deepchand commented on issue #19192:
URL: https://github.com/apache/airflow/issues/19192#issuecomment-1217403771

   I have also facing the same problem, below are the details 
   airflow - 2.2.2 
   providers apache-airflow[celery,redis,microsoft.azure,slack] - 2.2.2
   constraints - 
https://raw.githubusercontent.com/apache/airflow/constraints-2.2.2/constraints-3.6.txt
 
   postgres version - 11
   executor - celery executor
   we have total  1 - webserver, 2 - scheduler and 6 - workers in cluster
   When the problem occurs - When we ran around more then 100 parallel dag run 
it starts to show this warning 
   **The scheduler does not appear to be running. Last heartbeat was received X 
minutes ago** and the number of minutes  get increasing till 20-22 minutes if 
we have scheduled more dag run in parallel and scheduler does not schedule new 
tasks while we have enough resources available on all components. 
   These are some config parameters which we are using 
   job_heartbeat_sec = 10
   scheduler_heartbeat_sec = 5
   scheduler_idle_sleep_time = 1
   min_file_process_interval = 300
   dag_dir_list_interval = 300
   scheduler_health_check_threshold = 300
   scheduler_zombie_task_threshold = 600
   max_tis_per_query = 512
   use_row_level_locking = True
   max_dagruns_to_create_per_loop = 50
   parsing_processes = 8
   
   While we search in schedulers log we have found scheduling loop time gets 
increase 
   <img width="1536" alt="Screenshot 2022-08-17 at 8 33 10 AM" 
src="https://user-images.githubusercontent.com/18530606/185025568-4f5c5ea7-6588-4c62-a1c2-a5c0dd0edb52.png";>
   <img width="1536" alt="Screenshot 2022-08-17 at 8 33 27 AM" 
src="https://user-images.githubusercontent.com/18530606/185025583-6e306ec9-bdad-41de-ae68-4dae6fe026c6.png";>
   
   
   
   


-- 
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]

Reply via email to