vincbeck commented on code in PR #39650:
URL: https://github.com/apache/airflow/pull/39650#discussion_r1602230619


##########
docs/apache-airflow/administration-and-deployment/logging-monitoring/metrics.rst:
##########
@@ -242,6 +242,8 @@ Name                                                
Description
 ``pool.scheduled_tasks``                            Number of scheduled tasks 
in the pool. Metric with pool_name tagging.
 ``pool.starving_tasks.<pool_name>``                 Number of starving tasks 
in the pool
 ``pool.starving_tasks``                             Number of starving tasks 
in the pool. Metric with pool_name tagging.
+``task.cpu_usage_percent.<dag_id>.<task_id>``       Percentage of CPU used by 
a task

Review Comment:
   > I don't think we need "_percent".
   
   100 % fine
   
   > also, given this notation, does it mean we emit multiple DAG runs 
aggregated into 1 metric? What happens when multiple DAG runs are active? I am 
not against skipping <dag_run_id> as it will save cost, but curious if you 
thought about this case?
   
   The task ID is enough, the task ID is unique already. The `dag_id` is here 
only so that it simpler to find out in which DAG the task is running. We could 
drop `dag_id` is you think this is not necessary. I also followed what is being 
done today, every metric on task level includes the `dag_id`.
   



-- 
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