Kanak Biscuitwala created HELIX-384:
---------------------------------------
Summary: Ideal state replica count behavior is confusing
Key: HELIX-384
URL: https://issues.apache.org/jira/browse/HELIX-384
Project: Apache Helix
Issue Type: Bug
Reporter: Kanak Biscuitwala
Right now, the replica count in an ideal state is used in the following ways:
1. If the state model has a state with upper bound R
2. In HelixAdmin#rebalance to compute the preference lists/maps
But it's ignored in every other case, and this is particularly confusing when
the state upper bound N is used. The question is: should this replica count
trump every other upper bound constraint, even when the state model doesn't
have an R bound?
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
