[
https://issues.apache.org/jira/browse/AIRFLOW-4596?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Himanshu Garg updated AIRFLOW-4596:
-----------------------------------
Description:
With respect to the metric we are capturing in airflow version v1-10 -
[https://github.com/apache/airflow/blob/v1-10-stable/docs/metrics.rst]. The
Timer merics `dagrun.schedule_delay.<dag_id>` `Seconds of delay between the
scheduled DagRun start date and the actual DagRun start date`
we are capturig the delay of only schduled Dags.
But we should also capture the delay for the adhoc/non-scheduled dags by
capturing the difference between `start_date` in `dag_run` table and
`start_date` in `task_instance` table keeping the `dag_id` and `excecution`
date as the cardinality for performing join.
was:
With respect to the metric we are capturing in airflow version v1-10 -
[https://github.com/apache/airflow/blob/v1-10-stable/docs/metrics.rst]. The
Timer merics `dagrun.schedule_delay.<dag_id>` `Seconds of delay between the
scheduled DagRun start date and the actual DagRun start date`
we are capturig the delay of only schduled Dags.
But we shpuld also capture the delay for the adhoc/non-scheduled dags by
capturing the difference between `start_date` in `dag_run` table and
`start_date` in `task_instance` table keeping the `dag_id` and `excecution`
date as the cardinality for performing join.
> Capture stats for dagrun schedule delay for adhoc/nonscheduled dag
> ------------------------------------------------------------------
>
> Key: AIRFLOW-4596
> URL: https://issues.apache.org/jira/browse/AIRFLOW-4596
> Project: Apache Airflow
> Issue Type: New Feature
> Components: models
> Affects Versions: 1.10.3
> Reporter: Himanshu Garg
> Assignee: Himanshu Garg
> Priority: Major
>
> With respect to the metric we are capturing in airflow version v1-10 -
> [https://github.com/apache/airflow/blob/v1-10-stable/docs/metrics.rst]. The
> Timer merics `dagrun.schedule_delay.<dag_id>` `Seconds of delay between the
> scheduled DagRun start date and the actual DagRun start date`
> we are capturig the delay of only schduled Dags.
> But we should also capture the delay for the adhoc/non-scheduled dags by
> capturing the difference between `start_date` in `dag_run` table and
> `start_date` in `task_instance` table keeping the `dag_id` and `excecution`
> date as the cardinality for performing join.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
