Azure Pipeline容器中Pip安装超时及Pytest未找到问题求助

Azure DevOps Pipeline配置 #

stages:
- stage: unit_tests
  displayName: Run Tests
  jobs:
  - job: create_image
    displayName: Create image
    pool:
      name: SomePool
    steps:
    - script: |
        docker build -t someImage:latest -f Dockerfile .
        docker push repo/someImage:latest
      displayName: Create Base Image
  - job:
    displayName: Unit tests
    dependsOn: create_image
    pool: 
      name: SomePool
    container:
      image: someImage:latest
      endpoint: someServiceConnection
    steps:
      - script: |
          python -m pip install --upgrade pip
          python -m pip install --user pytest pytest-cov
          <some other commands>
        displayName: Install Dependencies
      - script: |
          python -m pytest -c pytest.ini tests/
        displayName: Run pytest

初始报错（pip安装超时） #

WARNING: Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ConnectTimeoutError(<pip._vendor.urllib3.connection.HTTPSConnection object at 0x7f7eb95ff2b0>, 'Connection to pypi.org timed out. (connect timeout=15)')': /simple/pip/
WARNING: Retrying (Retry(total=4, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ConnectTimeoutError(<pip._vendor.urllib3.connection.HTTPSConnection object at 0x7f34a25d4670>, 'Connection to pypi.org timed out. (connect timeout=15)')': /simple/pytest/

尝试操作后的问题 #

将pip安装命令移至Dockerfile：

RUN pip install -r requirements.txt
 RUN python -m pip install --upgrade pip
 RUN python -m pip install --user pytest

但Pipeline运行时提示pytest未找到，执行pip list确认pytest未安装：

/usr/local/bin/python: No module named pytest
##[error]Bash exited with code '1'.

注：在同一池机器上手动执行操作可得到正常结果。

一、解决pip安装超时问题 #

• 配置代理：容器内可能未继承池机器的代理配置，需手动添加。可在Dockerfile中设置环境变量：

  ENV http_proxy=http://your-proxy-address:port
  ENV https_proxy=http://your-proxy-address:port
  

  或在pip命令中直接指定代理：

  python -m pip install --upgrade pip --proxy http://your-proxy-address:port
  

• 更换PyPI镜像源：使用国内镜像源提升下载速度，避免超时。比如在pip安装时添加-i参数：

  python -m pip install --upgrade pip -i https://pypi.tuna.tsinghua.edu.cn/simple
  python -m pip install pytest pytest-cov -i https://pypi.tuna.tsinghua.edu.cn/simple
  

  也可在Dockerfile中配置全局镜像源：

  RUN mkdir -p ~/.pip
  RUN echo "[global]\nindex-url = https://pypi.tuna.tsinghua.edu.cn/simple" > ~/.pip/pip.conf
  

• 延长超时时间：通过--default-timeout参数增加pip的连接超时时间：

  python -m pip install --upgrade pip --default-timeout=60
  

二、解决pytest未被识别的问题 #

• 移除--user参数：--user会将包安装到当前用户的私有目录，容器运行时可能使用不同用户，导致包不在系统PATH中。直接全局安装：

  RUN python -m pip install --upgrade pip
  RUN pip install -r requirements.txt
  RUN pip install pytest pytest-cov
  

• 验证Dockerfile执行顺序：先升级pip，再安装依赖和pytest，避免旧版本pip导致安装失败。
• 强制重新构建镜像：Docker默认会使用缓存，可能导致修改后的Dockerfile未生效。在构建命令中添加--no-cache参数：

  docker build --no-cache -t someImage:latest -f Dockerfile .
  

• 检查容器内Python环境：在测试job中添加调试步骤，确认环境一致性：

  - script: |
      which python
      pip --version
      echo $PATH
      pip list
    displayName: Check Python Environment
  

  对比手动操作时的输出，确认是否使用了相同的Python解释器和环境变量。

内容的提问来源于stack exchange，提问作者Dense04