Roll back disks by using a snapshot-consistent group - Elastic Compute Service

Restore disk data with a snapshot-consistent group to recover from system faults or accidental operations.

Prerequisites

Before you roll back disks with a snapshot-consistent group, ensure the following requirements are met:

A snapshot-consistent group is created for the instance.

Important
The rollback operation cannot be reversed. All data changes made after the snapshot-consistent group was created are lost. Create a snapshot-consistent group to back up disk data before you roll back disks.
The instance is in the Stopped state. If not, stop the instance.

Important
If an ECS instance uses the pay-as-you-go billing method and resides in a VPC, you must enable the standard mode when you stop the instance. If you enable the economical mode, you may be unable to start the instance after you roll back the disks attached to the instance.
The disks in the snapshot-consistent group are not released, detached, or undergoing snapshot creation.
The instance operating system has not been replaced since the snapshot-consistent group was created.
No disk snapshots have been deleted from the snapshot-consistent group. If a disk's snapshot is deleted, that disk cannot be rolled back. Only disks with snapshots still in the group can be rolled back.

Go to ECS console - snapshot consistency group.
In the top navigation bar, select the region where the target ECS instance is located.
On the Snapshot-consistent Groups tab, find the target snapshot-consistent group and click Roll Back in the Actions column.
In the Roll Back dialog box:
1. Review the notes.
2. In the Disk Snapshot section, select the disk snapshots to roll back.
3. Select Start Instance After Disk Rollback as needed.
4. Click OK.
  
  After rollback, the Rolled back message appears.

After rollback, log on to the instance to verify that the disk data matches the snapshot state.