[GitHub] [airflow] potiuk edited a comment on pull request #11383: Spark Integration Configuration in Helm Chart

[GitBox <gi...@apache.org>]

potiuk edited a comment on pull request #11383:
URL: https://github.com/apache/airflow/pull/11383#issuecomment-706675482


   Yeah. With the JSON schema it looks better.
   
   And I would be all for merging it even without further modularization. I think what we've done with our project was a real-life example that corporates heavily relying on secure authentication mechanisms that can use the helm chart easily :). 
   
   We have now:
   * Possibility to rebuild all the dependent images by the customer
   * Kerberos support for authentication of workers including Kerberos sidecar
   * Presto DB supporting kerboeros
   * PostgreSQL mTLS authentication
   * Hadoop/Spark seamless integration (including Kerberos authentication)
   * Workfloo Identity capability via annotations
   * Capability of building the Airflow custom image using vetted binary dependencies on an air-gaped system
   * Custom Cluster Policy Violation per-Dag
   
   All this tested with a real Enterprise customer, based on their expectations and the limitations they had.
   
   I'd say all together this might be a super-powerful message to Enterprise customers once we release the Helm Chart. In many ways with all those changes, it became production-ready from the Enterprise customers point of view.
   
   We need MySQL support (But I believe someone is working on it), support for multiple schedulers for 2.0, a bit more comprehensive tests and I think we should be getting closer to releasing it as "production-ready" @dimberman.
    


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org