#celery

crohling88: Happy to report that setting BROKER_POOL_LIMIT = None solved our connection issue with the RabbitMQ cluster behind ELB ;) It was the connection pooling that fucked things over.

Ohhh … I guess he left … :/

is his name actually ask?

ooh i like these new scheduling options

"Execute when the upper edge of the sun appears over the eastern horizon in the morning."!

"Execute when there is enough light for objects to be distinguishable so that outdoor activities can commence; formally, when the Sun is 6 degrees below the horizon."

celery/celery#3182 (master - 9078905 : Dmitry Malinovsky): The build was fixed.

Change view : https://github.com/celery/celery/compare/dd711f...

Build details : https://travis-ci.org/celery/celery/builds/1066...

I need to use messaging service (will be using Rabbit MQ provided by CloudAMPQ). So I want to understand celery correctly. I have two servers running, one is powered by tornado and another is Django. The tornado server puts the messages in Queue and Django server fetches the messages. Is celery right library for this? because all the examples I see, both

producer and consumer code exists in same codebase

av_ you can use celery as communication between different systems. app.send_task sends tasks by name so you don't have to import the task function

as long as one server puts the order in the queue that the other one is listening to

adrianlc: thanks, do you have any examples?

and I didn't get this: as long as one server puts the order in the queue that the other one is listening to

the broker. i.e. RabbitMQ server must be accessible by both your servers. You could/might have several queues in the same RabbitMQ server, for example if you ever need to send tasks from your Tornado server to another independent system

adrianlc: oh cool, I will take care of it. Thanks for the tip :)

What's important is that you have the queues configured so that tornado puts the task in the same queue that the celery worker is listening to

yes, I understand that

it's easy really. Specially if you only need one queue

I am starting with one queue as of now :)

so I need to use this: http://docs.celeryproject.org/en/latest/referen... and I have to specify the same task name which is in Django code base. Is that right?

yep, that's it

great

Thanks! :D

you can see all the available the task names when starting the worker with --debug

cool~

thanks again!

adrianlc: is this correct: http://dpaste.com/1761BQB `add` is a task function on Django code and `message` is the parameter to it

adrianlc: The celery worker is not consuming the code :( Here's my code: http://dpaste.com/0T0K2SQ

av_: Are you using rabbitMQ behind a loadbalancer?

Djarnis: no

ok

I am using a single instance on cloudampq. shall I post the worker debug info?

let me try setting up CELERY_ACCEPT_CONTENT

ok

To me, it looks like cloudampq is a rabbitmq cluster?

yes