Remote logging using GCP connection from connection table broken

[potatochip]

Apache Airflow version

2.2.0 (latest released)

Operating System

linux

Versions of Apache Airflow Providers

No response

Deployment

Other 3rd-party Helm chart

Deployment details

No response

What happened

The following commit removed the ability to use the connections table for remote logging with GCP. Can we roll it back?

083c3c1#diff-6af0d1d2bb4aa467255c0b73646429b820cdfea4f8bd6eb2415d5f997e7ce6d7

What you expected to happen

I expect to be able to use the connections table for remote logging.

How to reproduce

Try and log with a gcp connection defined using AIRFLOW__LOGGING__REMOTE_LOG_CONN_ID

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]

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

[mik-laj]

Why do you want to configure the logginfusing connection? This does not sound safe as every user has the option to edit connections and therefore damage the instances. Can you tell more about your user case?

[potatochip]

that is the standard way of doing it. it is referenced in the airflow configuration https://airflow.apache.org/docs/apache-airflow/stable/configurations-ref.html#remote-log-conn-id. It was done that way for gcp before the regression. It is still done that way for aws s3 (https://github.com/apache/airflow/blob/main/airflow/providers/amazon/aws/log/s3_task_handler.py#L48), aws cloudwatch (https://github.com/apache/airflow/blob/main/airflow/providers/amazon/aws/log/cloudwatch_task_handler.py#L60), and azure (https://github.com/apache/airflow/blob/main/airflow/providers/microsoft/azure/log/wasb_task_handler.py#L61).

it is no less safe than the multitude of other airflow infra using connection ids.

[potatochip]

I would also add that all users do not have the ability to edit connections, only admin and ops roles.

[github-actions]

This issue has been automatically marked as stale because it has been open for 365 days without any activity. There has been several Airflow releases since last activity on this issue. Kindly asking to recheck the report against latest Airflow version and let us know if the issue is reproducible. The issue will be closed in next 30 days if no further activity occurs from the issue author.

[github-actions]

This issue has been closed because it has not received response from the issue author.