Ubuntu环境下Gunicorn无法识别虚拟环境中的Python依赖

在Ubuntu 23.10上基于Nginx、Flask、Gunicorn部署的服务器，Python模块安装在venv虚拟环境中，此前部署正常，现在Gunicorn无法启动，报错找不到Flask模块。怀疑Gunicorn未使用指定的虚拟环境执行。

用户systemd配置文件 #

Description=description
After=network.target

[Service]
User=<my user name>
Group=www-data
WorkingDirectory=/var/www/<website name>.org/<project name>/<project name>/
Environment="PATH=/home/<user name>/.local/share/virtualenvs/<site name>-rbL82YaN/bin/"
ExecStart=/home/<user name>/.local/share/virtualenvs/<site name>-rbL82YaN/bin/gunicorn --workers 3 --bind unix:/var/www/<site name>/<project name>/<project name>/<project name>.sock wsgi:app

[Install]
WantedBy=multi-user.target

虚拟环境bin目录文件列表 #

activate      activate.fish  activate.ps1      flask     normalizer  pip3      pip3.11  python3     wheel   wheel-3.11
activate.csh  activate.nu    activate_this.py  gunicorn  pip         pip-3.11  python   python3.11  wheel3  wheel3.11

Gunicorn报错信息 #

[2023-11-22 04:17:49 +0000] [95711] [INFO] Starting gunicorn 20.1.0
[2023-11-22 04:17:49 +0000] [95711] [INFO] Listening at: http://127.0.0.1:8000 (95711)
[2023-11-22 04:17:49 +0000] [95711] [INFO] Using worker: sync
[2023-11-22 04:17:49 +0000] [95712] [INFO] Booting worker with pid: 95712
[2023-11-22 04:17:49 +0000] [95712] [ERROR] Exception in worker process
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/gunicorn/arbiter.py", line 589, in spawn_worker
    worker.init_process()
  File "/usr/lib/python3/dist-packages/gunicorn/workers/base.py", line 134, in init_process
    self.load_wsgi()
  File "/usr/lib/python3/dist-packages/gunicorn/workers/base.py", line 146, in load_wsgi
    self.wsgi = self.app.wsgi()
                ^^^^^^^^^^^^^^^
  File "/usr/lib/python3/dist-packages/gunicorn/app/base.py", line 67, in wsgi
    self.callable = self.load()
                    ^^^^^^^^^^^
  File "/usr/lib/python3/dist-packages/gunicorn/app/wsgiapp.py", line 58, in load
    return self.load_wsgiapp()
           ^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3/dist-packages/gunicorn/app/wsgiapp.py", line 48, in load_wsgiapp
    return util.import_app(self.app_uri)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3/dist-packages/gunicorn/util.py", line 384, in import_app
    mod = importlib.import_module(module)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/importlib/__init__.py", line 126, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
  File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
  File "<frozen importlib._bootstrap>", line 1147, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 690, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 940, in exec_module
  File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
  File "/var/www/zf47r85u8jkg.org/<project name>/<project name>/__init__.py", line 1, in <module>
    from flask import Flask
ModuleNotFoundError: No module named 'flask'
[2023-11-22 04:17:49 +0000] [95712] [INFO] Worker exiting (pid: 95712)
[2023-11-22 04:17:49 +0000] [95711] [INFO] Shutting down: Master
[2023-11-22 04:17:49 +0000] [95711] [INFO] Reason: Worker failed to boot.

问题分析与解决 #

从报错信息看，Gunicorn确实在使用系统全局安装的版本（路径/usr/lib/python3/dist-packages/gunicorn），而非虚拟环境内的版本，导致无法找到虚拟环境中安装的Flask模块。问题出在systemd服务的执行上下文，即使指定了PATH，可能因为环境变量传递或执行方式的问题，导致Gunicorn调用了全局版本。

修复步骤： #

1. 确认虚拟环境内的依赖安装
   手动激活虚拟环境并验证依赖：

   source /home/<user name>/.local/share/virtualenvs/<site name>-rbL82YaN/bin/activate
   pip install flask gunicorn
   pip list | grep -E "Flask|gunicorn"
   

   确保输出中存在这两个模块。

2. 修改systemd配置，强制使用虚拟环境Python执行Gunicorn
   将ExecStart改为通过虚拟环境的Python调用Gunicorn，避免路径歧义：

   ExecStart=/home/<user name>/.local/share/virtualenvs/<site name>-rbL82YaN/bin/python -m gunicorn --workers 3 --bind unix:/var/www/<site name>/<project name>/<project name>/<project name>.sock wsgi:app
   

3. 完善环境变量设置（可选）
   在[Service]段添加虚拟环境标识，确保上下文明确：

   Environment="VIRTUAL_ENV=/home/<user name>/.local/share/virtualenvs/<site name>-rbL82YaN/"
   Environment="PATH=/home/<user name>/.local/share/virtualenvs/<site name>-rbL82YaN/bin:$PATH"
   

4. 重新加载配置并重启服务

   sudo systemctl daemon-reload
   sudo systemctl restart <your-service-name>.service
   

5. 验证服务状态

   sudo systemctl status <your-service-name>.service
   

   同时检查socket文件是否生成：

   ls -l /var/www/<site name>/<project name>/<project name>/<project name>.sock
   

内容的提问来源于stack exchange，提问作者Alex Michael