I swapped out the offending line with:
to get rid of any possibility of race conditions.
Running `cassandraCluster.ring();` before and after the shutting down and starting up of the node shows the cluster switched to vnodes on restart. Not sure if this affects the test result.
That's when the session.execute() command returns the following error:
Real world use case: If a cluster is running and a network partition happens, or the cluster really shuts down, on restart, it doesn't appear as though the driver will connect again without manual intervention.