juhai opened a new pull request, #32278:
URL: https://github.com/apache/airflow/pull/32278

   <!--
   Thank you for contributing! Please make sure that your code changes
   are covered with tests. And in case of new features or big changes
   remember to adjust the documentation.
   
   Feel free to ping committers for the review!
   
   In case of an existing issue, reference it using one of the following:
   
   closes: #ISSUE
   related: #ISSUE
   
   How to write a good git commit message:
   http://chris.beams.io/posts/git-commit/
   -->
   This PR provides support for creating unique but sensible names to k8s pods 
based on the given name and `ti.job_id`. The KubernetesPodOperator supports 
adding random suffix but that isn't suitable for our use case. The random 
suffix can't be templated and isn't available via DAG context.
   
   The background for this is the need to run Spark tasks in k8s. In order to 
use the preferred `client` mode, the pod name needs to be placed in spark 
configuration so that driver-executor communication can take place via a 
headless k8s service. This requires that the name of the (driver) pod is unique 
and known at the time the pod is created. The `ti.job_id` seems to be a good 
way to make pod names unique and is available via pod context when pod is 
created.
   
   The change adds a new argument `job_id_as_suffix` to the 
`KubernetesPodOperator` class with backwards compatibility with 
`random_suffix`. In case `job_id_as_suffix=True` and 
`random_name_suffix=False`, the `ti.job_id` value from context will be appended 
to the pod name, separated by `-`. The same value can then be provided to spark 
configuration via Airflow templating to use as the driver hostname for the 
headless service.
   ---
   **^ Add meaningful description above**
   
   Read the **[Pull Request 
Guidelines](https://github.com/apache/airflow/blob/main/CONTRIBUTING.rst#pull-request-guidelines)**
 for more information.
   In case of fundamental code changes, an Airflow Improvement Proposal 
([AIP](https://cwiki.apache.org/confluence/display/AIRFLOW/Airflow+Improvement+Proposals))
 is needed.
   In case of a new dependency, check compliance with the [ASF 3rd Party 
License Policy](https://www.apache.org/legal/resolved.html#category-x).
   In case of backwards incompatible changes please leave a note in a 
newsfragment file, named `{pr_number}.significant.rst` or 
`{issue_number}.significant.rst`, in 
[newsfragments](https://github.com/apache/airflow/tree/main/newsfragments).
   


-- 
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]

Reply via email to