victorsheng commented on issue #17835: URL: https://github.com/apache/dolphinscheduler/issues/17835#issuecomment-3707534797
Thanks for your feedback and for pointing out the complexities. You are absolutely right that DolphinScheduler's upgrades involve more than just schema changes (e.g., configuration migration, plugin compatibility, and internal protocol shifts), which traditionally require downtime. Regarding your concerns: - Beyond DB Changes: How do you suggest we handle non-database state during a rolling upgrade? Would a "Side-by-Side" deployment or a compatibility layer for the API/Master-Worker protocol be a more feasible direction? - Flink CDC Limitations: If Flink CDC is too heavy or limited, do you have recommendations for more lightweight data migration tools or a built-in version-mapping mechanism? The goal of this issue is to explore the possibility of reducing RTO (Recovery Time Objective) for enterprise users. I'd love to hear your thoughts on what a "minimum viable" zero-downtime upgrade might look like. -- 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]
