So I have been able to get the jar and required configuration files onto the Confluence Kubernetes cluster using the information in the "Using external libraries" reference located here: https://atlassian.github.io/data-center-helm-charts/examples/external_libraries/EXTERNAL_LIBS/
The changes I am performing require a restart of Confluence to take effect as they enable an alternate login facility (ie, CAS). These changes work fine when I make them on a stand alone EC2 instance without load balancers. However, when I make the *exact same changes* on the Kubernetes cluster, it only works about once every 15 or 20 times I re-deploy the Confluence helm chart.
To me, this issue seems a lot like a load balancer caching issue, but I am unable to determine for sure if this is the case. This is running on a single node Confluence cluster on EKS. I am using the aws-load-balancer-controller facility but I don't see any annotations in there that might impact the behavior (see: https://kubernetes-sigs.github.io/aws-load-balancer-controller/v2.6/guide/ingress/annotations/). My load balancer annotations are:
alb.ingress.kubernetes.io/certificate-arn: arn:aws:acm:us-east-1:<info redacted>
alb.ingress.kubernetes.io/listen-ports: '[{"HTTPS":443}, {"HTTP":31373}]'
alb.ingress.kubernetes.io/scheme: internet-facing
alb.ingress.kubernetes.io/target-group-attributes: stickiness.enabled=true,stickiness.lb_cookie.duration_seconds=60
alb.ingress.kubernetes.io/target-type: ip
alb.ingress.kubernetes.io/healthcheck-port: '8090'
alb.ingress.kubernetes.io/healthcheck-path: /status
Does anyone have suggestions as to what might be going on?
