Roll back disk using snapshot
If you lose data on a disk due to accidental operations, ransomware attacks, or other incidents, you can use a snapshot of that disk to roll it back. Rolling back a disk restores it to the exact state it was in when the snapshot was created. This topic describes how to roll back a disk by using a snapshot.
Limitations
A snapshot can only roll back the disk it was created from. To restore data to a different disk, create a disk from a snapshot.
After replacing the operating system of an ECS instance, snapshots of the original system disk cannot roll back the new system disk. As a workaround, create a pay-as-you-go disk from the snapshot, attach it to the instance, copy the data, and release the temporary disk to avoid unnecessary charges. See Create a data disk from a snapshot, Attach a data disk, and Release a cloud disk.
Shared snapshots and snapshot copies cannot be used for disk rollback.
Prerequisites
Before you begin, verify the following:
A snapshot exists for the disk, is available for rollback, and no snapshot creation is in progress.
The disk has not been released.
If the disk is attached to an ECS instance, the instance is stopped.
If the disk is part of a dynamic extended volume or RAID array, all services and applications using the disk are stopped with no active I/O operations.
Procedure
Disk rollback is irreversible. All data changed between the snapshot creation time and the rollback time is permanently lost. Create a snapshot of the current disk state before proceeding. See Create a snapshot manually.
The following steps use the Snapshots page. You can also roll back from the Instance page.
Go to ECS console - Snapshots.
In the upper-left corner of the page, select a region and resource group.
On the Disk Snapshot tab, find the target snapshot. In the Actions column, click Roll Back Disk.
In the Roll Back Disk dialog, select the confirmation checkbox, and click OK.
What to do next
After a rollback, the disk reverts to the exact state captured in the snapshot. Be aware of the following impacts:
System disk rollback: host configuration reset
After a rollback, all data that was added, removed, or modified since the snapshot was created is lost. This includes host configuration settings such as the hostname, SSH settings, passwords, network settings, system software repository settings, and clock source. These settings revert to the state they were in when the snapshot was created. You must reconfigure them based on your business requirements.
Note
Rolling back the system disk of an ECS instance does not affect the current SSH key pair, username, or password. You can continue to use them to log on to the instance after the rollback.
Data disk rollback: disk size may revert
If you resized a data disk after creating the snapshot and then use that snapshot to roll back the disk, the disk reverts to its original size at the time of the snapshot. The additional disk space is lost. To restore the disk to the expanded size, log on to the ECS instance and resize the disk again:
Linux instance: Extend the partitions and file systems of disks on a Linux instance
Windows instance: Extend the partitions and file systems of disks on a Windows instance
FAQ
How do I restore a specific file or folder without rolling back the entire disk?
Disk rollback restores the entire disk. To restore specific files from a data disk, use a temporary disk:
Create a temporary pay-as-you-go disk from the snapshot. See Create a data disk from a snapshot.
Attach the disk to an ECS instance. See Attach a data disk.
Bring the disk Online.
Copy the target files or folders to the original disk.
Detach and release the temporary disk. See Detach a data disk and Release a cloud disk.