แก้ไข

แชร์ผ่าน


Quickstart: Run end-to-end tests at scale with Microsoft Playwright Testing Preview

In this quickstart, you learn how to run your Playwright tests with highly parallel cloud browsers and troubleshoot failed tests easily using Microsoft Playwright Testing Preview. Use cloud infrastructure to validate your application across multiple browsers, devices, and operating systems. Publish the results and artifacts generated by Playwright to the service and view them in the service portal.

After you complete this quickstart, you have a Microsoft Playwright Testing workspace to run your Playwright tests at scale and view test results and artifacts in the service portal.

Important

Microsoft Playwright Testing is currently in preview. For legal terms that apply to Azure features that are in beta, in preview, or otherwise not yet released into general availability, see the Supplemental Terms of Use for Microsoft Azure Previews.

Prerequisites

Create a workspace

To get started with running your Playwright tests at scale on cloud browsers, you first create a Microsoft Playwright Testing workspace in the Playwright portal.

  1. Sign in to the Playwright portal with your Azure account.

  2. If you already have a workspace, select an existing workspace, and move to the next step.

    Tip

    If you have multiple workspaces, you can switch to another workspace by selecting the workspace name at the top of the page, and then select Manage all workspaces.

  3. If you don't have a workspace yet, select + New workspace, and then provide the following information:

    Field Description
    Workspace name Enter a unique name to identify your workspace.
    The name can only consist of alphanumerical characters, and have a length between 3 and 64 characters.
    Azure subscription Select the Azure subscription that you want to use for this Microsoft Playwright Testing workspace.
    Region Select a geographic location to host your workspace.
    This is the location where the test run data is stored for the workspace.

    Screenshot that shows the 'Create workspace' page in the Playwright portal.

  4. Select Create workspace to create the workspace in your subscription.

    During the workspace creation, a new resource group and a Microsoft Playwright Testing Azure resource are created in your Azure subscription.

When the workspace creation finishes, you're redirected to the setup guide.

Install Microsoft Playwright Testing package

To use the service, install Microsoft Playwright Testing package.

npm init @azure/microsoft-playwright-testing

This generates playwright.service.config.ts file which serves to:

  • Direct and authenticate Playwright to the Microsoft Playwright Testing service.
  • Adds a reporter to publish test results and artifacts.

If you already have this file, the package asks you to override it.

Configure the service region endpoint

In your setup, you have to provide the region-specific service endpoint. The endpoint depends on the Azure region you selected when creating the workspace.

To get the service endpoint URL, perform the following steps:

  1. In Add region endpoint in your setup, copy the region endpoint for your workspace.

    The endpoint URL matches the Azure region that you selected when creating the workspace.

    Screenshot that shows how to copy the workspace region endpoint in the Playwright Testing portal.

Set up your environment

To set up your environment, you have to configure the PLAYWRIGHT_SERVICE_URL environment variable with the value you obtained in the previous steps.

We recommend that you use the dotenv module to manage your environment. With dotenv, you define your environment variables in the .env file.

  1. Add the dotenv module to your project:

    npm i --save-dev dotenv
    
  2. Create a .env file alongside the playwright.config.ts file in your Playwright project:

    PLAYWRIGHT_SERVICE_URL={MY-REGION-ENDPOINT}
    

    Make sure to replace the {MY-REGION-ENDPOINT} text placeholder with the value you copied earlier.

Set up Authentication

To run your Playwright tests in your Microsoft Playwright Testing workspace, you need to authenticate the Playwright client where you're running the tests with the service. This could be your local dev machine or CI machine.

The service offers two authentication methods: Microsoft Entra ID and Access Tokens.

Microsoft Entra ID uses your Azure credentials, requiring a sign-in to your Azure account for secure access. Alternatively, you can generate an access token from your Playwright workspace and use it in your setup.

Set up authentication using Microsoft Entra ID

Microsoft Entra ID is the default and recommended authentication for the service. From your local dev machine, you can use Azure CLI to sign-in

az login

Note

If you're a part of multiple Microsoft Entra tenants, make sure you sign in to the tenant where your workspace belongs. You can get the tenant ID from Azure portal. See Find your Microsoft Entra Tenant. Once you get the ID, sign-in using the command az login --tenant <TenantID>

Set up authentication using access tokens

You can generate an access token from your Playwright Testing workspace and use it in your setup. However, we strongly recommend Microsoft Entra ID for authentication due to its enhanced security. Access tokens, while convenient, function like long-lived passwords and are more susceptible to being compromised.

