Restoring an Amazon Aurora cluster

Use the AWS Backup console to restore Aurora recovery points

AWS Backup restores your Aurora cluster; it does not create or attach an Amazon RDS instance to your cluster. In the following steps, you will create and attach an Amazon RDS instance to your restored Aurora cluster using the CLI.

Restoring an Aurora cluster requires that you specify multiple restore options. For information about these options, see Overview of Backing Up and Restoring an Aurora DB Cluster in the Amazon Aurora User Guide. Specifications for the restore options can be found in the API guide for RestoreDBClusterFromSnapshot.

To restore an Amazon Aurora cluster

1. Open the AWS Backup console at https://console.aws.amazon.com/backup.

2. In the navigation pane, choose Protected resources and the Aurora resource ID that you want to restore.

3. On the Resource details page, a list of recovery points for the selected resource ID is shown. To restore a resource, in the Backups pane, choose the radio button next to the recovery point ID of the resource. In the upper-right corner of the pane, choose Restore.

4. In the Instance specifications pane, accept the defaults or specify the options for the DB engine, DB engine version, and Capacity type settings.

   Note

   If Serverless capacity type is selected, a Capacity settings pane appears. Specify the options for the Minimum Aurora capacity unit and Maximum Aurora capacity unit settings, or choose different options from the Additional scaling configuration section.

5. In the Settings pane, specify a name that is unique for all DB cluster instances owned by your AWS account in the current Region.

6. In the Network & Security pane, accept the defaults or specify the options for the Virtual Private Cloud (VPC), Subnet group, and Availability zone settings.

7. In the Database options pane, accept the defaults or specify the options for Database port, DB cluster parameter group, and IAM DB Authentication Enabled settings.

8. In the Backup pane, accept the default or specify the option for the Copy tags to snapshots setting.

9. In the Backtrack pane, accept the default or specify the options for the Enable Backtrack or Disable Backtrack settings.

10. In the Encryption pane, accept the default or specify the options for the Enable encryption or Disable encryption settings.

11. In the Log exports pane, choose the log types to publish to Amazon CloudWatch Logs. The IAM role is already defined.

12. In the Restore role pane, choose the IAM role that AWS Backup will assume for this restore.

13. After specifying all your settings, choose Restore backup.

    The Restore jobs pane appears. A message at the top of the page provides information about the restore job.

14. After your restore finishes, attach your restored Aurora cluster to an Amazon RDS instance.

    Using the AWS CLI:

    • For Linux, macOS, or Unix:

      aws rds create-db-instance --db-instance-identifier sample-instance \ 
                    --db-cluster-identifier sample-cluster --engine aurora-mysql --db-instance-class db.r4.large

    • For Windows:

      aws rds create-db-instance --db-instance-identifier sample-instance ^ 
                    --db-cluster-identifier sample-cluster --engine aurora-mysql --db-instance-class db.r4.large

See continuous backups and point-in-time restore (PITR) for information about continuous backups and restoring to a chosen point in time.

Use the AWS Backup API, CLI, or SDK to restore Amazon Aurora recovery points

Use StartRestoreJob. The metadata you can include for a restore job will depend if you are restoring a continuous backup to a point in time (PITR) or if you are restoring a snapshot.

Restore a cluster from a snapshot

You can specify the following metadata for an Aurora snapshot restore job. See RestoreDBClusterFromSnapshot in the Amazon Relational Database Service API Reference for additional information and accepted values.

// Required metadata:
dbClusterIdentifier // string
engine // string

// Optional metadata:          
availabilityZones // array of strings
backtrackWindow // long
copyTagsToSnapshot // Boolean
databaseName // string
dbClusterParameterGroupName // string
dbSubnetGroupName // string
enableCloudwatchLogsExports // array of strings
enableIAMDatabaseAuthentication // Boolean
engineMode // string
engineVersion // string
kmsKeyId // string
optionGroupName // string
port // integer
scalingConfiguration // object
vpcSecurityGroupIds // array of strings

Example:

"restoreMetadata":"{\"EngineVersion\":\"5.6.10a\",\"KmsKeyId\":\"arn:aws:kms:us-east-1:234567890123:key/45678901-ab23-4567-8cd9-012d345e6f7\",\"EngineMode\":\"serverless\",\"AvailabilityZones\":\"[\\\"us-east-1b\\\",\\\"us-east-1e\\\",\\\"us-east-1c\\\"]\",\"Port\":\"3306\",\"DatabaseName\":\"\",\"DBSubnetGroupName\":\"default-vpc-05a3b07cf6e193e1g\",\"VpcSecurityGroupIds\":\"[\\\"sg-012d52c68c6e88f00\\\"]\",\"ScalingConfiguration\":\"{\\\"MinCapacity\\\":2,\\\"MaxCapacity\\\":64,\\\"AutoPause\\\":true,\\\"SecondsUntilAutoPause\\\":300,\\\"TimeoutAction\\\":\\\"RollbackCapacityChange\\\"}\",\"EnableIAMDatabaseAuthentication\":\"false\",\"DBClusterParameterGroupName\":\"default.aurora5.6\",\"CopyTagsToSnapshot\":\"true\",\"Engine\":\"aurora\",\"EnableCloudwatchLogsExports\":\"[]\"}"

Restore a cluster to a point in time (PITR)

You can specify the following metadata when you want to restore an Aurora continuous backup (recovery point) to a specific point in time (PITR). See RestoreDBClusterToPointInTime in the Amazon Relational Database Service API Reference for additional information and accepted values.

// Required metadata:
dbClusterIdentifier // string
engine // string
restoreToTime // timestamp; must be specified if UseLatestRestorableTime parameter isn't provided

// Optional metadata:          
backtrackWindow // long
copyTagsToSnapshot // Boolean
dbClusterParameterGroupName // string
dbSubnetGroupName // string
enableCloudwatchLogsExports // array of strings
enableIAMDatabaseAuthentication // Boolean
engineMode // string
engineVersion // string
kmsKeyId // string
optionGroupName // string
port // integer
scalingConfiguration // object
vpcSecurityGroupIds // array of strings