Traffic was fine. Servers were fine. The problem was the authentication service sitting behind the wrong piece of network hardware.
Authentication through an external load balancer can be the difference between a system that scales under pressure and one that implodes. Done right, you deliver secure, fast, consistent login flows at any volume. Done wrong, you create a choke point that turns minor problems into outages.
An external load balancer is not just a traffic cop. In authentication paths, it becomes part of the security perimeter, the routing logic, and the performance profile. It needs to handle SSL/TLS termination, session persistence, and sometimes even token validation. When authentication services sit behind it, every detail matters — from how sticky sessions are configured to the way health checks probe your endpoints.
The most common mistakes come from assuming application defaults will play nicely with the load balancer. Authentication often uses stateful protocols, and external load balancers are built for stateless scaling. That mismatch can break single sign-on flows, drop JWTs, or create timeouts that appear random but are entirely predictable.
A proper setup starts with clear identity flows. Map requests from client to load balancer, from load balancer to auth service, and back. Decide if you need SSL passthrough or SSL offloading. Configure idle timeouts to match token validity. Avoid IP-based session affinity if clients move between networks. Route based on host and path rules that match your environment exactly.