I have a similar question. I setup a k8s load balancer for gRPC service, my client could connect to the service (both service and client use gRPC-java) using load balancer external IP, and my service received the request and could complete process, however gRPC client just can't receive the response. if i don't use k8s load balancing, and deploy client code to same cluster, or simply setup a local port forwarding proxy on client side, then it works without any issue. anyone know why k8s load balancing is not working for gRPC?
On Friday, June 24, 2016 at 9:12:04 AM UTC-7, Didier Colens wrote: > > Let's say we have pod A scaled to 4 replicas and pod B scaled to 2. pod B > is a nodejs grpc client and pod A is a go grpc server, we can't use > kubernetes' builtin service and kube-proxy load balancing because the > client maintains a permanent connection to the server, so how can a client > distribute its queries to all 4 pods and not only 2 (in the best case). > > What is the proper way of doing this ? > -- You received this message because you are subscribed to the Google Groups "grpc.io" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/grpc-io/bb1419ed-c524-4a78-a23b-41f76667aca5%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
