This tool provides a tautological implementation of a Kubernetes LoadBalancer. Rather than allocating dynamic IP addresses and ports on cloud infrastructure, it simply sets the LoadBalancer ingresses to the node IP addresses on the nodePort's assigned by Kubernetes. This allows constrained environments like KinD, k3s, etc, to have load balancer support.
go build main.go
# Or
docker build .
Requires ginkgo, docker, kind, kubectl, and helm to be on the path, and docker to be running.
ginkgo run -vv .
helm upgrade --install ./deploy/helm/nodeport-load-balancer