How to restore VMDK disks that are part of a LVM on the original host

Summary

You cannot attach a restored VMDK disk, mount the file system and copy files, due to LVM and the use of UUIDs for logical volumes.

Step By Step

1. Restore the VMDK(s) and attach them to the VM as “Existing hard disks”.
2. Scan the SCSI bus:
   1. echo "- - -" > /sys/class/scsi_host/host0/scan
3. List and identify the restored disk devices by running: lsblk.
4. Run the command to import and rename the duplicated volume group, by default an incremental suffix to the original name will be used:
   1. vgimportclone --basevgname <VolumeGroupName> <PhysicalVolume> [PhysicalVolume...]
5. Verify the logical volumes by running: lvdisplay.
6. Make the logical volume(s) available:
   1. lvchange -ay /dev/<VolumeGroupName>1/LogicalVolumeName>
7. Mount the restored logical volume to a temporary mount point to access the file system:
   1. mount -o nouuid /dev/<VolumeGroupName>1/<LogicalVolumeName> /mnt
8. Unmount the restored logical volume when the restore operation is complete:
   1. umount /mnt
9. Remove the restored volume group and physical volume:
   1. vgremove <VolumeGroupName>1
10. Delete the restored VMDK(s) from the VM.
11. Perform cleanup of the restore job in ECX.