If you want to run graphical user interface tests with a locked screen, i.e. test keyboard or mouse operations without a user logged in to the agent computer, see chapter "Unattended execution".
How Tosca Distributed Execution works
A tester creates a TestEvent in Tosca Commander.
Once the tester presses Execute now, the system forwards the TestEvent and its ExecutionLists to the Tosca Distribution Server.
The Tosca Distribution Agents synchronize with the Tosca Distribution Server on a regular basis.
Once an agent contacts the server, the server hands over the first ExecutionList of the TestEvent to this agent. As soon as the next agent contacts the server, the server hands over the second ExecutionList of the TestEvent to this agent.
If there are no free agents contacting the server, the system puts the ExecutionList into a queue until an agent becomes available.
Once a Tosca Distribution Agent has executed an ExecutionList, it saves the results to the common repository.
Tosca Distribution Agents
Tosca Distributed Execution works with two types of Tosca Distribution Agents:
Distributed execution with Full Execution Agents:
Create TestEvents and execute TestEvents. You can create and execute TestEvents using classic engines or mixed TestEvents.
Distributed execution with Slim Execution Agents:
Slim Execution Agents are execution-only agents and exclusively for TestCases created with 3.0 engines.
Full Execution Agents
Full Execution Agents can execute TestEvents created with classic engines and/or mixed TestEvents. They work with Tosca Unattended Execution and Tosca Continuous Integration.
Slim Execution Agents
Use Slim Execution Agents if you want to save memory and disc space.
These agents are execution-only agents and exclusively for TestCases created with 3.0 engines. You can use Slim Execution Agents with Tosca Unattended Execution and Tosca Continuous Integration, provided that you are not running mixed and/or classic tests.