Hi team. We're trying to test Druid hot and cold tiering setup with 2 historical hot nodes and 1 cold node. We're using the same broker for all the tiers. Currently we're seeing a strange behaviour that only 1 historical hot node is responding with queries. This seems to be incorrect as both the historical hot tiers have data to provide. Can anyone help out with this please.

Do your retention policies have two replicas per segment?

This issue is resolved : Both pods were getting. situated on the same node.By correcting the requests we were able to get to desired behaviour