About Execution Environments and Agents
Execution environments are the workhorses of DAI. They require network access to your DAI Server, and you connect them to your Server on the DAI Environments page. Execution environments are were test developers author scripts ("snippets"), connect the snippets to DAI for use in models and scripted tests, connect to systems under test (SUTs), and run tests. An execution environment can be either of the following types:
- A DAI Design Environment, which consists of a DAI Design Agent and an Eggplant Studio instance (also known as "Eggplant Functional" or "EPF") running with its graphical user interface (GUI).
- A DAI Run Environment, which consists of a DAI Run Agent and an EPF instance running in command line mode (also known as "headless" mode).
See the DAI Components page for a high-level overview of all of the components comprising DAI.
What are the DAI Design Agent and DAI Run Agent?
In an execution environment, the agent makes the connection between Eggplant DAI and Eggplant Functional.
Within the DAI Design Agent, we offer the possibility of managing your suites within our internal storage. You can also set up a Git connection to manage your suites. The Design Agent provides suite management support only within the internal storage. Please see the Git integration docs for more information on how to manage suites coming from your Git repositories. The DAI Design Agent can only run Live Run tests.
The DAI Run Agent is a lightweight command-line application used solely to run tests without needing the other features that the DAI Design Agent offers such as managing suites. The DAI Run Agent can run from the DAI Runner, from the command line (using a batch file, for example), or from a DevOps pipeline.
With the DAI Design Agent or DAI Run Agent software running, and the execution environment connection active, you can run either Script-based tests or Model-based tests:
- For Script-based tests, you can create a list of test steps by adding script-based test cases associated to the suite which you uploaded. Read more here
- For Model-based tests, you can attach SenseTalk snippets to actions and states in any associated Eggplant DAI model. When you run the model, it executes the SenseTalk snippets against the connected SUT. The test results can then be viewed in the console in the Model workspace.
You can set up and switch between execution environments in Eggplant DAI to run different sets of tests.
The agent can run on the same machine as the Eggplant DAI server, or any local or hosted machine that can connect to the Eggplant DAI server via the network. To learn how to set up an agent in an execution environment, go to Configure an Execution Environment.
For SenseTalk snippets/scripts to execute, Eggplant Functional must be installed on the same machine as the DAI Design Agent or DAI Run Agent software.
In Eggplant DAI 6.0 and later, the agents use a secure access key to authenticate with DAI's identity and access management system (Keycloak).
For DAI Eggplant Cloud (Hosted DAI) users, the agent, and Fusion Engine are already included and so are the Cloud SUTs. Contact your account manager or ep-sales@keysight.com to request for DAI Eggplant Cloud.