Deploy an Application to Google Cloud Run

In addition to Kubernetes, Waypoint provides a plugin that works with Google Cloud Run. This is a way to run containers on Google's cloud infrastructure.

Prerequisites

You'll need to install the waypoint binary locally, clone the examples repository (detailed in the next section). In addition, you may need to signup for Google Cloud, create a project, install the gcloud command, prepare your Google credentials, and enable Google Cloud Run.

You must also be running Docker Desktop locally.

Clone the examples repository

The code for this tutorial is in the hashicorp/waypoint-examples repository. Clone the repository with git.

$ git clone https://github.com/hashicorp/waypoint-examples.git

Navigate to the subdirectory for the Google Cloud Run project. This project uses NodeJS but the following instructions will work with any language that can be built with a cloud native buildpack.

$ cd waypoint-examples/gcp/google-cloud-run/nodejs

Sign up

Visit Google Cloud and sign up for an account. You must create a project under which your service accounts and cloud resources will be organized.

In this tutorial, we will refer to waypoint-project-id as the project ID.

Install gcloud command

Follow Google's instructions for installing the Google Cloud SDK. This will install a command on your system named gcloud.

Enable Google Cloud Run

You must enable Google Cloud Run for your project. Visit Google Cloud Run and click "Enable" to add it to your project.

You must also enable Google Cloud Run for the user or service account you will use for this tutorial. Go to the IAM & Admin console. Find the account you want to use and click the pencil icon for "Edit Member". Add the "Cloud Run Admin" (or similar) role to the account you will use for this tutorial.

Authenticate to Google Cloud

Waypoint will use your locally configured Google Cloud credentials to deploy your application.

Login to Google Cloud from your machine.

$ gcloud auth application-default login

Find your Google Cloud project ID. Use the value under PROJECT_ID.

$ gcloud projects list

PROJECT_ID                     NAME                            PROJECT_NUMBER
production-project             Production Environment          11111111111
development-project            Development Environment         22222222222
waypoint-project-id            Waypoint Project                33333333333

Set your project ID (replace waypoint-project-id with your Google Cloud project's ID).

$ gcloud config set project waypoint-project-id

In order to communicate between your local Docker instance, the remote container registry, and Google Cloud Run, you must configure Docker for Google Cloud. Run the auth configure-docker command.

$ gcloud auth configure-docker

Install the Waypoint server

You're now ready to configure and run Waypoint.

The Waypoint server can be run locally or on a remote server. The local installation requires fewer steps to setup. A remote installation will enable the full functionality of Waypoint, including logs, exec, and other commands.

$ waypoint install --platform=docker -accept-tos

$ waypoint install --platform=kubernetes -accept-tos

service/waypoint created
statefulset.apps/waypoint-server created

Configure waypoint

Edit waypoint.hcl to specify your Google Cloud project ID instead of waypoint-project-id. Let's examine a few relevant snippets first.

In the build section, this configuration uses the standard docker plugin to find the Docker repository. In this case, the URL to gcr.io with your Google Cloud project ID and the name of this Waypoint application are all that's needed. There is no GCR-specific plugin for Waypoint since this URL is all that's needed.

build {
  use "pack" {}

  registry {
    use "docker" {
      image = "gcr.io/waypoint-project-id/example-nodejs"
      tag   = "latest"
    }
  }
}

In the deploy section, you can specify the project and location. There are many other options for memory, cpu_count, and machine-specific configuration. See the Google Cloud Run plugin documentation for other arguments.

deploy {
  use "google-cloud-run" {
    project  = "waypoint-project-id"
    location = "us-east1"
    // ...

    capacity {
      memory                     = 128
      cpu_count                  = 1
      // ...
    }
  }
}

Here is the full configuration found in waypoint.hcl.

project = "example-nodejs"

app "example-nodejs" {
  labels = {
    "service" = "example-nodejs",
    "env"     = "dev"
  }

  build {
    use "pack" {}

    registry {
      use "docker" {
        image = "gcr.io/waypoint-project-id/example-nodejs"
        tag   = "latest"
      }
    }
  }

  deploy {
    use "google-cloud-run" {
      project  = "waypoint-project-id"
      location = "us-east1"

      port = 5000

      static_environment = {
        "NAME" = "World"
      }

      capacity {
        memory                     = 128
        cpu_count                  = 1
        max_requests_per_container = 10
        request_timeout            = 300
      }

      auto_scaling {
        max = 2
      }
    }
  }

  release {
    use "google-cloud-run" {}
  }
}

Initialize waypoint

Now, let's run initialize and run Waypoint.

$ waypoint init

You'll see messages indicating that the configuration is valid and the project was correctly set up.

Run waypoint up

Deploy the application to Google Cloud Run with up.

$ waypoint up

This may take a few minutes to run since it needs to copy the Docker image to the Google Container Registry.

Visit either URL shown in the output to see the running application.

Destroy the deployment

When you are done, destroy the deployment.

$ waypoint destroy -auto-approve

If you see an error message, it may be that the container is completing another task. To solve this, wait a minute or two and run destroy again.

Next steps

Read the plugin API documentation for more details.

Learn more about Waypoint by following along with the other tutorials for AWS, Azure, and others, or read the documentation for other Waypoint plugins.