Implementation of task scheduling on Celery with Redis and RabbitMQ as broker for performing the following -
- Creating tasks
- Pausing tasks
- Resuming tasks
- Terminating tasks
Tasks are considered here to be iterative in nature and have lots of iterations. The task done in each iteration here is just to wait for given number of seconds (just to replicate long iterative tasks), and APIs are created to enable all the four operations.
You can also go to flower to view all tasks at port 5555
The API can be tested via the swagger UI at :8000/docs where 8000 is the port
The project uses docker compose to create the containers and expose these ports. This can eaisly be deployed on Kubernetes using kompose or kubectl (after convertion of the resources though)
A pre-made docker image can be found here and the kubernetes resource files to apply that using kubectl can be found here
The USER ID and PASSWORD for flower are user and test respectively