django-celery-beat VS kombu

I was looking for a nice way to manipulate periodic tasks in Celery. I found an amazing django-celery-beat package that provides PeriodicTask database objects. With PeriodicTask objects, you can dynamically add/remove/update periodic tasks in Celery. I want to share my approach. I've created an example GitHub repository and wrote step-by-step article.

Please take a look at https://github.com/celery/django-celery-beat with this package you can dynamically define periodic tasks. They are stored in your database, so they are persistent between deployments. I'm using this package with Celery in my project for server uptime monitoring, it is working well. I can dynamically add/delete/update any task.

Something similar is presented in django-celery-beat where you’re able to create schedule for running repeated task. So, there is choice field in admin, and choices are literally names of registered celery tasks.

I would recommend using celery in combination with celery beat scheduling and this package https://github.com/celery/django-celery-beat . What this means is if you have a task that you want to dynamically schedule for different times for different users all you have to do as add a new row to a database table and celery will handle the rest for executing.

If you want more control and/or visibility into your periodic tasks, I'd go with something like Celery Beat, that's easily integrated with Django: https://github.com/celery/django-celery-beat

For me it was django-celery-beat

I use a Django extension called django-celery-beat. It creates several models & one of those models is called crontab. It has a few fields & I would like to add one more field so that I can name these crontabs for easy reference. So I need to override the model & add the field.

It is feasible and there are a number of ways to do it. There are a few issues, though, mainly how to make sure that you are accessing the latest data and not getting part of the data mid write. A database handles these issues pretty well, since generally they are meant to be an intermediary between different processes. Another solution, that is often more flexible, is a message queue, like this one: https://github.com/celery/kombu

Above I use the library Kombu to create some connections and send some stuff. I started by setting up our environment variables. Then created exchange and queue objects. Finally, I made our connection object and the producer object, and then we sent a simple “Hello” message.