TL;DR: Learn how to take your own Node.js app (like this one) and deploy it to Google Cloud Run using Docker, Cloud Build, and Artifact Registry — fully managed, zero maintenance, and highly scalable.
Why Cloud Run?
Deploying applications typically requires setting up virtual machines or managing Kubernetes clusters. While this works for complex infrastructures, it introduces overhead for simple web apps.
Cloud Run solves this by allowing you to:
- Deploy containerized apps in a fully managed serverless environment
- Scale from 0 to N based on traffic automatically
- Pay only when your app is running
- Update apps with zero downtime
It's ideal for small services, APIs, microservices, and experiments.
What You’ll Do
In this hands-on guide, you will:
- Clone a Node.js app from GitHub
- Dockerize it and build the container with Google Cloud Build
- Push the container to Artifact Registry
- Deploy the image to Cloud Run
- Modify and redeploy your app with zero downtime
Step 1: Activate Cloud Shell and Set Region/Zone
Activate Cloud Shell
Cloud Shell is a VM provided by Google Cloud with the gcloud CLI, Docker, and other dev tools pre-installed. It comes with a 5GB persistent home directory and is ready to go.
- Go to Google Cloud Console
- Click the Activate Cloud Shell icon (top-right corner)
- Continue through the setup prompts
- Click Authorize when asked
Once ready, you'll see a message like:
Your Cloud Platform project in this session is set to "PROJECT_ID"
To confirm settings:
gcloud auth list
gcloud config list project
Set Region and Zone
Some resources (e.g., VMs, Artifact Registry) are regional. Set your default region and zone:
gcloud config set compute/zone "us-central1-a"
export ZONE=$(gcloud config get-value compute/zone)
gcloud config set compute/region "us-central1"
export REGION=$(gcloud config get-value compute/region)
Step 2: Clone the GitHub Repository
We’ll use a minimal Node.js app hosted at https://github.com/niiiixd/sampla-app.
Clone the repo:
git clone https://github.com/niiiixd/sampla-app.git
cd sampla-app
This app uses Express to serve a simple homepage message from app.js.
Step 3: Build the Docker Image with Cloud Build
Enable Required Services
gcloud services enable artifactregistry.googleapis.com \
cloudbuild.googleapis.com \
run.googleapis.com
Create a Docker Repository in Artifact Registry
gcloud artifacts repositories create sample-app-repo \
--repository-format=docker \
--location=$REGION \
--description="Docker repo for sample Node app"
Configure Docker Authentication
gcloud auth configure-docker $REGION-docker.pkg.dev
Build and Push Docker Image
gcloud builds submit --tag $REGION-docker.pkg.dev/$PROJECT_ID/sample-app-repo/sample-app:1.0.0
This command builds your Docker image and pushes it directly to your private Docker repository.
Step 4: Deploy to Cloud Run
Use the following command to deploy your app:
gcloud run deploy sample-app \
--image $REGION-docker.pkg.dev/$PROJECT_ID/sample-app-repo/sample-app:1.0.0 \
--platform managed \
--region $REGION \
--allow-unauthenticated
Once deployed, the terminal will return a live URL like:
https://sample-app-xyz-uc.a.run.app
Open the URL in your browser. You should see:
Hello, world! This is Sample app version 3🚀
Step 5: Modify & Redeploy (Zero Downtime)
Let's simulate an update to your app (e.g., a marketing text change).
Update the Message in app.js
sed -i 's/This is Sample app version 3🚀/This is Sample app version 4🔥/' app.js
This command updates the welcome message inside your Express app.
Rebuild and Push the Updated Image
gcloud builds submit --tag $REGION-docker.pkg.dev/$PROJECT_ID/sample-app-repo/sample-app:2.0.0
Redeploy to Cloud Run
gcloud run deploy sample-app \
--image $REGION-docker.pkg.dev/$PROJECT_ID/sample-app-repo/sample-app:2.0.0 \
--platform managed \
--region $REGION \
--allow-unauthenticated
Cloud Run will automatically create a new revision and switch traffic to it with zero downtime.
Refresh your app’s URL — your updated message should now appear:
Hello, world! This is Sample app version 4🔥
Optional: Adjust Concurrency
To simulate function-like concurrency limits, you can set it to 1:
gcloud run deploy sample-app \
--image $REGION-docker.pkg.dev/$PROJECT_ID/sample-app-repo/sample-app:2.0.0 \
--platform managed \
--region $REGION \
--concurrency 1 \
--allow-unauthenticated
This limits each container to handling 1 request at a time. You can later restore it by setting:
--concurrency 80
Learn More
Final Thoughts
In this guide, you:
- Cloned and understood a real Node.js app
- Built and stored a Docker container in Artifact Registry
- Deployed it on Cloud Run with zero configuration
- Rolled out updates with zero downtime
This workflow scales beautifully from MVPs to microservices and production workloads.
Stay Connected
Want more tutorials on DevOps, cloud, and serverless technologies? Subscribe to the blog for hands-on walkthroughs, cheat sheets, and more!
