Compute Nest provides centralized backup and restoration to simplify data protection for your service instances. This topic describes its use cases, how it works, associated costs, and how to enable it.
Background
You may need to back up data for your services for various reasons, such as:
-
Game servers: Private servers for games like Palworld can experience data loss. Regular backups help you recover quickly.
-
Database services: Back up your databases to protect against data loss and minimize downtime from potential failures.
-
Pre-upgrade protection: Create a backup before a major service upgrade. If the upgrade fails, you can restore the service to its previous state.
Previously, protecting the data in a service instance was a manual and complex task. You had to navigate to the console of each underlying cloud product, such as Elastic Compute Service (ECS), to find the correct resources and perform backups. For example, you would need to manually create a snapshot for each EBS disk. When it was time to restore, you had to find the correct snapshot or RDS instance backup and perform the rollback. This process was tedious and error-prone, especially for services with many EBS disks or RDS instances. To solve this, Compute Nest introduces backup management, which lets you back up or restore all cloud resources for a service instance with a single click from the Compute Nest console.
How it works
The backup and restoration capabilities in Compute Nest depend on the native backup features of the underlying cloud resources. A resource can be backed up via Compute Nest only if both Compute Nest and the underlying cloud product support it. Currently, Compute Nest supports backup and restoration for the following cloud resources:
-
Elastic Block Storage (EBS) disks.
-
Alibaba Cloud RDS instances with the MySQL, SQL Server, or PostgreSQL engine.
Backup
When you start a backup, Compute Nest first creates a backup record. This record stores the start and end times, the overall backup status, and the IDs and individual backup statuses of all cloud resources being backed up.
To perform the backup, Compute Nest uses the service instance ID and resource tags to identify all associated cloud resources, filtering for EBS disks and supported RDS instances. It then calls the API of each underlying cloud product to initiate the backup process for each resource. While the backups are in progress, Compute Nest continuously polls their status. The overall status in the backup record reflects the collective state of all operations. The record shows "Backing up" as long as any single resource is still being backed up. If any resource backup fails, the entire job is marked as "Backup failed". The final status is marked as "Backup successful" only after every resource has been backed up successfully.
Restoration
The restoration process is similar to the backup process. Before starting, Compute Nest creates a restoration record that stores the start and end times, the ID of the backup being used, the overall restoration status, and the restoration status of each cloud resource.
Because an EBS disk can only be restored when its attached ECS instance is in a stopped state, Compute Nest automatically stops the entire service instance before beginning the restoration. After the instance stops, Compute Nest uses the specified backup ID to find the backup data for each resource and calls the respective cloud product APIs to begin the restoration. Compute Nest continuously monitors the progress of each restoration job and updates the restoration record accordingly. The overall status shows "Restoring" as long as any resource is still being restored. If any restoration fails, the overall status becomes "Restoration failed". The job is marked as "Restoration successful" only after all resources have been restored successfully. Once the restoration is complete, Compute Nest automatically starts the service instance. Note: The restoration process for RDS instances running MySQL or PostgreSQL temporarily alters database and table names during the process. To resolve this, Compute Nest uses Function Compute (FC) to execute SQL statements that rename the databases and tables back to their original names. As a result, you may incur charges for FC when restoring a service instance that includes RDS instances with these engines.
Billing
The Compute Nest backup management feature is free, but you will incur costs from the underlying cloud products used for backup and restoration. These costs primarily fall into three categories:
-
Costs for Elastic Block Storage (EBS) disk snapshots. For more information, see Snapshot billing.
-
Costs for RDS backup storage. For more information, see the following documents:
-
MySQL engine: Backup fees.
-
SQL Server engine: Backup fees.
-
PostgreSQL engine: Backup fees.
-
-
Costs for Function Compute (FC), which may be incurred during restoration as explained in the section. For more information, see Billing overview.
Enable backup management
To enable the backup and restoration feature for your service, perform the following two steps:
-
In the Service O&M (Optional) section, click + Add O&M Operation. In the dialog box that appears, select the Basic O&M Item tab, set the template name to SingleNode, select the start and stop O&M items, and then click OK.
-
Turn on the Backup and Recovery button.
After completing these steps, the backup and restoration features are enabled for all new instances of this service.
Next steps
To learn how to use the feature, see Use instance backup.