This message was deleted.

You can create a new datasource from an existing one, and use this process to change the rollups etc. Have a look at https://druid.apache.org/docs/latest/multi-stage-query/examples#insert-for-reindexing-an-existing-datasource On some level this is re-ingesting but should be quicker etc.

Thanks a lot, Does it support streaming ingestion?

SQL based ingestion does not support streaming yet

and there is no way to run batch ingestions each 5 minutes for example?

Druid does not have an internal job scheduling mechanism as yet (although I have heard it being discussed for possible future implementation). So for now you would have to schedule your jobs externally, for example using Unix cron or Apache Airflow.

Thanks a lot 🙏