Murrstock - stock.adobe.com
From post-breach recovery operations to workload migrations, there are plenty of reasons why Azure admins must restore VMs.
Walk through the most common methods to restore Azure VMs, how each one works and why you may want to use one restoration method over another.
Why restore Azure VMs
There are several scenarios where Azure administrators might need to restore a VM:
- Workload disruption. If production VMs are disrupted or breached, restoring VMs based on backups is the most straightforward way to resume operations.
- Workload migration. Restoration based on a VM's disk or image can be a simple way to move a VM from one Azure cloud region to another. You could use a tool, like Azure Resource Mover, to accomplish a region-to-region migration. However, it doesn't support all VM types so occasionally you need to perform a manual restore.
- Suspended VM. Restoration makes sense if you stopped running a VM because it was only for testing purposes, but then want to restart it. The simplest method is to restore from a backup image.
Methods to restore Azure VMs
Azure supports several methods to restore VMs. Here's a summary of the most common ones:
Restore to a new VM
You can restore a VM by creating an entirely new VM based on a restore point from a previous VM. You can accomplish this only if you have an available restore point.
This approach gives you a complete, standalone copy of your original VM. It's the best way to replicate an existing VM. It also allows you to keep the original VM in place and have it continue to operate.
Replace an existing VM
Under this approach, you create a snapshot of your existing VM and then launch a new VM to replace it. To do this, you need to have access to the original VM. If you deleted it or otherwise can't access it, this VM restore method won't work.
The advantage of this method is that it doesn't change the total number of operating VMs. It also doesn't increase the chances that you'll accidentally leave multiple copies of the same VM running, which wastes money. However, it doesn't allow you to launch additional copies of the VM. This is not a good method if you're looking to scale infrastructure to accommodate additional instances of your applications.
Restore Azure VM disks
This restore method makes sense if you need to restore the entire file system of a VM, but the rest of the VM is intact. Rather than restore an entire Azure VM, you can replace just the virtual disks attached to the VM. The only component of the VM that changes under this method is the disks associated with it.
This method is useful if an OS problem corrupted a VM's file system and you want to restore to a stable copy of the file system.
Restore individual files
If individual files used by the VM were corrupted, but the file system is otherwise intact, you can restore specific files. This method is valuable if specific files were accidentally deleted or removed by attackers. It's also faster than restoring the entire disk. However, to use this method, you need to know which files were damaged or removed. If it is not apparent which files, or how many files, were affected by a file system problem, it's safer to restore the entire disk.
Restore across regions
Azure enables IT teams to restore VMs to a different Azure cloud region. This approach is useful under two scenarios:
- The Azure region hosting your VMs had an outage, and you want to run the VMs in a different region to restore operations.
- You want to make your VMs available in a backup region to prepare for the possibility of an outage in your primary region.
However, running Azure VMs in multiple regions at the same time will double your costs. While this approach offers strong resilience benefits, it may strain your budget.