This message was deleted.

Hi JK, Druid does not store hard schemas ... the datasource schema is the union of the schemas of the segments of data ingested. So you only need to create a Supervisor ... as soon as it starts ingesting data the datasource will "appear" and be queryable. ref: https://docs.imply.io/latest/druid/api-reference/tasks-api#submit-a-task ref: https://docs.imply.io/latest/druid/development/extensions-core/kafka-supervisor-reference

https://druid.apache.org/docs/latest/development/extensions-core/kafka-supervisor-reference#ingesting-from-multiple-topics - This might be useful.

@Abhishek Agarwal Getting it, Anyway I need to have a supervisor right? How about submitting a task? Submitting a task also does creation of datasource... Or may be I'm wrong, What's right way of doing it like using supervisor or task?

Hi JK, Sorry, I posted the wrong doc page for Supervisor API ... here it is: https://druid.apache.org/docs/latest/api-reference/supervisor-api The "multi-topic" link that Abhishek posted is also a new feature in the product, something you could potentially use if you want to combine the data from both topics into one datasource. And to answer your question on regular tasks ... and ingestion and/or reindex task that creates segments with a new datasource name will implicitly create that new datasource .. again there is no hard schema for datasources in Druid ... it will appear when the first segment with that datasource name becomes announced by either a Peon (ingestion task, as a real-time segment) or Historical (as a historical segment, after being loaded from deep storage following a real-time or batch ingestion.) Let us know if you have further questions on this. Thanks. John

Sure ,Thank You @John Kowtko for the information...