KeycloakX chart caching / autoscaling

[iamandymcinnes]

Hi,

I'm just getting prepared for going to production in the next few weeks with a keycloak deployment using the Keycloak-x helm chart. With the quarkus change I believe some of the caching complexities of firefly are removed? and I've been looking at the example in this repo with kube_ping, but the more I read the more confused I get about what is actually needed to be configured to configure both the caching and requirements to allow the auto clustering to work for the auto scaling feature.

I also notice there is a PR open (#716) to use the dns_ping by default which if I'm correct is now a default part of quarkus requiring no custom docker image to run a headless dns service?

Sorry for all the questions, but I'm fairly new to keycloak and the seems to be no clear documentation as to how to approach this now vs the old way with firefly based keycloak.

For context I'm running in AKS and attempting to get an auto scaling HA instance or "production" instance, which seems to be a flag in keycloak too but not sure if that is a thing in the codecentric charts? Also I see there are different "stacks" like "azure" for azure_ping etc, but not sure whether or not this is necessary.

Thanks in advance for any insight you can provide!

Andy

[github-actions]

This issue has been marked as stale because it has been open for 30 days with no activity. It will be automatically closed in 10 days if no further activity occurs.