getting-started-grpc/

README

Google Cloud Endpoints Sample for Go using gRPC

This sample demonstrates how to use Google Cloud Endpoints using Go and gRPC.

For a complete walkthrough, see the following guides:

• Quickstart with gRPC on Container Engine
• Quickstart with gRPC on Compute Engine

Test the code locally (optional)

Run the backend using go run:

$ go run server/main.go

Send a request from another terminal:

$ go run client/main.go
2017/03/30 17:08:32 Greeting: Hello world

Deploying service config

1. First, generate out.pb from the proto file:

   protoc --include_imports --include_source_info helloworld/helloworld.proto --descriptor_set_out out.pb
   

2. Edit api_config.yaml. Replace YOUR_PROJECT_ID:

   name: hellogrpc.endpoints.YOUR_PROJECT_ID.cloud.goog
   

3. Deploy your service:

   gcloud service-management deploy out.pb api_config.yaml
   

   Your config ID should be printed out, it looks like 2017-03-30r0. Take a note of it, you'll need it later.

   You can list the config IDs using this command:

   gcloud service-management configs list --service hellogrpc.endpoints.YOUR_PROJECT_ID.cloud.goog
   

Building the server's Docker container

Build and tag your gRPC server, storing it in your private container registry:

gcloud container builds submit --tag gcr.io/YOUR_PROJECT_ID/go-grpc-hello:1.0 .

Deploy to GCE or GKE

Deploy to GCE

1. Create an instance and SSH into it:

   gcloud compute instances create grpc-host --image-family gci-stable --image-project google-containers --tags=http-server
   gcloud compute ssh grpc-host
   

2. Set some environment variables (you'll need to manually set the service config ID):

   GOOGLE_CLOUD_PROJECT=$(curl -s "http://metadata.google.internal/computeMetadata/v1/project/project-id" -H "Metadata-Flavor: Google")
   SERVICE_NAME=hellogrpc.endpoints.${GOOGLE_CLOUD_PROJECT}.cloud.goog
   SERVICE_CONFIG_ID=<Your Config ID>
   

3. Pull your credentials to access your private container registry:

   /usr/share/google/dockercfg_update.sh
   

4. Run your gRPC server's container:

   docker run -d --name=grpc-hello gcr.io/${GOOGLE_CLOUD_PROJECT}/go-grpc-hello:1.0
   

5. Run Endpoints proxy:

   docker run --detach --name=esp \
       -p 80:9000 \
       --link=grpc-hello:grpc-hello \
       gcr.io/endpoints-release/endpoints-runtime:1 \
       -s ${SERVICE_NAME} \
       -v ${SERVICE_CONFIG_ID} \
       -P 9000 \
       -a grpc://grpc-hello:50051
   

6. Get the IP address of your secured gRPC server:

   gcloud compute instances list --filter=grpc-host
   

7. Send a request to the API server (see "Running the client" below)

Deploy to GKE

If you haven't got a cluster, first create one.

1. Edit container-engine.yaml. Replace <YOUR_PROJECT_ID> and <SERVICE_CONFIG_ID>.

2. Create the deployment and service:

   kubectl apply -f container-engine.yaml
   

3. Wait until the load balancer is active:

   kubectl get svc grpc-hello --watch
   

4. Send a request to the API server (see "Running the client" below)

Running the client

1. First, create a project API key.

2. Then, after you have your server's IP address (via GKE's kubectl get svc or your GCE instance's IP), run:

   go run client/main.go -api-key=AIza.... -addr=YOUR_SERVER_IP_ADDRESS:80 [message]