Hello forum, I've bumped into an issue that puzzles me. I have a client that has a load balancing rule active for years, to 4 backend servers. It uses a TCP health check. This has been running fine until they did an upgrade of the backend last night…