If you’re running vSphere Essentials Plus or higher editions, you may have noticed the ‘VMware vSphere replication feature. Here’s everything you need to know to use this feature.
What is VMware vSphere Replication, anyway?
The replication offered by this feature allows you to configure on a per-VM basis, in as little as 5 minutes. It’s fully compatible with a variety of datastore types, from VMFS through vSphere virtual volumes, NFS and more. It’s a great option for an automated disaster recovery protection that is not only dependable but also integrates well. It’s budget-friendly, offering storage at an outside site for additional insurance that your precious data is not lost, and can even handle a multitude of VMs concurrently. You don’t need to know that this is commonly called ‘boot storm’ to realize this is the most likely scenario you will face in recovery from a disaster. If you’re looking for some nitty-gritty stats, try recovering 1000VMs, from a 4-node all-flash SAN cluster, in under half an hour.
How do I make the most of this?
This VR method is already configured for each VM, meaning you have precision control each time, regardless of how your storage is actually configured. You can pick the storage policy that will be applied to recovery as you store each VM. Figuring out the policy applied at the target location before recovery can be a lot more tricky. You can’t see from the user interface of the vSAN what’s being assigned. You can get around this using an RVC command line interface, however.
We ran a basic test. While we won’t bore you with the exact details here, but 5 stored VMs, all using the FTT=0 rule, were placed into the location. RVC reported this all correctly. However if you select the ‘reconfigure’ option from the VR and change the storage policy, nothing happens- the original policy remains in place. At least, that is what appears to be the case- when the VM was recovered, however, the newly applied policy was assigned.
What does this mean for recovery?
The long and short of this is, of course, that when you reconfigure for a different policy on the VM the replication does not change the policy immediately. It will only be assigned when the VM is actually recovered. So the VR replica assigns the policy at initial configuration, and any reconfigurations will not take place until the VM is recovered, at which point the new policy will come into place.
