在Elastic Beanstalk部署Django-Channels时WebSocket异常问题排查

背景 #

在AWS Elastic Beanstalk上部署Django应用，采用Gunicorn处理WSGI请求、Daphne处理ASGI请求。应用部署成功，但WebSocket功能无法正常工作。

Redis连接验证（结果正常） #

通过以下步骤验证EC2实例与Redis缓存的连接：

eb ssh

source /var/app/venv/*/bin/activate
cd /var/app/current/
python manage.py shell

在Python Shell中执行：

import channels.layers
from asgiref.sync import async_to_sync
channel_layer = channels.layers.get_channel_layer()
async_to_sync(channel_layer.send)('test_channel', {'foo': 'bar'})
async_to_sync(channel_layer.receive)('test_channel')
# 成功收到响应：{'foo': 'bar'}

手动启动Daphne失败 #

在EC2实例执行启动命令：

daphne -b 0.0.0.0 -p 5000 tattoovalley.asgi:application

报错信息：

django.core.exceptions.ImproperlyConfigured: Requested setting INSTALLED_APPS, but settings are not configured. You must either define the environment variable DJANGO_SETTINGS_MODULE or call settings.configure() before accessing settings.

已配置的.ebextensions文件 #

.ebextensions/01_env.config内容：

option_settings:  
    aws:elasticbeanstalk:environment:proxy:staticfiles:
        /static: static
        value: static/

    aws:elasticbeanstalk:container:python:     
        WSGIPath: tattoovalley/wsgi.py
    aws:elasticbeanstalk:application:environment:
        DJANGO_SETTINGS_MODULE: tattoovalley.settings
        PYTHONPATH: /opt/python/current/app/tattoovalley:$PYTHONPATH

    aws:elbv2:listener:80:
        DefaultProcess: http
        ListenerEnabled: 'true'
        Protocol: HTTP
        Rules: ws
    aws:elbv2:listenerrule:ws:
        PathPatterns: /ws/*
        Process: websocket
        Priority: 1
    aws:elasticbeanstalk:environment:process:http:
        Port: '80'
        Protocol: HTTP
    aws:elasticbeanstalk:environment:process:websocket:
        Port: '5000'
        Protocol: HTTP

端口检查结果 #

检查端口5000，未发现任何活动进程。

更新：ASGI配置文件内容 #

asgi.py文件：

import os
import django
from channels.auth import AuthMiddlewareStack
from channels.routing import ProtocolTypeRouter, URLRouter
from channels.security.websocket import AllowedHostsOriginValidator
from django.core.asgi import get_asgi_application

from chat.routing import websocket_urlpatterns

os.environ.setdefault("DJANGO_SETTINGS_MODULE", "tattoovalley.settings")
django.setup()
django_asgi_app = get_asgi_application()

import chat.routing

application = ProtocolTypeRouter(
    {
        "http": django_asgi_app,
        "websocket": AllowedHostsOriginValidator(
            AuthMiddlewareStack(URLRouter(websocket_urlpatterns))
        ),
    }
)

问题排查与解决方案 #

1. 手动启动时加载环境变量 #

通过eb ssh进入实例后，Elastic Beanstalk配置的环境变量不会自动加载到当前Shell会话，需手动加载：

# 导出环境变量到临时文件
sudo /opt/elasticbeanstalk/bin/get-config environment | jq -r 'to_entries|map("\(.key)=\(.value)")|.[]' > /tmp/env.sh
source /tmp/env.sh

# 激活虚拟环境并启动Daphne
source /var/app/venv/*/bin/activate
cd /var/app/current/
daphne -b 0.0.0.0 -p 5000 tattoovalley.asgi:application

2. 配置Daphne为后台服务 #

Elastic Beanstalk默认仅管理Gunicorn进程，需配置让其自动启动Daphne。可添加.ebextensions/daphne.config：

container_commands:
  01_kill_existing_daphne:
    command: "pkill -f daphne || true"
  02_start_daphne:
    command: "nohup /var/app/venv/*/bin/daphne -b 0.0.0.0 -p 5000 tattoovalley.asgi:application > /var/log/daphne.log 2>&1 &"
    leader_only: true

或者用systemd配置持久化服务，创建.ebextensions/daphne.service.config：

files:
  "/etc/systemd/system/daphne.service":
    mode: "000644"
    owner: root
    group: root
    content: |
      [Unit]
      Description=Daphne ASGI Server for Django Channels
      After=network.target

      [Service]
      User=webapp
      WorkingDirectory=/var/app/current
      EnvironmentFile=/opt/elasticbeanstalk/deployment/env
      ExecStart=/var/app/venv/*/bin/daphne -b 0.0.0.0 -p 5000 tattoovalley.asgi:application
      Restart=always

      [Install]
      WantedBy=multi-user.target

container_commands:
  01_enable_daphne:
    command: "systemctl enable daphne.service"
  02_start_daphne:
    command: "systemctl start daphne.service"

3. 验证负载均衡器路由配置 #

在AWS控制台检查：

• Elastic Beanstalk环境的负载均衡器HTTP 80端口监听器规则，确认/ws/*路径已指向websocket进程（端口5000）
• 对应目标组的健康检查状态，确保实例5000端口可正常访问

4. 检查安全组配置 #

• 确保EC2实例安全组允许来自ELB安全组的5000端口入站流量
• 确保ELB安全组允许客户端的80/443端口入站流量（WebSocket基于HTTP/HTTPS协议）

5. 确认Django Channels配置正确性 #

确保settings.py中：

• INSTALLED_APPS包含channels和channels_redis
• CHANNEL_LAYERS配置正确指向Redis：

CHANNEL_LAYERS = {
    'default': {
        'BACKEND': 'channels_redis.core.RedisChannelLayer',
        'CONFIG': {
            "hosts": [('redis-endpoint', 6379)],
        },
    },
}

内容的提问来源于stack exchange，提问作者paul