Prerequisites:
1: Install RabbitMQ as it would be used as message broker for Celery. In windows, it would create a service, make sure the service is started.
2: Install Celery: pip install celery
Meat and Potatoes:
Senario 1: don‘t specify the backend for celery, if we don‘t care about the result
1. Create a module named tasks.py
from __future__ import absolute_import
from celery import Celery
import time
app = Celery(‘tasks‘, broker=‘amqp://[email protected]:5672//‘)
@app.task
def add(x, y):
print ‘hello celery‘
time.sleep(10)
return x + y
2. Start Celery worker
celery worker -A tasks --loglevel=INFO
You would see the console output like below,
-------------- [email protected] v3.1.10 (Cipater) ---- **** ----- --- * *** * -- Windows-7-6.1.7601-SP1 -- * - **** --- - ** ---------- [config] - ** ---------- .> app: tasks:0x36871d0 - ** ---------- .> transport: amqp://[email protected]:5672// - ** ---------- .> results: disabled - *** --- * --- .> concurrency: 8 (prefork) -- ******* ---- --- ***** ----- [queues] -------------- .> celery exchange=celery(direct) key=celery [tasks] . tasks.add [2014-03-26 15:43:11,263: INFO/MainProcess] Connected to amqp://[email protected]:5672// [2014-03-26 15:43:11,285: INFO/MainProcess] mingle: searching for neighbors [2014-03-26 15:43:12,293: INFO/MainProcess] mingle: all alone [2014-03-26 15:43:12,302: WARNING/MainProcess] [email protected] ready.
3. Test the method
Call the function "add",
>>> from tasks import add >>> result = add.delay(3,5) >>>
You would see something like below from Celery worker console,
[2014-03-26 15:55:04,117: INFO/MainProcess] Received task: tasks.add[0a52fd72-c7cd-4dc7-91a8-be51f1ff4df2] [2014-03-26 15:55:04,118: WARNING/Worker-1] hello celery [2014-03-26 15:55:14,130: INFO/MainProcess] Task tasks.add[0a52fd72-c7cd-4dc7-91a8-be51f1ff4df2] succeeded in 10.0110001564s: 8
If you want to see task status from client, you can use call "result.ready()". However, as we didn‘t specify the backend for Celery, by defualt it would use "DisabledBackend", you would encounter the following error,
>>> result.ready()
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "C:\Python27\lib\site-packages\celery\result.py", line 254, in ready
return self.state in self.backend.READY_STATES
File "C:\Python27\lib\site-packages\celery\result.py", line 390, in state
return self._get_task_meta()[‘status‘]
File "C:\Python27\lib\site-packages\celery\result.py", line 327, in _get_task_meta
meta = self.backend.get_task_meta(self.id)
File "C:\Python27\lib\site-packages\celery\backends\base.py", line 291, in get_task_meta
meta = self._get_task_meta_for(task_id)
AttributeError: ‘DisabledBackend‘ object has no attribute ‘_get_task_meta_for‘
To resolve this issue, here comes the following second senario.
Senario 2: Specify the backend for celery, if we do care about the result
1. Update the module tasks.py to specify parameter "backend" as "amqp". For other backend specification, refer to doc
from __future__ import absolute_import
from celery import Celery
import time
app = Celery(‘tasks‘, backend="amqp", broker=‘amqp://[email protected]:5672//‘)
@app.task
def add(x, y):
print ‘hello celery‘
time.sleep(10)
return x + y
2. Restart celery worker and open a new python shell. (This is important, otherwise the code update above won‘t take effect)
3. Test
>>> from tasks import add >>> result = add.delay(3,5) >>> result.ready() False >>> result.state ‘PENDING‘ >>> result.status ‘SUCCESS‘ >>> result.state ‘SUCCESS‘ >>> result.ready() True >>> result.get() 8 >>>
See also: https://denibertovic.com/posts/celery-best-practices/
A simple case to use Celery: