Skip to content

Latest commit

 

History

History
181 lines (131 loc) · 6.06 KB

esp-v2-on-cloudrun.md

File metadata and controls

181 lines (131 loc) · 6.06 KB
Raw

Run ESPv2 on Google Cloud Run

This tutorial describes how to run ESPv2 as a gateway for the Bookstore endpoint, on Google Cloud Run.

Before you begin

  1. Select or create a Cloud Platform Console project.

  2. Enable billing for your project.

  3. Note the project ID, because you'll need it later.

  4. Install CURL for testing purposes.

  5. Enable Cloud Endpoints API, Cloud Service Management API, and Cloud Service Control API for your project in the Google Cloud Endpoints page in the API Manager. Ignore any prompt to create credentials.

  6. Download the Google Cloud SDK.

Deploying Backend service on Cloud Run

For this tutorial, we will deploy a simple HTTP bookstore manager as the backend application. We supply the Docker image for the backend at gcr.io/endpoints-release/bookstore:1, which is built from this Dockerfile.

To deploy Bookstore service on Cloud Run, you can either do it on Pantheon UI, (by choosing Cloud Run, then create Service), or you can directly run the following command, with the name that you want to use for the service, as well as the project ID created above.

gcloud run deploy CLOUD_RUN_SERVICE_NAME \
    --image="gcr.io/endpoints-release/bookstore:1" \
    --allow-unauthenticated \
    --platform managed \
    --project=YOUR_PROJECT_ID

On successful completion, the command displays a message similar to the following:

Service [bookstore] revision [bookstore-00001] has been deployed and is serving
traffic at https://BACKEND_SERVICE_URL

You can verify its status by sending a request to the service by:

curl https://BACKEND_SERVICE_URL/shelves

Deploying ESPv2

Similarly, you need to deploy ESPv2 on Google Cloud Run using a docker image. We supply the Docker image for ESPv2 at gcr.io/endpoints-release/endpoints-runtime-serverless:2. Note the -serverless suffix in this image, which denotes this is specifically for use on Cloud Run.

gcloud run deploy ESPv2_SERVICE_NAME \
    --image="gcr.io/endpoints-release/endpoints-runtime-serverless:2" \
    --allow-unauthenticated \
    --platform managed \
    --project=YOUR_PROJECT_ID

Replace ESPv2_SERVICE_NAME and YOUR_PROJECT_ID accordingly.

On successful completion, similar message is displayed:

Service [apiproxy] revision [apiproxy-00001] has been deployed and is serving
traffic at https://PROXY_SERVICE_URL

Configuring Endpoints

You must have an OpenAPI document based on OpenAPI Specification v2.0 that describes the surface of your backend service and any authentication requirements. You also need to add a Google-specific field that contains the URL for each service so that ESPv2 has the information it needs to invoke a service.

We supply a template for the bookstore service. You must make the following changes to it:

  1. Change the host name to the PROXY_SERVICE_URL, without the protocol identifier.
  2. Add the x-google-backend object with the address to BACKEND_SERVICE_URL, with the protocol identifier.

For example:

...

"host": "PROXY_SERVICE_URL",
"x-google-backend": {
  "address": "https://BACKEND_SERVICE_URL"
},
...

Deploying the Endpoints configuration

To deploy the Endpoints configuration, use the gcloud endpoints services deploy command. This command pushes the configuration to Google Service Management, Google's foundational services platform used by Endpoints and other services for creating and managing APIs and services.

To deploy the Endpoints configuration:

gcloud endpoints services deploy bookstore_swagger_template.json

As it is creating and configuring the service, Service Management outputs information to the terminal. When it finishes configuring the service, Service Management outputs the service configuration ID and the service name, similar to the following:

Service Configuration [ENDPOINTS_SERVICE_CONFIG_ID] uploaded for service [ENDPOINTS_SERVICE_NAME]

Note that on Cloud Run, ENDPOINTS_SERVICE_NAME is usually the same as PROXY_SERVICE_URL (minus the protocol identifier).

Build the service config to the ESPv2 docker image

You need to build the service config into a new ESPv2 image and redeploy that new image to Cloud Run. We provide a bash script to automate this process. Ensure you have the gcloud SDK installed and download this script.

Run it with the following commands:

chmod +x gcloud_build_image &&
./gcloud_build_image -s ENDPOINTS_SERVICE_NAME -c ENDPOINTS_SERVICE_CONFIG_ID -p YOUR_PROJECT_ID

It will use gcloud command to download the service config, build the service config into a new docker image, and upload the new image to your project container registry located here:

gcr.io/YOUR_PROJECT_ID/apiproxy-serverless:ENDPOINTS_SERVICE_NAME-ENDPOINTS_SERVICE_CONFIG_ID

Redeploy the ESPv2 Cloud Run service with the new image

Replace ESPv2_SERVICE_NAME with the name of your Cloud Run service.

gcloud run deploy ESPv2_SERVICE_NAME \
  --image="gcr.io/YOUR_PROJECT_ID/apiproxy-serverless:ENDPOINTS_SERVICE_NAME-ENDPOINTS_SERVICE_CONFIG_ID" \
  --allow-unauthenticated \
  --platform managed \
  --project=YOUR_PROJECT_ID

Testing the API

Now you can sent request to the backend service through the ESPv2 proxy:

curl https://PROXY_SERVICE_URL/shelves