Reputation: 307
We have an airflow setup with rabbitmq, mysql db and celery executor setup in a docker instance. Trigerring the first individual step/opertator via a run command as below works.
airflow run wfm_model_training_client_1_queue_11 fetch_top_funnel_data 2016-03-03
However when we trigger the entire dag via a backfill command as below, we encounter the following error during celery execution. Clearly this is an error with celery, however how do i extract more logs to debug the same.
backfill command
airflow backfill wfm_model_training_client_1_queue_11 -s 2016-03-03 -e 2016-03-03
Celery error in celery_taskmeta table in traceback column
Traceback (most recent call last):
File "/usr/lib/python2.7/site-packages/celery/app/trace.py", line 367, in trace_task
R = retval = fun(*args, **kwargs)
File "/usr/lib/python2.7/site-packages/celery/app/trace.py", line 622, in __protected_call__
return self.run(*args, **kwargs)
File "/usr/lib/python2.7/site-packages/airflow/executors/celery_executor.py", line 59, in execute_command
except subprocess.CalledProcessError as e:
AirflowException: Celery command failed
Upvotes: 1
Views: 3065
Reputation: 307
The error was due to some pickle related issue. I bypassed this issue by passing the --donot_pickle args to the backfill command.
It was resolved via the updated command
airflow run wfm_model_training_client_1_queue_11 fetch_top_funnel_data 2016-03-03 --donot_pickle
Upvotes: 2