In spark_ec2.py#355 one can see that no zookeeper nodes are reated and neither do docs on "Running Spark on EC2" mention options related to creating a zookeeper nodes.
Is it a legacy or on the contrary, something not implemented yet?
Can this script be used to create a cluster with zookeeper installed and running?