account. Format depends on the account type: - Default account: Enter the password only. The username is `default`, or use the instance ID (for example, `r-bp10noxlhcoim2**`). - Custom account: Enter `<username>:<password>`. For example, if the username is `testaccount` and the password is `Rp829dlwa`, enter `testaccount:Rp829dlwa`. This format also works for the default account. After entering the credentials, click Test Connection to verify. If the test passes, a Test Passed message appears. If it fails, click Check next to Test Failed** and update the credentials accordingly.
	Compliance warranty regarding cross-border data transfers	Read and select the check box to agree to the compliance commitment.
Backup destination information	Backup storage type	Where backup data is stored: - DBS Storage (recommended): Data Disaster Recovery manages storage. No OSS bucket required. Charged by data volume. See Storage fees. - OSS For User: Stores data in your own OSS bucket. Create the bucket in advance (Standard storage class only). See Create buckets. If you select this option, also configure the OSS Bucket Name parameter. For large data volumes, a subscription storage plan is more cost-efficient than pay-as-you-go.
	Storage encryption	Encryption method for stored data: - Encrypted (recommended): Uses AES-256 with OSS server-side encryption. OSS encrypts objects on upload and decrypts them on download transparently. See Server-side encryption. - Non-encrypted: Data is stored without encryption.

In the Edit Backup Objects step, find the database or table to back up in the Available section. Click the rightwards arrow to move it to the Selected section, then click Next.

- To back up all databases, click Select All in the lower-left corner of the Available section. Supported backup granularity varies by database type. See Supported database types and features. - Databases created after the backup schedule is set up are not included by default. To add them, update the backup objects on the Configure Task page. See Modify backup objects.

In the Configure Backup Time step, configure the following parameters and click Next.

Parameter	Description
Full-scale backup frequency	Periodic Backup or Single Backup. If you select Periodic Backup, also configure Full Data Backup Recurrence and Start At.
Full data backup recurrence	The days of the week to run the backup. Select at least one day.
Start at	The start time. Set this to an off-peak period (for example, 01:00) to avoid impacting production workloads. If a previous full backup is still running at the scheduled start time, Data Disaster Recovery skips the next run.
Incremental backup	Whether to enable incremental backup. Displayed only when Full-scale backup frequency is set to Periodic Backup. Enable this to support point-in-time recovery.
Maximum concurrent threads for full data backup	The maximum number of threads for a full backup. Reduce this value to minimize the impact on database performance during backup.
Maximum speed for full data backup	The maximum speed for full data backup in KB/s. This parameter reduces the impact of backup on the single-thread performance of the Tair (Redis OSS-compatible) instance. You can use the default value.

In the Edit Lifecycle step, configure the retention period for full backup data in the Configure Full Data Backup Lifecycle section. If you enabled incremental backup in step 6, also configure the lifecycle for incremental backup data. For lifecycle rule details, see How do I manage the lifecycle rules of backup sets?
Click Precheck.
When the Precheck Passed message appears, click Start Task. The backup schedule is active when its state changes to Running.

If an error occurs when starting the backup schedule, troubleshoot it as soon as possible. See How do I fix errors for an abnormal backup schedule? If the issue persists, contact technical support in the DingTalk group (ID: 35585947).

Restore a Tair (Redis OSS-compatible) database

This procedure restores a backup set to an existing Tair (Redis OSS-compatible) instance. Create the destination instance before you begin. See Step 1: Create an instance.

Log on to the DMS console V5.0.
In the top navigation bar, choose Security and Specifications (DBS) > Disaster Recovery for Data (DBS) > Backup Plan.

In simple mode, move the pointer over the icon in the upper-left corner and choose All Features > Security and Specifications (DBS) > Disaster Recovery for Data (DBS) > Backup Plan.
On the Backup Schedules page, find the backup schedule and click Manage in the Actions column.
On the Configure Task page, click Restore Database in the upper-right corner.

In the Set Time Restored To step, configure the following parameters and click Next.

