We are planning on building a Workflow farm to use with an existing SharePoint 2013 farm. We have 7 servers in total split between 2 physical locations:
Location 1: Server 1, Server 2, Server 3, Server 4
Location 2: Server 5, Server 6, Server 7
The requirement is to have high availability in case one data centre goes up into the air. This would be the perfect for disaster recovery scenario, but at the same time we would like to utilize all of the servers as well and not keep half of them running but being useless at the same time.
Since Service Bus Farm can either be 1,3 or 5 servers, the following configuration should work:
Location 1: Server 1 (WF + SB), Server 2 (WF + SB), Server 3 (WF + SB), Server 4 (SQL, Clustered)
Location 2: Server 5 (WF + SB), Server 6 (WF + SB), Server 7 (SQL, Clustered)
What would happen if all servers in Location 1 would go down? That would leave us with 2 WF servers (2 Service Bus servers as well) + 1 SQL running. Would the whole Workflow Farm stop working since there would only be 2 Service Bus instances instead of supported 1, 3 or 5? What would be the best way to utilise those 7 servers in 2 physical locations to squeeze the most of them?
I assume that in case of a disaster there would be no need to restore the databases as clustered SQL would take care of that?