Not able to run DAG manually with Airflow version 2.3.3

[uscairflow]

Apache Airflow version

Other Airflow 2 version (please specify below)

What happened

When manually run a DAG from the dashboard (...8080/home), the DAG is stuck in the queue and never triggered, however, the DAG runs with the scheduled time. This happens with Airflow version 2.3.3

What you think should happen instead

We have the exactly the same DAG running Airflow version 2.2.2 and it worked with manual run.

How to reproduce

Click on the DAG, click on the upper right arrow (Trigger DAG) and the DAG is not running but the status shows its queued.

Operating System

Rocky Linux

Versions of Apache Airflow Providers

2.3.3

Deployment

Other

Deployment details

Azure Pipeline/Releases

Anything else

No response

Are you willing to submit PR?

• Yes I am willing to submit a PR!

Code of Conduct

• I agree to follow this project's Code of Conduct

[boring-cyborg[bot]]

Thanks for opening your first issue here! Be sure to follow the issue template!

[potiuk]

It's reproducible. Note that issues are not there to debug problem with your deployent

You need to look at your deployment and logs and check what happens. Likely you have some resource problem. And this is not a "generic" problem - people were using and triggering DAGs in 2.3.3 for months. Without any specific details that you provide, this is not possible to help - you need to dig deeper.

You need to look at your logs, talk to the admin of your deployment and try to see why it is happening.

converted into a discussion.

[potiuk]

For example if you run on K8S and task cannot be started - you will see in kubernetes logs why . But that is something that only you can know about and can act upon. There is absolutely nothing anyone else execptyou having access to your system and knowing how to monitor running apps can do.

[uscairflow]

I have compared the files and their sizes between the 2 Airflow servers (different versions) they are exactly the same (DAG files as well as Python files). What other things can go wrong with the deployment?

[potiuk]

Did you check reource usage?

[uscairflow]

I ran "top |grep airflow" and both servers show identical results

[potiuk]

Look at your deployment logs and check for anomalies.

Lookg @uscairflow . There is no way anyone can "guess" what's wrong in your deployment. If you will not provide any evidences I can only say "NO IDEA". You need to deal with it and dig deeper and find out what's going wrong.