UpCloud provides several ways to restore data from your disk backups. Most commonly you can either revert the entire disk to a previous backup snapshot, or you can clone the backup to a standard disk and restore individual files.

Restoring a backup replacing the current disk

This method will overwrite the existing disk entirely with the selected backup. All data on disk will be returned to the point at which the backup was taken. This is very useful for reverting mayor changes to your current system.

Note that your cloud server must be shut down during the backup restore procedure.

Start by going to your Server Settings at UpCloud Control Panel and open the Backups -tab. At the bottom of the page you’ll see a list of backups that have been taken of this server. On each backup line you can find the date and time of when the backup was taken and controls to restore or delete the backup.

Restore backup

Press the Restore -button to open the restoration options.

Select the Restore the storage resource with the backup -option, and then click the Restore backup -button to confirm the action.

Restore replacing current

The backup restoration might take some time depending on your storage size and type, MaxIOPS is considerably faster at all storage procedures than a HDD of the same size. You can follow the progress on the Backups -tab at the bottom of the list. Once the restore has finished, you can start the server as normal.

Cloning a backup to a new storage disk

It is also possible to clone a new standard storage media from a backup. You can then attach it to the original server, or to any other, as a secondary disk. Attaching a backup clone to your existing server allows you to access individual files from the disk without needing to revert the entire system.

To start, go to your Server Settings at UpCloud Control Panel and open the Backups -tab. Find the backup you wish to access on the list at the bottom of the page and click Restore.

Restore cloning

Select the Clone a new disk from the backup -option, and click the Start the cloning -button.

In the following setup dialog, give the new disk a name, something you can easily recognize later when attaching it. For disk type the MaxIOPS is ideal for a temporary backup clones, as it is much faster at data transfer than the HDD option. Keep the clone on the same zone as the server you wish to access the backups and press Accept to start the cloning process.

Clone the backup

The cloning will take some time depending on the size of your backup and the disk type you selected as target. You can check the status at your My disk resources section. While cloning, the disk actions will show Disk operation in progress, when the operation has finished you’ll see the same controls as for the other storage disks listen on the same page.

Note that your cloud server must be shut down before attaching new storage disks.

Once the cloning is complete go back to your server’s settings and open Disks -tab. Under attached disk resources you’ll find an option to Attach new disk.

Attached disk resources

Click on the line that says Attach new disk, this will open a new setup dialog for creating and attaching storage devices.

Attach existing disk

Select to Attach Existing disk, find the disk clone you made earlier and select it as Disk Resource, keep the Disk Controller set to virtio for best performance, and finally click Accept to confirm the action.

Afterwards you’ll see the cloned disk shows up as an attached disk resources, you can now start up the server again.

Linux servers will require you to mount the new disk to access the files from the backup clone. Do this by first creating a mounting point to your file system with the following command

sudo mkdir /media/backup

Check the disk name and partitions you just attached to the server with

lsblk -io KNAME,TYPE,SIZE

The output will show a list of disks usually named like vda, vdb or vdc and their partitions such as vda1 and vdb1.

If you’re not sure which of the partitions is on the cloned backup, you can check your current system partitions with this command

df -h

The command will list the partitions and their sizes currently configured on your server. Select a partition from the lsblk printout that is not yet in use, for example if your system partition is called vda1, select vdb1 for mounting.

Then simply mount the new storage disk partition to the directory you just created with the following command

sudo mount /dev/<partition> /media/backup

After mounting the disk you can restore any and all files you wish by simply coping them over.

Windows servers will also need you to take a couple of steps before you are able to access the backup files. Start by connecting to your server through Remote Desktop. Once logged on type in diskmgmt.msc to the Windows search and open it by pressing Enter.

In the Disk Management you’ll see your previously existing disks and the newly added backup clone which will show as offline. Right click the clone disk and select Online.

Disk manager

Depending on your Windows server version, you might get a pop-up asking to scan the new drive, you can skip this by clicking Continue without scanning.

With the disk online, you can now access it normally through the file browser to copy over any files you wish to restore.

Removing extra storage disks once you’ve got everything sorted is as simple as shutting down the server again, detach the backup clone and delete it. Go again back to your server’s settings and open Disks -tab.

Under attached disk resources you’ll find an options to Detach and Delete.

If you do not wish to delete the backup clone just yet, you can simply use detach and leave the disk for later use.

If then again you do not need the cloned disk any longer, you may just delete it and the disk will get detached in the process.

Don’t forget to make regular backups of your server again after restoring files.

Questions and Answers

Why it is not possible to restore individual files automatically?

We handle storage disks and their backups as full media images. Access to individual files would require knowledge of used partitioning, file system and possible disk encryption. Due to this it is not possible for us to either restore backups for you or provide GUI access to browse contents of the backups.

Why I can’t attach backup directly to my server?

Backups are stored in separate storage systems which maximize storage capability over IO and access properties. Due to these limitations it is not possible to directly attach backup as a disk. Instead a new standard storage can be created by cloning the backup (note that cloning is usually possible only to the same storage tier as the original disk image was from).