You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@airflow.apache.org by "jack (JIRA)" <ji...@apache.org> on 2019/04/24 11:27:00 UTC

[jira] [Commented] (AIRFLOW-3296) Taks Failed with empty Log and without sending email

    [ https://issues.apache.org/jira/browse/AIRFLOW-3296?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16825038#comment-16825038 ] 

jack commented on AIRFLOW-3296:
-------------------------------

It's possible that this issue was resolved by [https://github.com/apache/airflow/pull/3137/commits/db29af4ffb3d120ad55cd089a44b99feb7b8bf38]

Can you please recheck against newer version of Airflow?

> Taks Failed with empty Log and without sending email
> ----------------------------------------------------
>
>                 Key: AIRFLOW-3296
>                 URL: https://issues.apache.org/jira/browse/AIRFLOW-3296
>             Project: Apache Airflow
>          Issue Type: Bug
>            Reporter: Maria
>            Priority: Critical
>
> I have a DAG with 60 tasks (PythonsOperators) and in some executions different tasks are marked as failed, but I don’t know the reason, when I go to "View Log" the log is empty and when i pass over the red square it says Operator:null what does that mean?
>  
> It seems like it hasn't executed the task, but I don't understand why.
>  
> The questions are:
>  * Why Airflow mark it as failed but there is no execution showed on log?
>  * Why hasn't it send email of error if the tasks it's marked as failed?
>  
> Here is the python code associated to the dag:
> {code:java}
> DEFAULT_ARGS = {
> 'owner': 'blablabla',
> 'depends_on_past': False,
> 'start_date': datetime(2018, 5, 8),
> 'catchup': False,
> 'email': ['mail@mail.com'],
> 'email_on_failure': True,
> 'email_on_retry': False,
> 'retries': 3,
> 'max_active_runs': 1,
> 'retry_delay': timedelta(minutes=5)
> }
> dag = DAG('dag_name',
> default_args=DEFAULT_ARGS,
> schedule_interval='20 0 * * *')
> mylist = get_codes_list()
> for item in mylist:
> healthcheckerName = 'healthchecker_' + item
> healthchecker = PythonOperator(
> dag=dag,
> task_id=healthcheckerName,
> python_callable=prime_ops.check_last_budget_calculation(item),
> queue=SPECIFIC_QUEUE,
> pool=DEFAULT_PPC_POOL
> ){code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)