fcoelho/celery repository overview

⁠Celery

This is an image for running Celery⁠ as a Docker container.

⁠Usage

To make proper use of the image, you'll need to provide at the very least a /code volume with a configuration file for celery, by default celeryconfig.py. If the current directory holds a celeryconfig.py file, the Celery worker can be started as:

docker run --volume $PWD:/code fcoelho/celery

The image will install by default the newest version of the celery package. It's possible to install a specific version of any libraries needed using a requirements file for pip. If a file named requirements.txt on the /code volume is found when the container is started, it will be fed to pip install -r. Note: if this file is supplied, the celery package has to be installed from it, it won't be installed separately.

⁠Configuration

The container is configured through the following environment variables:

• REQUIREMENTS_FILE This sets the name of the requirements file to be used during installation of the dependencies. The default value is requirements.txt
• CELERY_MODULE Defines the module to be loaded by Celery. By default, Celery loads the configuration from a module called celeryconfig.py
• RUN_CELERY_BEAT If this environment variable is set to true, the container will run Celery beat along the workers
• CELERY_CONCURRENCY Sets the number of workers. By default, runs only 1 worker
• CELERY_LOG_LEVEL Sets the log level for the celery worker⁠, with notice as default
• CELERY_LOG_FILE Specifies the celery output log file to be used. By default, /log/celery.log is used

This container makes it possible to use a PyPI mirror during package installation if one is available, such as a container running devpi-server⁠. To use it, link to the container using pypi as its name:

docker run --link devpiserver_container:pypi fcoelho/celery

⁠Volumes

The image exposes the following volumes:

• /code This is where your code should reside. Paths for REQUIREMENTS_FILE and CELERY_MODULE are considered relative to /code
• /env Path for the virtualenv. Can be shared between two containers if needed - for example between the Celery client container and the worker container.
• /log Place to store logs