Narrative

Upstream GHI (https://github.com/keycloak/keycloak/issues/11632) enabled Horizontal Pod Autoscaling for Keycloak deployed with the new Operator.
So the Keycloak CR natively supports HPA via subresources scaling, and which should enable customers to use an HPA of their choice.

However, we don't provide any guidelines to specific HPA configuration for now, nor do we know how this works at scale since it's not tested. A minimal piece of documentation is available at https://www.keycloak.org/getting-started/getting-started-scaling-and-tuning#_horizontal_autoscaling which doesn't include a working example. 

Providing some guidelines and a reliable example would require configuring the HPA to use some metrics on which it'd scale Keycloak. We don't have such knowledge at the moment, and therefore we need to investigate this and test it properly, so that we can extend the HA guide with guidelines and examples for OCP's HA.

Value Proposition

• Customers inquiries about how to configure HPA using the RHBK Operator for production-scale deployments
• Allows DevOps to use a horizontal pod autoscaler (HPA) to specify how OpenShift Container Platform should automatically increase or decrease the scale of a RHBK deployment configuration, based on metrics collected from the pods that belong to that deployment configuration.

Goals

• Test, verify, and document configuration examples from the usage reference of the OCP's HPA with the subresources scaling in the RHBK HA Guide.
• Ensure that RHBK is perfectly working fine, while scaling-in/out accordingly to the configuration from HPA

Implementation note

• References to existing issues: 
  https://github.com/keycloak/keycloak-benchmark/issues/441