In a previous post, we looked at instant VM Recovery. Now we’ll take a look at Full VM Recovery.

How restoration works

Veeam Backup & Replication lets you restore a virtual machine using a backup file. During the restore step, you can select the last restore point to obtain the latest state of the VM (last saved state). It is of course possible to select an earlier restore point.

During the restore operation, the VM image is extracted from the backup. It is then copied to production storage. This type of restoration takes longer than an Instant VM Recovery. Nevertheless, it offers the advantage of not requiring any additional operations following the restore. Once the data has been restored to storage, the VM is saved to the desired ESXi host. A complete VM restore can be performed using one of the following modes:

• Direct access to storage: this mode is used if the backup proxy is connected directly to the SAN array (or has access to the NFS datastore). The Veeam Data Movers component present on both the repository and the backup proxy enables VM data to be recovered via the backup file. It is important to note that Veeam Backup & Replication can only restore thick disks in this mode. Thin disks will be restored via the virtual appliance or network mode.

• Virtual appliance: It’s important to note that with this mode, the backup proxy must be positioned at the ESXi host that hosts the restored VM. The proxy is installed on a virtual machine, and during backup, the VM disks concerned are added to the proxy. The proxy then proceeds to backup or restore these disks. In this way, data transfers over the network are avoided, resulting in no disruption to production.

• Network: If the first two modes cannot be used, the network transport mode can be used. Both backup and restore use the network to transfer data from the ESXi to the Veeam repository and vice versa. It is very important to note that this mode is very costly in terms of network resources. The performance of the latter will be reduced during backup and restore. Wherever possible, this mode should only be selected as a last resort.

The VM can be restored to its original location or to a new location. If the original location is chosen, the production virtual machine is first stopped, then the disks present in the backup file are restored. Restoring to a new location offers the option of configuring a new name for the virtual machine. Host and datastore selection can be carried out using the restore wizard. It is also possible to select the desired disk format (thin or thick). Finally, network properties can be configured. It is interesting to note that when restoring a VM with multiple disks, restore operations on these disks are performed in parallel.

Restore a VM

From the Veeam Backup & Replication console, select the Home view and click on Restore. In the context menu, select VMware vSphere.

A wizard appears, click on Restore from backup.

Select Entire VM restore in the Restore from backup window.

Click on the Entire VM restore option in the wizard

Click on the Add button to select the desired resource

The two available options were detailed in the previous section (The different recovery modes – Instant VM Recovery – Restoring a VM). You can change the view (VM and templates, Host and cluster, etc.).

Select the virtual machine and click on Point. Select the desired restore point, then click OK.

Select the restore mode and click on Next.

Enter a reason for the restore and click Next. A message appears warning that the restored VM will be stopped during the restore operation. Click OK to continue the operation.

Click Finish to proceed with the restore. You can see from the Log tab which mode was used for the restore. At the end of the restore, click on Close. The VM has been restored.