azure - 自动化测试在 Azure 中失败，但该步骤的结果是通过，这是为什么？

Question

我有一套在代码提交时执行的自动化 API 测试。自动化测试运行，但其中许多测试按预期失败。但是，执行测试的步骤会成功返回。因此，整个作业都会成功，这是不正确的行为。

parameters:
- name: testScriptDisplayName
  type: string
- name: testScript
  type: string

trigger:
  branches:
    include:
      - main

steps:
- task: NodeTool@0
  inputs:
    versionSpec: '18.x'
  displayName: 'Install Node.js'

- script: |
    ls -a
  displayName: 'Explore output directory'
  workingDirectory: '$(Build.Repository.LocalPath)/API/API-Tests'

- script: |
    npm install
    npx playwright install --with-deps
  displayName: "npm install"
  workingDirectory: '$(Build.Repository.LocalPath)/API/API-Tests'

- script: |
    ${{ parameters.testScript }}
    ls -a
  displayName: ${{ parameters.testScriptDisplayName }}
  workingDirectory: '$(Build.Repository.LocalPath)/API/API-Tests'
  failOnStderr: true
  continueOnError: false
  env:
    CI: true

- task: PublishTestResults@2
  displayName: 'Publish test results'
  condition: succeededOrFailed()
  inputs:
    testRunner: JUnit
    testResultsFiles: '**/e2e-junit-results.xml'
    # don't need to use searchFolder
    #searchFolder: 'ts/playwright'

Answer 1

添加failTaskOnFailedTests PublishTestResults 任务的参数，它会将管道标记为失败 - 但仍为您提供测试报告。

- task: PublishTestResults@2
  displayName: 'Publish test results'
  condition: succeededOrFailed()
  inputs:
    testRunner: JUnit
    testResultsFiles: '**/e2e-junit-results.xml'
    failTaskOnFailedTests: true