Ingress is often treated as a solved problem in Kubernetes until clusters grow, teams multiply, and production traffic becomes unforgiving. In this session, I’ll share real operational lessons from running Nginx ingress in Kubernetes and the challenges that pushed us toward evaluating Gateway API.

We’ll walk through the practical problems teams encounter in real clusters: ingress configuration sprawl, unclear ownership, security boundaries, and brittle deployments. I’ll show how Kubernetes-native constructs Ingress, Gateway API, and related policies behave under real pressure, what improved, and what new complexities emerged.

This talk is not about tools, but about decisions, trade-offs, and lessons learned operating Kubernetes ingress in production environments.