istio: Internal Loadbalancer not being created on GKE

Describe the bug When creating an istio-ingressgateway in gcp and adding “internal” annotation load balancer is stuck in pending

Expected behavior Loadbalancer is created

Steps to reproduce the bug add serviceAnnotations: cloud.google.com/load-balancer-type: “internal” to the “istio-ingressgateway:” in the values file

Version {{ What version of Istio and Kubernetes are you using? Use istioctl version and kubectl version }} not using istioctl Server Version: version.Info{Major:“1”, Minor:“11+”, GitVersion:“v1.11.6-gke.3”, GitCommit:“04ad69a117f331df6272a343b5d8f9e2aee5ab0c”, GitTreeState:“clean”, BuildDate:“2019-01-10T00:39:15Z”, GoVersion:“go1.10.3b4”, Compiler:“gc”, Platform:“linux/amd64”}

Installation helm chart downloaded from https://istio.io/docs/setup/kubernetes/download-release/

Environment cloud provider: gcp nodes: n1-standard-4 (4 vCPUs, 15 GB memory)

I am aware of the issue with gcp internal load balancer limit (something that was not written in documentation when we moved to google cloud) “A maximum of 5 ports in a forwarding rule, unless you configure the forwarding rule to match all ports.”

there are 8 port configured on “istio-ingressgateway” Port 80 Port 443 Port 31400 Port 15011 - tcp-pilot-grpc-tls Port 8060 - tcp-citadel-grpc-tls Port 853 - tcp-dns-tls Port 15030 - http2-prometheus Port 15031 - http2-grafana

I would like to work with everything offered in the ingressgateway is there any way around this and still use internal load balancer? I tried splitting the LB and putting 80/443/31400/853 on one LB and 15011/8060/15030/15031 on the second LB But when i started with the bookstore app example everything did not work properly. When going back to regular LB (external) with all ports on one LB everything worked great. Due to limitations in our company we need to work with internal LB’s. is there any way around this?

Thanks