1.
You know, you can run persistent workflows also in a transient engine.
So I would create a transient engine, e.g. using org.copperengine.core.tranzient.TransientEngineFactory and use mocks in the workflows.
If you want to make sure, that all the data in your workflows is really serializable, but you don't want to use MySQL, Oracle or Postgres for your tests, you can also create a peristent engine with an in-process derbydb or H2.
org.copperengine.ext.persistent.RdbmsEngineFactory might be helpful.
2.
Check the JMX interfaces, e.g. ProcessingEngineMXBean