ElasticSearch operator pod placement tends to "stick" to first available node.
Issue
When ElasticSearch instance is created by Cluster Logging Operator, ES pods tend to stick to the very few nodes. This is not ideal, as we'd expect uniform distribution of ES pods over the cluster.
Environment
OCP 4
Logging stack
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
