Persistence configuration - when node down all new connections permanently in the backup node, even if the previous one goes up again.

You could also configure the the pool to use priority group activation and the monitor on the pool to have the Manual Resume feature enabled. This would send all traffic to your designated primary pool member, and if that pool member goes down, the Manual Resume feature will prevent it from re-activating even if the service on that member comes back up.

The downside of this technique is that you will have to manually reactivate the first pool member, but it gives you pretty solid control over which pool member is receiving traffic.