Question

I have a system set up currently that is using celery with a redis
backend to do a bunch of asynchronous tasks such as sending emails,
pulling social data, crawling,etc. Everything is working great, but I
am having group figuring out how to monitor the system (aka the number
of queue up messages). I started looking through the celery source but
I figured I would post my questions in here:
First off, here are my configurations:

BROKER_BACKEND                  = "redis" 
BROKER_HOST                     = "localhost" 
BROKER_PORT                     = 6379 
BROKER_VHOST                    = "1" 
REDIS_CONNECT_RETRY     = True 
REDIS_HOST                              = "localhost" 
REDIS_PORT                              = 6379 
REDIS_DB                                = "0" 
CELERY_SEND_EVENTS                      = True 
CELERYD_LOG_LEVEL               = 'INFO' 
CELERY_RESULT_BACKEND           = "redis" 
CELERY_TASK_RESULT_EXPIRES      = 25 
CELERYD_CONCURRENCY             = 8 
CELERYD_MAX_TASKS_PER_CHILD = 10 
CELERY_ALWAYS_EAGER                     =True

---

The first thing I am trying to do is monitor how many messages are in
my queue. I assume, behind the scenes, the redis backend is just
pushing/popping from a list, although I cannot seem to find that in
the code. So I mock up a simulation where I start about 100 tasks and
am trying to find them in redis:
My celeryd is running like this:
python manage.py celeryd -c 4 –loglevel=DEBUG -n XXXXX –logfile=logs/
celery.log
So I should only have 4 concurrent workers at once …..
Two thing I do not understand:
Problem 1:
After I have queued up 100 task, and look for them on redis, I only
see the following:

$ redis-cli 
redis 127.0.0.1:6379> keys * 
1) "_kombu.binding.celery" 
redis 127.0.0.1:6379> select 1 
OK 
redis 127.0.0.1:6379[1]> keys * 
1) "_kombu.binding.celery" 
2) "_kombu.binding.celeryd.pidbox" 
redis 127.0.0.1:6379[1]>

I cannot seem to find the tasks to get a number of how many are queued
(technically, 96 should be since I only support 4 concurrent tasks)

Problem 2

$ ps aux | grep celeryd | cut -c 13-120 
 41258   0.2  0.2  2526232   9440 s004  S+    2:27PM   0:07.35 python 
manage.py celeryd -c 4 --loglevel=DEBU 
 41261   0.0  0.1  2458320   2468 s004  S+    2:27PM   0:00.09 python 
manage.py celeryd -c 4 --loglevel=DEBU 
 38457   0.0  0.8  2559848  34672 s004  T    12:34PM   0:18.59 python 
manage.py celeryd -c 4 --loglevel=INFO 
 38449   0.0  0.9  2517244  36752 s004  T    12:34PM   0:35.72 python 
manage.py celeryd -c 4 --loglevel=INFO 
 38443   0.0  0.2  2524136   6456 s004  T    12:34PM   0:10.15 python 
manage.py celeryd -c 4 --loglevel=INFO 
 84542   0.0  0.0  2460112      4 s000  T    27Jan12   0:00.74 python 
manage.py celeryd -c 4 --loglevel=INFO 
 84536   0.0  0.0  2506728      4 s000  T    27Jan12   0:00.51 python 
manage.py celeryd -c 4 --loglevel=INFO 
 41485   0.0  0.0  2435120    564 s000  S+    2:54PM   0:00.00 grep 
celeryd 
 41264   0.0  0.1  2458320   2480 s004  S+    2:27PM   0:00.09 python 
manage.py celeryd -c 4 --loglevel=DEBU 
 41263   0.0  0.1  2458320   2480 s004  S+    2:27PM   0:00.09 python 
manage.py celeryd -c 4 --loglevel=DEBU 
 41262   0.0  0.1  2458320   2480 s004  S+    2:27PM   0:00.09 python 
manage.py celeryd -c 4 --loglevel=DEBU 

If anyone could explain this for me, it would be great.

Answer 1

Your configuration has CELERY_ALWAYS_EAGER = True. This means that the tasks run locally and hence you won’t see them in Redis. From the docs: http://celery.readthedocs.org/en/latest/configuration.html#celery-always-eager

CELERY_ALWAYS_EAGER

If this is True, all tasks will be executed
locally by blocking until the task returns. apply_async() and
Task.delay() will return an EagerResult instance, which emulates the
API and behavior of AsyncResult, except the result is already
evaluated.

That is, tasks will be executed locally instead of being sent to the
queue.