I have a druid cluster with version 0.23 and deployed on kubernetes using druid operator. When I submit indexing task the task doesn't go to middle manager node instead it goes to coordinator node. Since I have overlord also running on same node when I submit task with bigger data size and worker concurrency the whole cluster getting crashed. Isn't middle manager supposed to pick up indexing task ? Am I missing any configuration to achieve the same ? I could not see any errors in logs.
This is a companion discussion topic for the original entry at https://stackoverflow.com/questions/73305856/druid-indexing-task-not-picking-up-by-middlemanager