The notes below provide descriptions of the new features and changes introduced with each release of Eggplant DAI. You are strongly encouraged to read about the relevant changes whenever you upgrade from an earlier version.
Potential compatibility issues are highlighted in RED.
- Windows: Windows Server 2019, 2016 (recommended for production use), Windows 10, Red Hat Enterprise Linux (RHEL) 7.6/7.7 (supported in Eggplant DAI 4.0.1 only)
- RAM: 8 GB minimum
- Disk Space: 4 GB minimum
- Processor: 64-bit Intel Core 2.3 GHz or equivalent (4+ cores or 8+ threads recommended)
- When designing a model to run with the Cloud SUT in Eggplant DAI, make sure to add a waitfor command in the first snippet that gets executed, so that it waits for the OS user interface to load before interacting with the SUT.
- When upgrading to Eggplant DAI 5.1, it is recommended to update your version of Eggplant Functional to version 21.1.
- From version 4.2.x, Eggplant DAI is only compatible with Eggplant Functional 20.2 and later.
- As of version 4.1, Eggplant DAI is no longer compatible with Eggplant Manager.
Release 5.1 (1 February 2021)
- Adds an additional dashboard called QA Summary.
- Introduces a new method to run tests via Script Execution.
- The Models tab in Designer now provides the following additional functions:
- Fixes an error that would occur when typing the model name on the Capture page in Designer.
- Resolves an issue that slowed down the Modeler UI when opening a model that had been run a large number of times.
- Fixes the error that occurred while changing the model's group for test case creation.
- Resolves the issue which required entering modelId as a required parameter along with modelName and group parameters when running the Run Execution API endpoint.
- Fixes the issue with Coverage Report missing the latest model changes.
- Resolves the issue with test cases inside a submodel being labeled incorrectly when run from the parent model.
- We now enforce unique model names across systems. For example, if you have models of the same name with one in a different group, then they will be renamed to MODELNAME_UUID. You can choose the Save Model As menu option to save the model as your preferred unique name.
- We recommend using the Models tab (Designer) to import models instead of the Modeler UI due to a minor issue being found with re-importing a model and saving it with its original name.
Release 5.0.0 (18 November 2020)
This release is applicable to DAI Eggplant Cloud Customers only.
Adds capability to run end-to-end test execution in Eggplant Cloud. You can now run functional tests against a web application using Eggplant software without having to install the Fusion Engine locally or configure desktop machines for browser instances as SUTs.
When running the Run Execution API endpoint, along with entering modelName and group parameters, you also must now enter modelId as a required parameter.
Release 4.2.3 (29 October 2020)
Fixes an issue where error messages displayed with inaccurate descriptions in the Modeler console.
- Fixes an issue that stopped the Coverage Report from displaying.
- Resolves an issue where the DAI server failed to start due to incorrect allocation of engine processes.
- Fixes an issue to ensure that a valid error description appears on the Runner page when a test run results in error.
- Resolves an issue with model execution resulting in an infinite loop when the model has no error state.
- Fixes an issue with not being able to stop a model after adding a global variable.
- An error message appears when editing Eggplant Functional snippets from a suite that is not downloaded locally i.e., not available via the Manage Execution Environment window > Manage Suites option. To avoid seeing the error, download the required Eggplant Functional suite from your DAI Agent.
- The DAI Dashboard doesn't fully load the Coverage Report and Run Report tiles on Google Chrome when you hover over them.
Release 4.2.2 (11 September 2020)
- Fixes an issue with upgrading Eggplant DAI 4.2 over an existing installation of Eggplant DAI older than version 4.1.
- Resolves an issue with running tests in the Modeler interface for users with non-admin permissions.
- Resolves an issue with passing a non-string type of value via an Eggplant Functional snippet back to Eggplant DAI.
Release 4.2 (26 August 2020)
- Adds a new Screenshots tab in the Test Results tab > View Details pane to view and download screenshots for steps related to your model execution.
- Adds new agent installers for Linux OS: CentOS/Red Hat and Ubuntu/Debian.
- Adds a new Status tab on the System page that provides additional information for Eggplant DAI administrators. You can view the status of all the Eggplant DAI components, including the connected agents, from here.
- The Eggplant DAI UI now includes in-app help content.
- Eggplant DAI licenses can now be renewed with an extended expiration date for a license key including the same serial number that was installed before.
- Improves execution of models and suites.
- Improves database performance.
- Fixes an issue where failed events were not populating in the database for submodels.
- Resolves an issue which showed only the Selenium screenshot in the console when a test with VNC and Selenium connections failed.
- Fixes an issue with the agent not running the test after recovering from a closed websocket connection.