Section	Parameter	Description
N/A	Task name	A name for the restore task. Data Disaster Recovery auto-generates a name. Use a descriptive name to identify the task. Names do not need to be unique.
Set time restored to	Time range available to restore	The window from the first full backup to the latest full backup. The restore point must fall within this range.
	Restore to	The point in time to restore to. Must be within the Time range available to restore. - With incremental backup enabled: restore to any point from the first full backup to the last incremental backup. - With incremental backup disabled: restore to the exact completion time of any full backup only.
Configure destination database	Database location	The location of the destination database: - ApsaraDB for Redis - ECS-Hosted Database: Also configure ECS Instance ID and Port Number. - User-Created Database with Public IP Address \<IP Address:Port Number\>: Also configure Address and Port Number. - No public network IP: Port's self-built database (accessed through the database gateway): Configure a database gateway in advance. Also configure Gateway instance ID. - Express Connect DB/VPN Gateway/Intelligent Gateway: Configure a virtual private cloud (VPC) for the database in advance. Also configure Peer VPC.
	Instance region	The region where the destination instance resides.
	Database type	Defaults to Redis.
	Gateway instance ID	The ID of the Database Gateway. Displayed only when Database location is set to No public network IP: Port's self-built database (accessed through the database gateway).
	Address	The endpoint for the destination instance. If the server has firewall rules, click Set Whitelist to get the CIDR blocks for Data Disaster Recovery and add them to the server's allowlist. Displayed only when Database location is set to No public network IP: Port's self-built database (accessed through the database gateway), Express Connect DB/VPN Gateway/Intelligent Gateway, or User-Created Database with Public IP Address \<IP Address:Port Number\>.
	Port number	The port to connect to the destination database. Defaults to 6379.
	Redis instance ID	The ID of the destination Tair (Redis OSS-compatible) instance.
	Password	The password for the destination database account.
	Compliance warranty regarding cross-border data transfers	Read and select the check box to agree to the compliance commitment.

In the Configure Objects to Restore step, configure the following parameters and click Precheck.

Parameter	Description
Conflict handling	Defaults to Rename Object with the Same Name. If a restored object shares the same name as an existing object in the destination database, the restored object is renamed in this format: `<object_name>_dbs_<Restore task ID>_<Timestamp>`. For example, `job_info` becomes `job_info_dbs_<Restore task ID>_<Timestamp>`.
Objects to restore	Select the databases or tables to restore from the Available section and click the rightwards arrow to add them to the Selected section. Restoring at the database or table level reduces the amount of data to transfer and shortens the recovery time objective (RTO). For supported granularity, see Supported database types and features.

When the Precheck Passed message appears, click Start Task. To monitor progress, click Restore Tasks in the left-side navigation pane.

Restoration time depends on the backup schedule specifications and the database size. See Performance tests on backup and restoration.

More operations

Task	Reference
Create a backup schedule via API	CreateBackupPlan or CreateAndStartBackupPlan
Modify backup source, objects, time, or retention policy	Manage a backup schedule
Review backup schedule fees	Billing FAQ
Pause an unused backup schedule to reduce costs	Pause or start a backup schedule

FAQ

Why can't I select my Redis instance when configuring a backup schedule?

Data Disaster Recovery supports only Tair (Redis OSS-compatible) standard master-replica instances running Redis 4.0. Redis 5.0 and later, and cluster instances, are not supported. Check your instance version and architecture in the ApsaraDB for Redis console before configuring the backup schedule.

What point in time can I restore to if incremental backup is disabled?

If incremental backup is disabled, you can only restore to the exact completion time of a full backup — not to any arbitrary point in time. To enable point-in-time recovery, enable incremental backup in the Configure Backup Time step when setting up your backup schedule.

How are naming conflicts handled during restoration?

By default, if a restored database or table shares the same name as an existing object in the destination, Data Disaster Recovery renames the restored object using this format: <object_name>_dbs_<Restore task ID>_<Timestamp>.

Can I restore data to a region different from the backup source region?

Yes. The geo-redundancy feature lets you back up data to an OSS bucket in a specified destination region and restore it to an instance in the source region, the destination region, or another region. Configure the destination region in the Backup Destination Information section when setting up your backup schedule.

Why should I limit backup threads and speed?

Backup operations share system resources with your database workload. Reducing the maximum concurrent threads and backup speed lowers the impact on database performance during backup. Set both parameters to off-peak thresholds and adjust based on observed performance impact.