Supervisors in druid

Hi Folks ,

I was recently trying to explore the druid Kafka indexing tasks , and cam through this new word “supervisors” in documentation pages .

Can any one help me understand clearly on what are druid supervisors ?

Any thoughts in your own simple words ?



You can think of supervisor as a process whose job is to submit ingestion tasks to the middleManager.To start real time ingestion for a particular datasource, you create supervisor for that datasource.
The Supervisor has information about the kafka servers,topic name and taskcount etc.
I think a better way to understand it is to create a supervisor and see it in action.

awesome thanks swarth