I admit you're worrying me as I have yet to roll Celery out to my production site under uwsgi and have been working on solutions with a development server to date.
Still, I am left wondering that you would do yourself and us a great service if you clarified exactly what it is that you are experiencing. You say something is hanging. But what? And evidenced how?
What I can suggest is debugging the comms at two levels:
- flower will give you a good look at the Celery queues and what's in them
- RabbitMQ's web management interface will also give you a great look at its queues and what 's in them
They can tell you a lot about workers, clusters, nodes, and their status etc. If you have no workers running for example you won't get a response and that may be because you've not configured your system to run them porperly or they are crashing ... in which case checking logs may help or running a worker manually (logging info then comes straight to console).