- Fixes an issue with the agent where it tried to read the SSL certificates, even when the SSL Connection checkbox wasn't ticked.
- An error message appears when you try to edit a snippet on a connected agent and if that suite doesn't exist in the agent's suite location (either because it's been deleted, or the connected agent never had it).
- The Eggplant DAI Agent may crash when using it on a Mac machine with external monitors connected. As a workaround, use the agent on the Mac, without moving the agent window to the external monitor.
- Previously, the iterations for a model could be set to zero to run a model indefinitely. This behavior has been removed as part of our performance and scalability improvements.
Release 4.1 (12 May 2020)
- Adds the capability to upload suites to Eggplant DAI and to associate them with models for use in test execution. This feature also adds support in the following areas:
- Execution Environment: Provides an option to upload and download suites via the AI Agent so that you can manage scripts and image assets associated with your models. If an execution environment has previously been used with a specific suite, only differences need to be uploaded, which helps speed up the execution.
- Test Execution: If a model has a suite attached, a test configuration can be executed with an execution environment without any previous access to that suite.
- Suites Tab: View all suites that are available (via uploading suites in AI Agent) in your instance of Eggplant DAI.
- Models Tab: Associate suites to a model or modify an existing suite association.
- Updates the Eggplant AI API:
- Adds endpoints to associate a suite with a model.
- Improves the Composer endpoint to allow you to:
- Delete an asset from Eggplant DAI and any associated model asset from the AI API.
- Get models and any associated suites.
- Run and Abort endpoints now allow you to:
- Cancel the suite download on an Agent.
- Specify an asset version while downloading it.
- The Eggplant DAI UI now reports errors when a SUT Connection fails.
- Links to Test Results in the Dashboard are now filtered by model.
- Fixes a performance issue that caused delays with starting the execution of test cases.
- Fixes an issue where some test configurations with failed statuses reported as passed.
- Fixes an issue with previous test cases not clearing in a test configuration after the model selection is changed.
- Fixes an issue with the Agent Command window appearing in debug mode.
- Resolves an issue with SUT connection failing Test Case steps.
- Improves error handling for large volumes of scheduled test configurations.
- Fixes an issue with agent that showed tests that weren't executing in In Progress state.
- Resolves the issue with the agent not running test configurations despite being set to Run mode.
- Test configurations no longer get stuck in the Canceling state.
- After upgrading to Eggplant DAI, it is recommended to:
- Wait for some time after the installation of Eggplant DAI to launch it in your browser, to ensure the database and services are up and running.
Perform a hard refresh on your browser.
- Do not enable the Eggplant Functional GUI check box from the agent when using the Run mode, if the Eggplant DAI GUI is running several tests on an Execution Environment (e.g., a scheduled run in Eggplant DAI > Runner tab) as continuous opening and closing of SUT connections may cause problems with Eggplant Functional.
- If a suite is uploaded without helper suites and modified later to refer to Helper suites, then you should follow this sequence while uploading the suite:
- Delete the main suite that has a dependency on the helper suite.
- Add the helper suite and then upload the main suite back.
Release 4.0.1 (17 February 2020)
- Runner enhancements:
- Adds a new in-progress icon if the status of the test configuration is in progress.
- Adds the option to view detailed results of passed, failed, and in-progress test runs along with a donut chart representing the total number of passed, failed, and incomplete runs.
- The Eggplant DAI Installer includes an option that allows Eggplant to collect usage statistics from the users. You can disable this option by clearing the Report usage statistics checkbox (only available if the Run Advanced Install Configuration checkbox is enabled).
- You can now download server logs with a single click in Eggplant DAI.
- Test configurations now resume after suspending or restarting a machine.
- The ping messages on the Agent dialog are displayed as before.
Selecting the Run cleanup after abort option in the Settings dialog doesn't initiate the cleanup script if you abort your test run.
- If you have test configurations running simultaneously with SUTs that are not tied to a specific Execution Environment (agent), you may get an Unable to make a connection to the Agent error. Therefore, it is recommended to separate your test configurations by at least one minute, or use SUTs tied to specific Execution Environments.
Release 4.0 (20 December 2019)
This is the first release of Eggplant DAI.
- Schedule and track test executions through the Runner tab.
- Create and execute Test Configurations that combine directed test cases with exploratory testing.
- Limit exploratory testing by duration, number of iterations, or a coverage target so that tests run until one of these conditions is met.
- Configure and reserve systems under test (SUTs) for testing.
- SUT connections don't disconnect after task execution. This can cause problems for users with RDP configurations. To work around this issue, either manually disconnect from the Connection List before the next test run, or add a disconnect snippet to your model.
- The agent isn't receiving ping messages from the server.