Cloud Backup provides a cloud-native, scheduled data protection solution for Object Storage Service (OSS). Back up objects in OSS buckets on a regular schedule, and restore them to any recovery point when needed.
Introduction
OSS backup is a fully managed, cloud-native data protection solution for Alibaba Cloud OSS. Define a backup policy to regularly back up objects in your OSS buckets to Cloud Backup. If objects are accidentally deleted or modified, restore them from any available recovery point.
OSS backup has usage limits. For details, see Limits.
Benefits
-
Native OSS integration
Protect data in OSS buckets without installing any backup agents or components.
-
Fine-grained data protection
Back up an entire bucket or only objects that match specified prefixes for targeted protection.
-
Flexible versioning
Schedule backup plans to run daily, weekly, monthly, or yearly, and set a retention period for each backup version.
-
Simple point-in-time recovery
Restore one or more objects to a specific recovery point directly from the Cloud Backup console, with no scripts or custom code required.
-
Deduplication and compression
Cloud Backup automatically deduplicates and compresses data during backup to minimize storage consumption.
-
Immutable backups
Enable backup vault locking to prevent backup data from being deleted, whether accidentally or maliciously, until the retention period expires.
How it works
OSS backup is natively integrated with OSS and requires no additional service deployment. For buckets that contain more than 100 million objects, use the OSS inventory feature to accelerate backup scans.
When a backup job runs, Cloud Backup scans the target bucket and retrieves the latest object versions through the OSS inventory or the specified prefix filter. Changed objects are identified, deduplicated, compressed, and encrypted before the incremental data is uploaded to the backup vault. Cloud Backup uses an incremental-forever strategy: only the first job performs a full backup, and each subsequent job uploads only the changed data and then synthesizes a full recovery point. This approach keeps storage consumption low.
OSS backup vs. OSS versioning
-
OSS backup creates recovery points based on a schedule (hourly, daily, weekly, or other intervals). This approach controls the number of backup versions and keeps storage costs predictable, even when objects change frequently.
-
OSS versioning creates a new version every time an object is modified, enabling retrieval of any historical version. Choose the approach that best fits your data criticality and recovery requirements.
Procedure
The following procedure shows how to back up objects from an OSS bucket in the Cloud Backup console.
You are not charged for activating Cloud Backup. You are charged for the storage usage of backup vaults created in Cloud Backup. For more information, see Billing methods and billable items.
When you create a backup plan, configure the backup vault, data source, backup cycle, and retention period. Cloud Backup starts the backup plan and continuously backs up objects from the specified OSS bucket.
ImportantYou can enjoy a 30-day free backup plan. You cannot edit the source path and backup cycle for a free backup plan.
If the region that you select supports backup policies, you can set a backup plan only by associating it with a backup policy. Cloud Backup periodically backs up OSS objects based on the backup policy.
To view the regions that support backup policies, click Policy Center in the left-side navigation pane of the Cloud Backup console. For more information about how to create a backup policy, see Create a backup policy.
After you create a backup plan, the first backup job performs a full backup and subsequent backup jobs perform incremental backups.
You are not charged for restoration in the same region. If a file exception occurs in an OSS bucket, you can restore all objects or some objects that meet the specified conditions based on historical backups. You can restore the objects to the source OSS bucket for free or restore the objects to a new OSS bucket for free. You are charged for the traffic and storage capacity of the remote mirror vault.
Usage notes
The first time you create a backup plan or attach a backup policy, Cloud Backup automatically creates a service-linked role for Cloud Backup named AliyunServiceRoleForHbrOssBackup to obtain permissions to access related resources. In the authorization dialog box that appears, grant the required permissions to the role.
Before you use Cloud Backup to protect OSS data, note the following:
Cloud Backup supports backing up and restoring buckets of the Standard and IA storage classes. It does not support backing up or restoring buckets of the Archive, Cold Archive, or Deep Cold Archive storage classes.
Cloud Backup supports backing up and restoring objects of the Standard and IA storage classes. Objects of the IA storage class are restored as objects of the Standard storage class. Cloud Backup does not support backing up or restoring objects of the Archive, Cold Archive, or Deep Cold Archive storage classes.
Cloud Backup does not support backing up or restoring the access control list (ACL) of an object.
You are charged OSS request fees for OSS API operations that Cloud Backup calls during the backup of OSS objects. These fees are charged by OSS. For more information, see Billing.
When Cloud Backup reads OSS data from IA buckets or objects for backup, you incur OSS data retrieval fees. You are charged based on the volume of retrieved data. These fees are charged by OSS. For more information, see Data processing fees.
Deleting an early backup version from the backup history does not affect the data integrity of other backup versions.
The scheduled backup plan that is enabled when you create an OSS bucket is a 30-day free trial. The plan is automatically disabled upon expiration. Switch to the paid plan promptly.
ImportantThe 30-day free trial has several key limitations: It does not support backup policies, and its source path and backup cycle are fixed. Most importantly, the trial uses an isolated backup vault that cannot be changed or merged after converting to a paid plan. Therefore, if you intend to consolidate all backups into a single vault, you should start directly with a paid plan instead of using the free trial.
Billing
OSS backup incurs the following fees:
-
Storage usage fees
Cloud Backup offers two storage redundancy types: locally redundant storage (LRS) and zone-redundant storage (ZRS). Fees are based on the actual storage consumption of your backup vaults. View vault storage usage on the Overview page in the Cloud Backup console. For pricing details, see Pricing.
-
API request fees
-
Before each backup job, Cloud Backup calls the ListObject operation to retrieve the object list. Each request lists up to 1,000 objects.
-
Cloud Backup then calls the HeadObject operation to retrieve metadata for each object, generating one request per object.
-
Cloud Backup calls the GetObject operation to download each object for backup, generating one request per object.
-
Cloud Backup calls the ListObject operation to track backup progress and verify backed-up objects. Each request covers up to 1,000 objects.
-
If the OSS inventory feature is enabled, the ListObject requests above are not charged. Instead, OSS inventory fees apply. For details, see Bucket inventory.
For OSS API request billing rules, see API operation calling fees.
-
-
OSS data retrieval fees
When Cloud Backup reads data from Infrequent Access (IA) buckets or objects for backup, data retrieval fees apply. OSS charges these fees based on the volume of retrieved data. For details, see Data processing fees.
In addition, Cloud Backup charges the following fees based on specific backup configurations or operations:
What to do next
-
Back up OSS objects. See Back up OSS objects and Back up many OSS objects using the OSS inventory feature.
-
Explore Cloud Backup Best practices.