This topic describes how to roll back a disk by using a snapshot. When the system is unresponsive or when accidental changes are made, you can roll back a disk to a previous version by using one of its snapshots. Before you roll back the disk, you must have created a snapshot of the disk. If you roll back system disks, the current SSH key pairs or usernames and passwords of the Elastic Compute Service (ECS) instances remain bound to the disks after the disks are rolled back.
- A snapshot of the disk to be rolled back is created, and no new snapshot is being
created for the disk. For more information, see Create a snapshot of a disk.
Warning The rollback operation is irreversible. After a disk is rolled back, the data that you added, removed, or changed from the time when the snapshot is created to the time when the disk is rolled back is lost. To avoid data loss caused by accidental changes, we recommend that you create a snapshot for the disk to back up data before you roll back the disk.
- The disk is not released.
- After the operating system is replaced, snapshots of the previous system disk cannot be used to roll back the new system disk.
- If you use a disk to create a dynamic extended volume or redundant array of independent disks (RAID), you must stop all I/O operations on the disk before you roll back the disk.
- The disk is attached to an ECS instance and the instance is in the Stopped state.
For more information, see Attach a data disk and Stop an instance.
Notice If the billing method of an instance is pay-as-you-go and the network type of the instance is VPC, you must enable the standard mode for the instance when you stop the instance. If you enable the economical mode for the instance, you may not be able to start the instance after its disk is rolled back.
Roll back a disk by using a snapshot
You can roll back a disk on the Snapshots or Instances page. The following example shows how to roll back a disk on the Instances page:
- Log on to the ECS console.
- In the left-side navigation pane, choose .
- In the top navigation bar, select a region.
- Find the instance whose disk you want to roll back and click Manage in the Actions column.
- The Instance Details tab appears. Click the Snapshot tab.
- Find the snapshot and click Roll Back Disk in the Actions column. Note You can roll back only one disk at a time. Other disks that are attached to the instance are unaffected. After the disk is rolled back, the disk is restored to the state of the entire disk at a certain point in time, instead of the state of a partition or a directory.
- In the Roll Back Disk dialog box, click OK. Note
- Before you click OK, we recommend that you click Create Snapshot to back up the latest data.
- If you select Start Instance After Disk Rollback, the instance automatically starts after the disk is rolled back.
(Optional) Synchronize data after a disk is rolled back
- Create Snapshot B for the original disk at the point in time T2.
- Roll back the original disk by using Snapshot A.
- Create a disk by using Snapshot B. For more information, see Create a disk from a snapshot.
- Attach the new disk to the same ECS instance. For more information, see Attach a data disk.
- Connect to an ECS instance. For more information about connection methods, see Connection methods.
- View the new disk.
- Windows instance: The new disk is displayed in the system.
- Linux instance: Run the mount command to mount the partitions of the new disk.
- Copy useful data from the new disk to the original disk.
- Optional:Release the new disk.
What to do next
- After a disk is rolled back, the host configuration file and the configurations such as the hostname, SSH key pair, password, network, system source, and clock source are initialized. You must reconfigure the information.
- If you resize a disk after you create a snapshot for the disk, the size of the disk is also rolled back after you roll back the disk. To make the disk revert to the size before the rollback, you must log on to the instance to resize the file system again.