SameerMesiah97 opened a new pull request, #69768:
URL: https://github.com/apache/airflow/pull/69768

   **Description**
   
   This change introduces automatic default pool management for multi-team 
deployments.
   
   With this implementation, a corresponding default pool named 
`default_pool_<team_name>` is automatically created on team creation. During 
DAG parsing, tasks that do not explicitly specify a pool are assigned to their 
team's default pool based on the DAG bundle association, while tasks with 
explicitly configured pools are left unchanged.
   
   **Rationale**
   
   In multi-team deployments, tasks without an explicit pool are currently 
assigned to the global `default_pool`, allowing one team's workload to consume 
capacity shared across all teams. Avoiding this requires users to manually 
create and configure team-specific pools and update every task to reference 
them.
   
   By automatically provisioning a per-team default pool and applying it during 
DAG parsing, Airflow provides sensible defaults while preserving explicit pool 
assignments. This reduces operational overhead and helps isolate execution 
capacity between teams without requiring changes to existing DAGs.
   
   **Tests**
   
   Added unit tests verifying that:
   
   * Tasks using `default_pool` are assigned to the appropriate team default 
pool during DAG parsing.
   * Tasks with explicitly configured pools are not modified.
   * DAGs without an associated team continue to use `default_pool`.
   * Creating a team automatically provisions its corresponding default pool.
   * Team default pool names are generated correctly.
   
   **Documentation**
   
   Updated the multi-team documentation to describe automatic creation of team 
default pools and how they are applied during DAG parsing.
   
   **Backwards Compatibility**
   
   Existing behaviour is preserved when multi-team mode is disabled. Tasks with 
explicitly configured pools continue to use their configured pool without 
modification.
   
   Closes: #69170


-- 
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