The best configuration for Citrix NetScaler in an active-passive data center for disaster recovery is by using the global server load balancing feature for all resources that need to be accessed in the event of a failover to the passive data center.

But a standard high availability would fit this scenario as well when the backend environment has the same setup as the active data center. This means that there are no servers active in the passive data center. The servers will become active only when the active data center stops working. The servers in the passive data center side will need to be started manually by the hypervisor environment.

To use the high availability feature on Citrix NetScaler instead of global server load balancing, we need to be sure that the servers are configured with the same configuration as in the active data center:

So basically, the virtual machines in the active data center should be running in the passive data center if the active data center stops working.

If the server configuration in the passive data center is different from that in the active data center (for example, because it's a copy), then we should use the global server load balancing feature. This feature allows us to configure Citrix NetScaler with two different configurations for both data centers.

So, we need to configure a Citrix NetScaler Enterprise appliance in both data centers and configure these to match the configuration.

After this configuration, we're going to configure the global server load balancing virtual servers. During the configuration, we should configure the passive data center load balancing virtual servers as backup failover servers. This means that the passive data center will become active only if the active data center stops running.

The global server load balancing feature is especially useful when there are two or more data centers available with different configurations.

As you can see in the preceding figure, the configuration is almost the same as that of the active/active data center for application hosting and the active/passive enterprise VDI environment.

The only main difference between the active/active data center for application hosting and this scenario is the failover between the data centers configured on the load balancing virtual servers and the actually active/passive configuration of GSLB.