如何在GitHub Actions作业间共享服务容器？当前构建后容器销毁

GitHub Actions中，每个作业（Job）运行在独立的Runner实例上，作业关联的服务容器会随作业结束被销毁，无法直接跨作业共享。针对你的场景，提供以下几种可行方案：

方案1：合并依赖服务容器的作业 #

如果build、test步骤不需要严格拆分，将它们合并到同一个作业中，这样Postgres服务容器会全程可用：

on:
  push:
    branches: [ main ]
  pull_request:
    branches: [ main ]

env:
  RAILS_ENV: test
  RACK_ENV: test
  RAILS_MASTER_KEY: ${{ secrets.RAILS_MASTER_KEY }}
  POSTGRES_PASSWORD: postgres15
  POSTGRES_USERNAME: postgres
  POSTGRES_HOST: localhost

jobs:
  build-test:
    runs-on: ubuntu-latest
    services:
      postgres:
        image: postgres:15.1-alpine
        ports:
          - 5432:5432
        env:
          POSTGRES_PASSWORD: ${{ env.POSTGRES_PASSWORD }}
          POSTGRES_USER: ${{ env.POSTGRES_USERNAME }}
        options: >-
          --health-cmd pg_isready 
          --health-interval 10s 
          --health-timeout 5s 
          --health-retries 5
    steps:
      - name: Checkout
        uses: actions/checkout@v3
      - name: Initialize Ruby
        uses: ruby/setup-ruby@v1
        with:
          bundler-cache: true
      - name: Setup Rails
        run: bin/setup
      - name: Lint Ruby
        run: bundle exec rubocop
      - name: Run tests
        run: bin/rails test:all

  coverage:
    runs-on: ubuntu-latest
    needs: [ build-test ]
    steps:
      - uses: joshmfrankel/simplecov-check-action@main
        with:
          minimum_suite_coverage: 98
          minimum_file_coverage: 90
          github_token: ${{ secrets.GITHUB_TOKEN }}
          check_job_name: coverage

方案2：每个作业单独配置服务容器，缓存数据库状态 #

若必须保持作业拆分，可为每个需要Postgres的作业单独配置服务容器，同时通过缓存数据库数据避免重复初始化：

修改build作业，缓存数据库数据 #

build:
  runs-on: ubuntu-latest
  services:
    postgres:
      image: postgres:15.1-alpine
      ports:
        - 5432:5432
      env:
        POSTGRES_PASSWORD: ${{ env.POSTGRES_PASSWORD }}
        POSTGRES_USER: ${{ env.POSTGRES_USERNAME }}
      options: >-
        --health-cmd pg_isready 
        --health-interval 10s 
        --health-timeout 5s 
        --health-retries 5
      volumes:
        - postgres-data:/var/lib/postgresql/data
  steps:
    - name: Checkout
      uses: actions/checkout@v3
    - name: Initialize Ruby
      uses: ruby/setup-ruby@v1
      with:
        bundler-cache: true
    - name: Setup Rails
      run: bin/setup
    - name: Cache Postgres data
      uses: actions/cache@v3
      with:
        path: /var/lib/postgresql/data
        key: ${{ runner.os }}-postgres-${{ hashFiles('db/schema.rb', 'Gemfile.lock') }}
        restore-keys: ${{ runner.os }}-postgres-

修改test作业，复用缓存数据 #

test:
  runs-on: ubuntu-latest
  needs: [ build ]
  services:
    postgres:
      image: postgres:15.1-alpine
      ports:
        - 5432:5432
      env:
        POSTGRES_PASSWORD: ${{ env.POSTGRES_PASSWORD }}
        POSTGRES_USER: ${{ env.POSTGRES_USERNAME }}
      options: >-
        --health-cmd pg_isready 
        --health-interval 10s 
        --health-timeout 5s 
        --health-retries 5
      volumes:
        - postgres-data:/var/lib/postgresql/data
  steps:
    - name: Checkout
      uses: actions/checkout@v3
    - name: Initialize Ruby
      uses: ruby/setup-ruby@v1
      with:
        bundler-cache: true
    - name: Restore Postgres data
      uses: actions/cache@v3
      with:
        path: /var/lib/postgresql/data
        key: ${{ runner.os }}-postgres-${{ hashFiles('db/schema.rb', 'Gemfile.lock') }}
        restore-keys: ${{ runner.os }}-postgres-
    - name: Lint Ruby
      run: bundle exec rubocop
    - name: Run tests
      run: bin/rails test:all

缓存key基于db/schema.rb和Gemfile.lock的哈希值，确保数据库结构或依赖变更时自动更新缓存。

方案3：使用外部托管数据库（可选） #

如果允许额外成本，可采用AWS RDS、Heroku Postgres等外部数据库服务，让所有作业连接同一数据库，彻底摆脱Runner上的服务容器限制。

内容的提问来源于stack exchange，提问作者bitsmyth