Automated Testing | ![]() |
Overview
Automated testing provides a reasonably realistic approximation to an actual experiment run. The server and all the clients are run in seperate Java virtual machine instances, meaning they can be run on separate machines in the lab. Once started, clients proceed without further user interaction; they make random moves as appropriate.
Running an Automated test
- Set the defaultMode property in conf/mult-client.properties to automatic for each machine taking part in the test.
- Build Multistage as described in setup.
- Execute edu.caltechUcla.sselCassel.projects.multistage.server.StartServer and set up the parameters as you would for a normal experiment.
- On each client, run edu.caltechUcla.sselCassel.projects.multistage.client.StartClient.
Fixme (diffusor)
TODO actually run through this process
The test should proceed to completion, at which point the server will varify the validity of the results.


