CYarros10 opened a new issue, #32429:
URL: https://github.com/apache/airflow/issues/32429
### Apache Airflow version
Other Airflow 2 version (please specify below)
### What happened
Airflow 2.4.3
I created a DAG with:
```
default_args = {
"owner": "me",
"depends_on_past": False,
"email": [""],
"email_on_failure": False,
"email_on_retry": False,
"retries": 1,
"retry_delay": timedelta(minutes=2),
"start_date": datetime(2023, 7, 4),
"sla": timedelta(minutes=25),
}
with models.DAG(
f"the_new_reporting_dag_{VERSION}",
description=f"Export metadata to BigQuery location and create reporting
table.",
schedule_interval="0 0 * * *", # daily
tags=tags,
default_args=default_args,
user_defined_macros=user_defined_macros,
is_paused_upon_creation=True,
catchup=True,
max_active_runs=1,
dagrun_timeout=timedelta(minutes=30),
) as dag:
```
and ran it on 7/7/2023.
This started three dag runs with logical dates of 7/4/2023, 7/5/2023,
7/6/2023.
An SLA Miss entry SHOULD have these logical dates. Yet it does not. For
whatever reason, SLA Miss entries were for 2023-07-06 00:00:00 and 2023-07-07
00:00:00, with SLA timestamps of 2023-07-07, 16:10:54 and 2023-07-07, 16:12:04
respectively.
Why? SLA Miss table seems to have incorrectly inserted the wrong logical
dates AND completely missed the third DAG run.
### What you think should happen instead
SLA Miss entries should have occurred for the correct logical dates of:
2023-07-04 00:00:00
2023-07-05 00:00:00
2023-07-06 00:00:00
Because they all missed their 25 minute SLA (since they were all ran on
2023-07-07)
### How to reproduce
1. Create a DAG with an SLA of 20 minutes, start date of 3 days ago, and
catchup = true.
### Operating System
cloud composer on GCP
### Versions of Apache Airflow Providers
_No response_
### Deployment
Google Cloud Composer
### Deployment details
Airflow 2.4.3, Cloud Composer 2.1.8
### Anything else
_No response_
### Are you willing to submit PR?
- [ ] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of
Conduct](https://github.com/apache/airflow/blob/main/CODE_OF_CONDUCT.md)
--
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]
