芹菜的定期任务已发送但未执行

Question

I am using celery in one of my django projects to perform some tasks. I recently required to add a periodic_task to update a field for the objects in one of my models. so I added a periocdic_task as follows.

import datetime
import logger.logger as logger

from celery.decorators import periodic_task



@periodic_task(run_every=datetime.timedelta(seconds=30))
def entries_about_to_expire():
   """ Changes the status of entries that are about to expire.
   """

   try:
       import entries.models as models
       startdate = datetime.datetime.today()
       enddate = startdate + datetime.timedelta(days=7)
       soon = models.EntryStatus.objects.get_or_create(
               name='Expiring Soon')[0]
       models.Entries.objects.filter(expiry_date__range=[startdate, enddate])\
               .update(status=soon)

   except Exception, err:
       logger.exception("Error while changing job status: %s" % (err))

Next thing I did was to run celery as follows:

>>python manage.py celeryd -l info --beat

Everything runs fine until now. I can even see the task is sent by celery as follows (this happens every 30 seconds as required).

[2013-11-15 15:01:10,690: INFO/MainProcess] Scheduler: Sending due task entries.tasks.entries_about_to_expire (entries.tasks.entries_about_to_expire)

But for some reason the task is not executed. I mean I can't find any changes in the status of Entries objects. I tested this using the shell.

In [2]: from entries.models import Entries, EntryStatus

In [3]: soon = EntryStatus.objects.get_or_create(name='Expiring Soon')[0]

In [4]: entries = Entries.objects.filter(status=soon)

In [5]: entries
Out[5]: []

I get the same results even after the task is sent multiple times. I am sure the code for entries_about_to_expire runs fine because if I call it manually it works. For example if I do the following in the shell

In [6]: from entries.tasks import entries_about_to_expire

In [7]: entries_about_to_expire()

In [8]: entries = Entries.objects.filter(status=soon)

In [9]: Entries
Out[9]: [<Entries: 223>, <Entries: 512>, <Entries: 526>, <Entries: 591>]

So why is the celery task is not executing ? any clues ?

Note: I've also tried the celerybeat_schedule method but the same results ie task is sent but not executed

Answer 1

The database scheduler will not reset when timezone related settings change, so you must do this manually.

PeriodicTask.objects.update(last_run_at=None)

Refer: http://docs.celeryproject.org/en/latest/userguide/periodic-tasks.html#time-zones