One of our client wants a zero RPO/Automated failover capability in a storage cluster for VMWare ESXi VMFS datastores.
Client has deployed two storage arrays within the same datacenter and in different racks
If one storage become offline for some reason and other storage should take-over all the workloads without any downtime.
Is it possible? any inputs please?
That is called a vSphere Metro Storage Cluster. Your array should support metro clustering however in order for this to work. The above linked document, which I wrote for the most, goes over the vSphere side of things when it comes to the configuration and failure scenarios etc. All of course depends on the storage that you have available.
You need a storage array(solution) which offer a transparent automatically fail-over or a solution on top of your storage like the DellEMC VPLEX or IBM SVC.
There is nothing in vSphere which can help you with your special requirement and setup.(i leave FT 2.0 and vSAN out of the game). The vMSC stuff mentioned above doesnt include a technical storage solution. It justs add the think abouts when it comes to side awareness and so on when dealing with a stretched cluster.