Set up Elastic Execution Grid
Once you have created your account, you can set up Elastic Execution Grid:
Set up Elastic Execution Grid Agents. Agents Agent machines are the machines on which you execute your tests. They host the Agent Service and the Runners. The Agent Service receives the Flow from the server and directs it to the required Runner. The Runner actually executes the test. are the machines on which you execute your tests.
If you use Tricentis Tosca, set up the Tricentis Tosca integration.
Set up Elastic Execution Grid Agents
Agents are the machines on which you execute your tests.
Agents are copy-deployable, so you don’t have to install anything specific for Elastic Execution Grid.
Setup considerations
Your Elastic Execution Grid Agent machine must meet the minimum system requirements.
Tricentis recommends that you set up only one Agent per machine.
You can set up a maximum of 50 Agents.
Since your Agents execute your tests, you do need to install the respective test framework. For instance, if you want to run Tricentis Tosca tests on an Agent, you need to install Tricentis Tosca.
Setup process
To set up your Elastic Execution Grid Agents, perform the following actions:
1. Download the Elastic Execution Grid Agent.Download the Agent package from the 
Downloads tab.
2. Get the Client Secret.Every Elastic Execution Grid Agent needs a Client Secret to access Elastic Execution Grid. You can use the same Client Secret for multiple Agents.
You will later enter the Client Secret into the Agent configuration.
If you are not an Okta Third-party identity provider for user management and authorization. admin, ask your Okta admin for
If you are an Okta admin, follow the steps below:
Go to https://<name of your cloud instance>-admin.oktapreview.com/.
In Okta, go to the Applications page as described in the Okta documentation.
Click on the Elastic Execution Grid Agents application and copy
3. Integrate the Agent into Elastic Execution Grid.To integrate an Agent into the Elastic Execution Grid environment, follow the steps below:
Open the appsettings.yaml file, which is in the Elastic Execution Grid Agent download package.
Modify the following AgentConnection settings:
Setting | Value |
|---|---|
Tenant | |
ClientSecret | Enter the Client Secret of the Elastic Execution Grid Agents Okta application. |
Identifier | Optionally, change the Agent identifier. You don’t have to do this; the default value already makes the Agent name unique. Allowed values are alphanumeric characters, underscores, or dashes. The identifier must have a minimum of 3 and a maximum of 63 characters. |
ExecutionConfirmation | The setting ExecutionConfirmation allows you to block unwanted executions on a machine. It specifies that the Agent requires manual confirmation before it runs a test. If you set ExecutionConfirmation to true, the Agent doesn't execute automatically. Instead, the Agent console displays a message, where you have to confirm or reject the execution. Rejected executions report as failed. This setting is helpful if you use your local machine as your personal playground Agent. Ideally, you set a Characteristic Characteristics help you define which Elastic Execution Grid Agent should execute which tests. When you set up an Agent, you define its Characteristics: one for every application that's relevant to your tests. When you push tests to Elastic Execution Grid, you define which Characteristics you need to run these tests. Elastic Execution Grid forwards tests only to Agents with matching Characteristics. Only Agents that have all Characteristics you specify can execute the tests. for this Agent that clearly identifies it as off-limits for production. However, if another user accidentally picks this machine for execution, you can block the test run. |
DefaultWorkingDirectoryCleanup | When an Agent runs tests, it downloads all test artifacts to a local directory on the Agent machine. The Agent also stores test results in this directory before it uploads them to the server. Elastic Execution Grid runs a clean-up after each test run. By default, it deletes all files in the working directory that are older than 7 days.
To turn off the working directory cleanup and keep all files indefinitely, set Enabled to false. |
To take screenshots of user interface tests, modify the LiveView section. Screenshots allow you to review your tests in action and check for areas that need improvement. To do so, change the following settings:
For Enabled, enter the value True.
For ScreenshotIntervalInSeconds, specify the time interval in which the Agent should take screenshots.
After the execution, you can download the screenshots from the Elastic Execution Grid REST API.
Alternatively, you can configure your Agent to record executions.
If needed, change log level and log path in the Logging section.
In the IdleTimeout section, configure your Agent to stop the service automatically if it's been idle for too long. Agent machines typically host a wide range of services. If you need to be diligent about conserving resources on your machines, perform the following actions:
For Enabled, enter the value True.
For Minutes, specify a timeout in minutes.
If the Agent doesn't receive any executions within the time frame you specified, it stops the service and shows as Disconnected in the Agents tab. When you need the Agent again, restart it.
Save, but don't close the file yet. Next, you need to set your Agent Characteristics.
4. Set Characteristics. Characteristics determine which Elastic Execution Grid Agent executes which tests.Let’s say you have tests that you need to run in Salesforce, but only one of your Elastic Execution Grid Agents has Salesforce installed. In this case, you can create a "Salesforce" Characteristic for this Agent.
Later on, users mark the respective tests as "Salesforce". Elastic Execution Grid then forwards these tests only to the "Salesforce" Agent (see this example).
You set Agent Characteristics in the CustomCharacteristics section of the appsettings.yaml file.
CustomCharacteristics:
# - <Name>: "<Value>"By default, Elastic Execution Grid adds the following Characteristics to any Agent:
Platform: kernel information of the Agent machine.
Runner A Runner is part of the Elastic Execution Grid Agent machine. It's the local installation of the test framework that executes your tests.: which Runners are on the Agent machine.
AgentIdentifier: Agent identifier according to the appsettings.yaml.
AgentVersion: Agent version number.
To add additional Characteristics, follow the steps below:
Delete the # sign.
Set the Characteristic:
Replace <Name> with the name of the Characteristic. For instance, Browser.
