rabbitmq - How to ACK celery tasks with parallel code in reactor? -

I have a celery that is called, just ignite the execution of some parallel code inside a twisted reactor. is. For example, there is no sample (runan) code:

  def run_task_in_reactor (): # to do this takes some time do_something () do_something_more () @ celery.task def run_task () : The print "reactor was started" reactor calfarththread (run_task_in_rector)   
 (For simplicity, please assume that the reactor is already running when the worker gets the work;   As soon as the other thread in my reactor  
 When I call  run_task.delay () , the task is finished very quickly (because it  run_task_in_reactor ()  does not wait to finish, only determines its execution time in reactor). And, when  run_task_in_reactor ()  runs at the end,  do_something ()  or  do_something_more ()  can throw an exception, which will be untold.  
 To use  pika  to consume my line, for example, to inform the worker about the correct completion of the job, I  do_something_more ()  However, inside the celery, this does not seem possible (or, at least, I do not know how to make the same effect complete)  
 Also, I can not remove the reactor because it One requirement is using some third-party code. I also appreciate other ways to achieve the same result.   
 
 
  Use reactor. BlockingCallFromThread  instead.