https://docs.openshift.com/container-platform/4.13/networking/metallb/about-metallb.html#nw-metallb-layer2_about-metallb-and-metallb-operator

"Since layer 2 mode relies on ARP and NDP, the client must be on the same subnet of the nodes announcing the service in order for MetalLB to work."

This sentence seems incorrect. If the MetalLB IP is in the same subnet as the node IPs, why would the client need to be on the same subnet as the nodes? If the node IP subnet is routable on the network, then the client can access the MetalLB IP just like it can access the node IPs