Docker环境下Django集成Celery时task.get()无法正常工作求助

Hey there! I’ve run into this exact same headache when setting up Celery with Django in Docker, so I totally get how frustrating this is. Let’s walk through the most common fixes for this scenario—where messages land in RabbitMQ but task.get() hangs or fails to return a result.

1. Your Celery Worker Isn’t Running (or Connected Properly) #

Even if RabbitMQ gets the task message, if your Celery worker isn’t up and listening, the task will never execute. Let’s verify:

• First, check if your Celery worker container is running:

  docker ps
  

• If it’s running, pull the logs to confirm it connected to RabbitMQ successfully:

  docker logs <your-celery-worker-container-name>
  

  Look for lines like Connected to amqp://user:**@rabbitmq:5672//—if you don’t see this, your worker can’t reach the RabbitMQ service.
• Double-check your Celery broker URL matches your Docker setup. In your celery.py or worker command, it should reference the RabbitMQ service name from your docker-compose.yml:

  app = Celery('myproject', broker='amqp://user:password@rabbitmq:5672//')
  

2. Mismatched Task Serialization Settings #

Sometimes the task is sent, but the worker can’t deserialize it because Django and Celery use different serialization formats.

• Add these settings to your Django settings.py to standardize on JSON:

  CELERY_ACCEPT_CONTENT = ['json']
  CELERY_TASK_SERIALIZER = 'json'
  CELERY_RESULT_SERIALIZER = 'json'
  

• Make sure your Celery worker is pulling these settings—if you start it with celery -A myproject worker -l info, it should inherit them automatically if your Celery app is linked to Django correctly.

3. Missing or Unreachable Result Backend #

When you call task.get(), Celery needs to fetch the task result from a backend. If you haven’t configured this, or the worker can’t reach it, get() will hang.

• Add a result backend (Redis is a reliable choice for Docker setups) to settings.py:

  CELERY_RESULT_BACKEND = 'redis://redis:6379/0'
  

  Again, redis should match the service name in your docker-compose.yml.
• Verify the Redis container is running and check worker logs for any connection errors to the backend.

4. Docker Network Isolation #

Docker containers run in isolated networks by default. Ensure all your services (Django, Celery, RabbitMQ, Redis) are on the same network.

• In your docker-compose.yml, group all services under a shared network:

  version: '3'
  services:
    web:
      build: .
      networks:
        - celery-network
    celery:
      build: .
      command: celery -A myproject worker -l info
      networks:
        - celery-network
    rabbitmq:
      image: rabbitmq:3-management
      networks:
        - celery-network
    redis:
      image: redis:alpine
      networks:
        - celery-network
  networks:
    celery-network:
      driver: bridge
  

5. Code Sync or Permission Issues (Local Dev) #

If you’re using volume mounts for your code, the Celery worker might not pick up the latest task changes.

• Restart the Celery worker container after editing tasks:

  docker-compose restart celery
  

• Ensure the user inside the Celery container has permission to read your task files—adjust your Dockerfile’s USER setting or volume permissions if needed.

If you’re getting a specific error when calling task.get(), drop that detail and we can narrow it down further!

内容的提问来源于stack exchange，提问作者Ryan Skene