I've set up a service in kubernetes. However, I cannot reach it over TCP. What are the next debugging steps?
For the kubernetes yaml I've setup for TCP service, and I've pretty much copy/paste/modify for another service exposing JSON REST api server. What are some next steps I can take into the debugging?
EDIT 1:
https://pastebin.com/PweEv6gc This is my kubernetes-helm config. I'm trying to reach it within the cluster. When I try to reach it outside the cluster I'm using kubectl port-forward 50051 50051
utility.
EDIT 2:
I've tried to spin up ordinary flask web server and it works with specific pod/k8s/port configuration. This issues is related to grpc. or it's port binding.
I've used:
server.add_insecure_port("0.0.0.0:50051")
server.start()
And managed to connect via node port (10.132.0.25:31128, which I got from kubectl get service oracle
but within the cluster it says unreachable.
EDIT 3:
I've tried the same script inside the kubernetes and my localhost.
With using 10.132.0.25:31158 (nodeport): On my localhost it normally connects to the gRPC server. Inside k8s, it doesn't work
with using oracle:50051 Inside k8s it doesn't work
Update: Found the bug....it was hard. The HTTP proxy was set up, and on the tested pod I've used HTTP proxy which probably doesn't implement http/2 (tinyproxy). How can I print gRPC debug logs to get this info?
Update2: https://github.com/grpc/grpc/blob/master/doc/environment_variables.md
for gRPC better debug logs