When I run my JHipster application in my homelab HA k3s cluster with a single replica all works just fine. But when I scale it to 2-3 instances, the backend starts to replay with 302 HTTP statuses and I`m stuck at the home page.

There is nothing in the logs that can direct me to the problem.

Also, maybe it makes sense to mention. I`m running this behind nginx proxy manager.

Any idea what needs to be changed to make it work?