Getting Started¶

In this guide you will learn how to deploy a simple Hello World image to Google Cloud Run using Helmless. Read what is Helmless first if you want to understand how it works. See the architecture for more information on how to adopt it to other cloud providers.

Deploying from a CI/CD pipeline is covered in a later guide.

Prerequisites¶

• A Google Cloud account
• The Google Cloud CLI
• The Helm CLI

Setting up the Configuration¶

Helmless uses the Helm pattern of a values.yaml file to configure the service. You can store this anywhere in your repository, but for this example we will use the config directory.

Create a config/values.yaml file, with the following content:

# yaml-language-server: $schema=https://raw.githubusercontent.com/helmless/cloudrun/main/values.schema.json
name: tutorial-service
image: 'us-docker.pkg.dev/cloudrun/container/hello'
env:
  COLOR: 'blue'

• yaml-language-server is used to provide syntax highlighting and validation for the values.yaml file
• name is the name of the service and must be unique per project
• image is the URL to the Docker image you want to deploy
• env is an optional set of environment variables to pass to the service

Deploying the Service¶

To deploy the service you need to template the Helm chart into a Cloud Run service and then use the Google Cloud CLI to deploy it.

Templating the Helm Chart¶

Use helm template command and our custom Helmless Cloud Run chart to template your values.yaml into a Cloud Run YAML¹ specification.

helm template oci://ghcr.io/helmless/cloudrun -f config/values.yaml --output-dir out/

• helm template is the command to template a Helm chart
• oci://ghcr.io/helmless/cloudrun is the URL to our Helmless Cloud Run chart in GitHub Container Registry
• -f config/values.yaml uses the values.yaml file in the config directory as the configuration for the service
• --output-dir out/ writes the templated Cloud Run YAML specification to the out directory

You should now have a out/cloudrun/templates/cloudrun.yaml file that looks like this:

---
# Source: cloudrun/templates/cloudrun.yaml
apiVersion: serving.knative.dev/v1
kind: Service
metadata:
    name: tutorial-service
    labels:
    annotations:
      run.googleapis.com/launch-stage: GA
      run.googleapis.com/ingress: all
      run.googleapis.com/region: europe-west1
  spec:
    template:
      metadata:
        annotations:
          run.googleapis.com/execution-environment: gen2
          run.googleapis.com/cpu-throttling: "true"
          autoscaling.knative.dev/maxScale: "100"
      spec:
        containerConcurrency: 80
        timeoutSeconds: 300
        containers:
          - image: us-docker.pkg.dev/cloudrun/container/hello
            env:
              - name: "COLOR"
                value: "blue"
            resources:
              limits:
                cpu: "1"
                memory: "512Mi"
            ports:
              - containerPort: 8080
            livenessProbe:
              failureThreshold: null
              httpGet:
                path: /
                port: 8080
              initialDelaySeconds: null
              periodSeconds: null
            readinessProbe:
              failureThreshold: null
              httpGet:
                path: /
                port: 8080
              initialDelaySeconds: null
              periodSeconds: null
            startupProbe:
              failureThreshold: null
              httpGet:
                path: /
                port: 8080
              initialDelaySeconds: null
              periodSeconds: null

Deploying the Service¶

Use the Google Cloud CLI to deploy the templated Cloud Run YAML specification to Google Cloud Run.

First login to Google Cloud and set the project and region² you want to deploy the service to.

gcloud auth login
gcloud config set project <your-project-id>
gcloud config set run/region <your-region> # e.g. europe-west1

Then deploy the service using the gcloud run services replace command.

gcloud run services replace out/cloudrun/templates/cloudrun.yaml

You should see an output like the following:

Applying new configuration to Cloud Run service [tutorial-service] in project [...] region [...]
✓ Deploying new service... Done.
  ✓ Creating Revision...
  ✓ Routing traffic...
Done.
New configuration has been applied to service [tutorial-service].
URL: https://tutorial-service-836227714099.europe-west1.run.app

But when you navigate to the URL you'll see a Error: Forbidden page. This is because the service is not publicly accessible by default.

To access your service locally without exposing it to the public internet, you can use Cloud Run proxy.

gcloud run services proxy tutorial-service

Cleaning up¶

You can delete the service using the gcloud run services delete command.

gcloud run services delete tutorial-service

Next Steps¶

From here you can head back to the overview to learn more about Helmless for Google Cloud Run or jump to the next guide to learn how to deploy from a CI/CD pipeline using Github Actions.