hussein-awala commented on PR #30669:
URL: https://github.com/apache/airflow/pull/30669#issuecomment-1510499899
I agree, but I will continue working on the PR at least to identify all the
problems, and I'm open for all suggestions that's why I opened this draft PR.
The new column is added to solve the issue with the poke number where it's
state is not stored between the different reschedules:
```
airflow=# select * from task_reschedule;
id | task_id | dag_id |
run_id | map_index | try_number | poke_number |
start_date | end_date | duration |
reschedule_date
----+-------------------+---------------------------+---------------------------------------------+-----------+------------+-------------+-------------------------------+-------------------------------+-----------------------------------------
81 | wait_for_external | example_sensor_reschedule |
scheduled__2023-04-16T14:22:26.249199+00:00 | -1 | 1 |
1 | 2023-04-16 14:24:32.003232+00 | 2023-04-16 14:24:32.061204+00 | 0 |
2023-04-16 14:24:42.056541+00
83 | wait_for_external | example_sensor_reschedule |
scheduled__2023-04-16T14:22:26.249199+00:00 | -1 | 1 |
2 | 2023-04-16 14:27:58.576074+00 | 2023-04-16 14:27:58.641273+00 | 0 |
2023-04-16 14:28:08.637106+00
85 | wait_for_external | example_sensor_reschedule |
scheduled__2023-04-16T14:22:26.249199+00:00 | -1 | 1 |
3 | 2023-04-16 14:28:09.70126+00 | 2023-04-16 14:28:09.847813+00 | 0 |
2023-04-16 14:28:20.112106+00
```
As I know, there is two methods to save the state between the different
executions, by adding a new column in the metadata db, or storing it in a xcom
(not a good solution in our case)
--
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]
