amrit2196 commented on issue #42383:
URL: https://github.com/apache/airflow/issues/42383#issuecomment-2402880774
@jscheffl The issue we are facing is , assume a scenario where the task pod
runs and the base container gets completed and the side cars are terminated
immediately along with the base container, the task event are properly sent to
the kubernetes executor, whereas when I add a delay logic in termination of
side cars, the information is not sent to the executor even though the
containers are terminated and the pod is deleted after the delay. This results
in parallelism count reaching to 0. I can see messages like skipping event for
this task pod as the event has been already sent as well.
This happens even with a simple hello world dag:
from airflow import DAG
from airflow.operators.python_operator import PythonOperator
from datetime import datetime
# Define a simple Python function that will print "Hello, World!"
def hello_world():
print("Hello, World!")
# Define the default arguments for the DAG
default_args = {
'owner': 'airflow',
'start_date': datetime(2024, 10, 8), # Update to the current date or
earlier
'retries': 1,
}
# Instantiate the DAG
with DAG(
dag_id='hello_world_dag',
default_args=default_args,
schedule_interval=None, # Set to None to disable scheduled runs
catchup=False, # Don't backfill or run previous missed schedules
) as dag:
# Define a task that uses PythonOperator to run the hello_world function
hello_world_task = PythonOperator(
task_id='print_hello_world',
python_callable=hello_world,
)
# Define the task dependencies
hello_world_task
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
