Isn't the whole point of Ingress abstraction, that you can switch providers (like, from nginx to traefik, or aws-alb) and everything more-or-less works the same? I know there can be some custom annotations or whatnot, but the basic stuff should stay the same.

The fragmentation in the ecosystem is getting pretty difficult to manage. Most tools are still just passive dashboards that do not show you these lifecycle issues until your deployment actually fails.

We recently started to build Kunobi internally because we think the community needs to move beyond simple resource viewers and into tools that actually help manage the health of the cluster. It is frustrating that critical stuff like the Ingress NGINX retirement is so easy to miss with current tooling.