How to restore individual files when FLR fails

To restore the files:

1. Launch Instant Recovery of the Virtual Machine you need to retrieve files from making sure to use “Restore to a new location” and leaving the "Connect VM to network" option disabled. See Performing Instant VM Recovery.
   Note: You will need to rename the Virtual Machine when performing the Instant Recovery. It is suggested to add a suffix, such as "_restored", etc.
2. Once the Instant Recovery is finished, in VMware vSphere Client, create a new virtual disk for the VM. The disk should be created on a different, production datastore instead of with the VM as the VM is running on the Veeam NFS mount. The disk needs to be large enough to contain only the files you wish to restore 

   

3. Use the VMware console to log in to the recovered VM’s guest operating system. As this is not on the network, you will need to use local credentials or cached credentials.
4. Initialize and format the disk as you would any other disk in that operating system.
5. Transfer the data you want to restore to that disk.
6. In the vSphere client, note the name and location of the disk, then remove it from the instant recovered VM.
   Note: After this step, you can stop publishing the Instant Recovery at any time.

   
   

7. Attach that disk to the machine in Production that needs these files.

   
   

This method can be used to restore any files from any file system, which helps bypass limitations of Windows FLR and multi-OS FLR.

Instant Recovery for VMware does not require pre-allocating disk space on any datastore. The vPower directory specified in the repository settings is presented directly to the ESXi host as an NFS datastore. Instant recovery can usually be accomplished in a matter of minutes, but recovery of large VMs from deduplicating storage can sometimes take over an hour.

Instant Recovery can also be used to recover files in Hyper-V. However, Instant Recovery for Hyper-V will pre-allocate the entire size of the virtual disks.