  1. Authentication using access tokens is disabled by default. To use, Enable access-token based authentication

  2. Set up authentication using access tokens

Caution

We strongly recommend using Microsoft Entra ID for authentication to the service. If you are using access tokens, see How to Manage Access Tokens

Enable artifacts in Playwright configuration

In the playwright.config.ts file of your project, make sure you are collecting all the required artifacts.

  use: {
    trace: 'on-first-retry',
    video:'retain-on-failure',
    screenshot:'on'
  }

Run your tests at scale with Microsoft Playwright Testing

You've now prepared the configuration for running your Playwright tests in the cloud with Microsoft Playwright Testing. You can either use the Playwright CLI to run your tests, or use the Playwright Test Visual Studio Code extension.

Run a single test at scale

With Microsoft Playwright Testing, you get charged based on the number of total test minutes and number of test results published. If you're a first-time user or getting started with a free trial, you might start with running a single test at scale instead of your full test suite to avoid exhausting your free trial limits.

Note

Reporting feature is enabled by default for existing workspaces. This is being rolled out in stages and will take a few days. To avoid failures, confirm that Rich diagnostics using reporting setting is ON for your workspace before proceeding. See, Enable reporting for workspace.

After you validate that the test runs successfully, you can gradually increase the test load by running more tests with the service.

Perform the following steps to run a single Playwright test with Microsoft Playwright Testing:

To use the Playwright CLI to run your tests with Microsoft Playwright Testing, pass the service configuration file as a command-line parameter.

  1. Open a terminal window.

  2. Enter the following command to run your Playwright test on remote browsers in your workspace:

    Replace the {name-of-file.spec.ts} text placeholder with the name of your test specification file.

    npx playwright test {name-of-file.spec.ts} --config=playwright.service.config.ts
    

    After the test completes, you can view the test status in the terminal.

    Running 1 test using 1 worker
        1 passed (2.2s)
    
    To open last HTML report run:
    
    npx playwright show-report
    

You can now run multiple tests with the service, or run your entire test suite on remote browsers.

Caution

Depending on the size of your test suite, you might incur additional charges for the test minutes and test results beyond your allotted free test minutes and free test results.

Run a full test suite at scale

Now that you've validated that you can run a single test with Microsoft Playwright Testing, you can run a full Playwright test suite at scale.

Perform the following steps to run a full Playwright test suite with Microsoft Playwright Testing:

When you run multiple Playwright tests or a full test suite with Microsoft Playwright Testing, you can optionally specify the number of parallel workers as a command-line parameter.

  1. Open a terminal window.

  2. Enter the following command to run your Playwright test suite on remote browsers in your workspace:

    npx playwright test --config=playwright.service.config.ts --workers=20
    

    Depending on the size of your test suite, this command runs your tests on up to 20 parallel workers.

    After the test completes, you can view the test status in the terminal.

    Running 6 tests using 6 workers
        6 passed (18.2s)
    
    Test report: https://playwright.microsoft.com/workspaces/<workspace-id>/runs/<run-id>
    

View test runs and results in the Playwright portal

You can now troubleshoot the failed test cases in the Playwright portal.

  1. After your test run completes, a link to the Playwright Portal is generated. Open this link to view detailed test results and associated artifacts. The portal displays essential information, including:

    • CI build details
    • Overall test run status
    • The commit ID linked to the test run

    Screenshot that shows list of tests in the test run.

  2. The Playwright portal provides all the necessary information for troubleshooting. You can:

    • Switch between retries.
    • View detailed error logs, test steps, and attached artifacts such as screenshots or videos.
    • Navigate directly to the Trace Viewer for deeper analysis.

    Screenshot that shows the preview of a test.

  3. The Trace Viewer allows you to step through your test execution visually. You can:

    • Use the timeline to hover over individual steps, revealing the page state before and after each action.
    • Inspect detailed logs, DOM snapshots, network activity, errors, and console output for each step.

    Screenshot that shows the trace viewer.

Tip

You can use Microsoft Playwright Testing service features independently. You can publish test results to the portal without using the cloud-hosted browsers feature and you can also use only cloud-hosted browsers to expedite your test suite without publishing test results.

Note

The test results and artifacts that you publish are retained on the service for 90 days. After that, they are automatically deleted.

Optimize parallel worker configuration

Once your tests are running smoothly with the service, experiment with varying the number of parallel workers to determine the optimal configuration that minimizes test completion time.

With Microsoft Playwright Testing, you can run with up to 50 parallel workers. Several factors influence the best configuration for your project, such as the CPU, memory, and network resources of your client machine, the target application's load-handling capacity, and the type of actions carried out in your tests.

You can specify the number of parallel workers on the Playwright CLI command-line, or configure the workers property in the Playwright service configuration file.

Learn more about how to determine the optimal configuration for optimizing test suite completion.

Next step

You've successfully created a Microsoft Playwright Testing workspace in the Playwright portal and run your Playwright tests on cloud browsers.

Advance to the next quickstart to set up continuous end-to-end testing by running your Playwright tests in your CI/CD workflow.