Task Killed because of Recorded pid does not match the current pid #38128
Replies: 3 comments 5 replies
-
|
Thanks for opening your first issue here! Be sure to follow the issue template! If you are willing to raise PR to address this issue please do so, no need to wait for approval. |
Beta Was this translation helpful? Give feedback.
-
|
Looks like troubleshooting issue. I remember seeing it a long time ago (as long as 2.5 indeed) - so it's very likely it's been fixed already (there are likely 1000s of fixes since 2.5.3) - my recommendation is to upgrade to 2.8.3 (latest released ) and check there. This will save you a lot of time for investigation in case it's already fixed. |
Beta Was this translation helpful? Give feedback.
-
|
I'm also facing this issue (2.7.3) where tasks using the bash operator along with run_as_user 'airflow' fail. The airflow user does have sudo permissions because before adding default_impersonation and run_as_user the tasks ran fine. I've seen this issue across multiple discussions and it seems most solutions are to simply not use run_as_user. Hoping to find an alternative though as we want to enforce default impersonation. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Apache Airflow version
Other Airflow 2 version (please specify below)
If "Other Airflow 2 version" selected, which one?
2.5.3
What happened?
The task is being killed due to Recorded pid 248 does not match the current pid 249
What you think should happen instead?
The task should run
How to reproduce
Operating System
Linux
Versions of Apache Airflow Providers
No response
Deployment
Official Apache Airflow Helm Chart
Deployment details
We use the helm chart but workers are started on the side using simple kubernetes deployment files.
We use CeleryKubernetesExecutor.
Airflow is installed as a pip package in a virtual environment baked in the worker docker image.
The Worker Docker Image is then started in Kubernetes with the following command
Anything else?
The other DAGs work fine. The crashing DAG has the following characteristics:
The only task is a BashOperator:
The user is added to the Docker Image as a sudo user; the sudoers file has been edited to allow impersonification:
The error happens as soon as the python script executes, everything else before logs successfully:
Full traceback from worker:
Are you willing to submit PR?
Code of Conduct
Beta Was this translation helpful? Give feedback.
All reactions