The functionality of recovery points allows creating a snapshot of a disk at a specific point in time. This snapshot can be used in the following ways:
Creating a recovery point can be done while the virtual server is running and has no impact on its operation.
The created disk image is consistent for a given moment in time - first, a snapshot of the disk is created, and then its content is copied to separate physical media.
As creating a recovery point requires copying data from the original disk, this operation is time-consuming. To optimize time and costs for the user, only the areas of the disk that have changed since the previous recovery point are copied. This means that the first recovery point requires copying the entire disk content, and each subsequent one will only copy changed areas (and thus occupy less space).
The rollback operation allows restoring the contents of the original disk to the moment when the recovery point was created. This operation - because it modifies the disk content - cannot be performed on a disk connected to a running virtual server (it must be turned off). This operation is irreversible - the current disk content will be overwritten.
The cloning operation allows creating a new disk whose contents are the data from the recovery point. The newly created disk is available immediately, without waiting for data to be copied, which are copied gradually (copy on read).
Due to the fact that the performance parameters of the space used to store recovery points are different from normal disk space, the performance of the newly created disk may be lower on the first access to it.
Cloning of recovery points is possible only in the location where the recovery point was created.