Replace <Value> with the value you want. For instance, GoogleChrome.
You can enter alphanumeric characters, underscores, or dashes. The entire line must have a minimum of 3 and a maximum of 63 characters.
Add as many Characteristics as you need. To do so, copy and paste the line and modify it to your needs.
Save and close the file.
This example shows how Elastic Execution Grid handles execution with multiple Agents and Characteristics. Let’s say you have 3 Agents.
Agent A has SAP, Salesforce, and the browsers Google Chrome and Firefox installed:
CustomCharacteristics:
- "Apps": "SAP"
- "Apps": "Salesforce"
- "Browser": "Chrome"
- "Browser": "Firefox"Agent B has Salesforce installed:
CustomCharacteristics:
- "Apps": "Salesforce" Agent C has SAP and the browser Microsoft Edge installed:
CustomCharacteristics:
- "Apps": "SAP"
- "Browser": "Edge"You have 3 test sets. When they push tests to Elastic Execution Grid, users specify which Characteristics apply to their tests:
Test set 1 requires SAP and Salesforce.
Test set 2 requires Salesforce.
Test set 3 requires SAP and Microsoft Edge.
At run-time, Elastic Execution Grid processes the tests like this:
Test set 1 goes to Agent A, because it’s the only Agent that has both SAP and Salesforce installed.
Test set 2 goes to either Agent A or B, because both Agents have Salesforce installed.
Test set 3 goes to Agent C, because it’s the only Agent that has SAP and Microsoft Edge installed.
5. Connect custom frameworks to the Agent service.By default, Elastic Execution Grid supports the following test frameworks: Tricentis Tosca and PowerShell.
If you plan to use a different framework to run your tests, you need to connect it to the Agent service. To do so, follow the steps below:
Open the Runners folder in the Agent download package.
Create a new folder for your Runner.
Copy the RunnerConfig.yaml file from any default Runner folder and paste it into the new Runner folder.
Open the RunnerConfig.yaml file and modify it to your needs. The file contains comments with instructions on how to do so. For more information on how to work with YAML files, see the YAML documentation.
6. (Optional) Configure the Agent for recording.Execution recordings show you how the Agent navigates the user interface tests you designed. This allows you to review your tests in action and detect areas that may need some improvement.
To configure your Agent for recording, follow the steps below:
Go to the Runners folder in the Agent download package.
Go to the folder for the Runner you want to use on this machine. For instance, if this Agent should run Tricentis Tosca tests, go to the TBox folder.
In the RunnerConfig.yaml file, modify the following settings in the Recording section:
Setting | Description |
|---|---|
Enabled | Enter true to enable recording. If you plan to trigger your executions via REST API, you can enter the value ${RecordingVideo=true} and use this variable in an argument. Then Elastic Execution Grid only records the Flows A Flow determines the sequence of your tests. It consists of other Flows, which in turn consist of Units. or Units The smallest unit of testing in your Flow. The concept is the equivalent of a TestCase in Tricentis Tosca. where you added the argument. |
UploadOnlyIffailed | By default, Elastic Execution Grid keeps recordings of all executions: successful and failed. If you're not interested in successful executions, enter true. In this case, Elastic Execution Grid only keeps recordings of failed executions. If you plan to trigger your executions via REST API, you can enter the value ${UploadOnlyIffailed=true} and use this variable in an argument. Then Elastic Execution Grid discards recordings of successful executions, but only for Flows or Units where you added the argument. |
Save and close the file.
After the execution, you can access your recordings in the following ways:
View them in the detail view of your Flow in the Executions tab.
Download them from the Elastic Execution Grid REST API.
Access them directly in the Agent's output folder. Recordings in the output folder are available until the next clean-up.
Start Agents
Starting your Elastic Execution Grid Agent connects the Agent to Elastic Execution Grid. Once it's connected, the Agent is ready to receive and run tests.
Note: the Agent runs under the user who started it. Make sure that this user has all necessary rights and permissions for test execution.
To start your Agents, run the Agent.exe file which is part of the Agent download package.
You can see all started Agents in the 
Agents tab.
Modify Agent configurations
You can still change an Agent configuration in the appsettings.yaml after you have started the Agent. For instance if you want to add or change Characteristics.
After you have modified the appsettings.yaml, you need to restart the Agent.
Update Agents
Elastic Execution Grid regularly offers updated Agent versions with new and improved features. If you want to use these enhancements in your tests, update your Agents to the latest version.
Uninstall Agents
If you no longer need an Agent on a particular machine, you can uninstall it.
Set up the Tricentis Tosca integration
If you want to run Tricentis Tosca tests in Elastic Execution Grid, you have to activate the integration in Tricentis Tosca.
Elastic Execution Grid offers two types of integration with Tricentis Tosca: a full integration and a trial integration.
Full integration
If you choose the full integration, you can integrate in the following ways:
Via Automation Object Service
Via Elastic Execution Grid Add-in
The main difference is that with the Automation Object Service integration, Elastic Execution Grid also reports the test results back to Tricentis Tosca. If you integrate via add-in, results are in Elastic Execution Grid only.
Full integration is available for Tricentis Tosca 14.2 LTS, 14.3 STS, and 15.0 LTS.
For more information on setup options and processes, see the Tricentis Tosca help.
Trial integration
The trial integration is a quick and easy way to try out Elastic Execution Grid with Tricentis Tosca versions that are not available for full integration.
Trial integration is available for Tricentis Tosca 14.0.
You can download the Tosca Add-on for Elastic Execution Grid from the Downloads area in the Elastic Execution Grid user interface. The download package contains instructions on how to set up the integration and trigger tests in Tricentis Tosca.