Signal handler teardown raised: RuntimeError: Working outside of application context

[bouk]

Bug description

I'm running the alerts and reports celery worker as follows: celery --app=superset.tasks.celery_app:app worker --pool=prefork -O fair -c 4

On every job (e.g. on the minute from the beat) I get an exception in the logs, as below. I'm not sure it causes any issues. Perhaps the teardown handler should check if there's an app context?

Screenshots/recordings

Jan 03 17:31:00 celery [2026-01-03 17:31:00,019: ERROR/MainProcess] Signal handler <function teardown at 0x7f09077da520> raised: RuntimeError('Working outside of application context.\n\nThis typically means that you attempted to use functionality that needed\nthe current application. To solve this, set up an application context\nwith app.app_context(). See the documentation for more information.')
Jan 03 17:31:00 celery Traceback (most recent call last):
Jan 03 17:31:00 celery   File "/nix/store/8z13kf9snjgs5nwvrvbn8sh4nwngz9vs-superset/lib/python3.12/site-packages/celery/utils/dispatch/signal.py", line 280, in send
Jan 03 17:31:00 celery     response = receiver(signal=self, sender=sender, **named)
Jan 03 17:31:00 celery                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Jan 03 17:31:00 celery   File "/nix/store/8z13kf9snjgs5nwvrvbn8sh4nwngz9vs-superset/lib/python3.12/site-packages/superset/tasks/celery_app.py", line 71, in teardown
Jan 03 17:31:00 celery     db.session.remove()
Jan 03 17:31:00 celery   File "/nix/store/8z13kf9snjgs5nwvrvbn8sh4nwngz9vs-superset/lib/python3.12/site-packages/sqlalchemy/orm/scoping.py", line 180, in remove
Jan 03 17:31:00 celery     if self.registry.has():
Jan 03 17:31:00 celery        ^^^^^^^^^^^^^^^^^^^
Jan 03 17:31:00 celery   File "/nix/store/8z13kf9snjgs5nwvrvbn8sh4nwngz9vs-superset/lib/python3.12/site-packages/sqlalchemy/util/_collections.py", line 1015, in has
Jan 03 17:31:00 celery     return self.scopefunc() in self.registry
Jan 03 17:31:00 celery            ^^^^^^^^^^^^^^^^
Jan 03 17:31:00 celery   File "/nix/store/8z13kf9snjgs5nwvrvbn8sh4nwngz9vs-superset/lib/python3.12/site-packages/flask_sqlalchemy/session.py", line 102, in _app_ctx_id
Jan 03 17:31:00 celery     return id(app_ctx._get_current_object())  # type: ignore[attr-defined]
Jan 03 17:31:00 celery               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Jan 03 17:31:00 celery   File "/nix/store/8z13kf9snjgs5nwvrvbn8sh4nwngz9vs-superset/lib/python3.12/site-packages/werkzeug/local.py", line 519, in _get_current_object
Jan 03 17:31:00 celery     raise RuntimeError(unbound_message) from None
Jan 03 17:31:00 celery RuntimeError: Working outside of application context.
Jan 03 17:31:00 celery This typically means that you attempted to use functionality that needed
Jan 03 17:31:00 celery the current application. To solve this, set up an application context
Jan 03 17:31:00 celery with app.app_context(). See the documentation for more information.

Superset version

6.0.0

Python version

Not applicable

Node version

Not applicable

Browser

Chrome

Additional context

Python 3.12

Checklist

• I have searched Superset docs and Slack and didn't find a solution to my problem.
• I have searched the GitHub issue tracker and didn't find a similar bug report.
• I have checked Superset's logs for errors and if I found a relevant Python stacktrace, I included it here as text in the "additional context" section.

[dosubot]

Hi @bouk! I'm Dosu and I’m helping the superset team.

This error happens because the Celery worker's teardown handler calls db.session.remove() outside of a Flask application context. Flask-SQLAlchemy requires an active app context to manage sessions, and without it, you get the "Working outside of application context" RuntimeError. The teardown handler in superset/tasks/celery_app.py is registered on the task_postrun signal but does not wrap its logic in flask_app.app_context(), which is necessary for Flask context-dependent operations. Other handlers in the same file (like reset_db_connection_pool) do use the app context correctly, so this looks like an oversight or bug in the teardown handler's implementation.

To resolve the error, you can modify the teardown handler to wrap the session removal in a Flask app context, like this:

with flask_app.app_context():
    db.session.remove()

This change should prevent the RuntimeError from appearing in your logs. There is no indication that this error breaks alerts/reports functionality, but it does mean session cleanup may not always run as intended, which could eventually cause resource leaks or other subtle issues if left unaddressed.
You can see the relevant code and context here: superset/tasks/celery_app.py teardown handler.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

   

[ulekarsiddhant0-boop]

Hello @bouk , I am interested in this issue #36892. I properly understood this issue. So, please assign this issue to me for working on it and create pr.
Thank You!

[rusackas]

@ulekarsiddhant0-boop sorry to ask, but are you a bot? I'm seeing a lot of comments like this from you and others. Feel free to open a PR if you feel like it :D Contributions always appreciated.