DEV Community

Ricards Taujenis
Ricards Taujenis

Posted on • Edited on

How to Deploy a Go service to GCP Cloud Run

Image description

Deploying a Go service to GCP Cloud Run involves several steps, including setting up a Dockerfile and configuring environment variables. 

This guide will walk you through the process.

If you would like to follow it in video format as well have it on *YouTube *here.

Set up your GCP Project

Start off by going to GCP create account if havent done so yet.

  1. Create a GCP Project.
  • Go to the GCP console and create a new project.

  • Note the project ID for deployment.

Image description

  1. Enable required APIs.
  • Enable the Cloud Run API and Container Registry API.
  1. Install Google Cloud SDK
  • Initialize your repository with gcloud init.

Create your Go Service

Ensure your Go app can run locally and set up a Dockerfile.

cmd/main.go

// cmd/main.go
func main() {
 flag.Parse()

 a := app.Application{}

 if err := a.LoadConfigurations(); err != nil {
        log.Fatalf("Failed to load configurations: %v", err)
    }

    if err := runtime.Start(&a); err != nil {
        log.Fatalf("Failed to start the application: %v", err)
    }

}
Enter fullscreen mode Exit fullscreen mode

runtime/base.go

func Start(a *app.Application) error {
    router := gin.New()

    router.Use(cors.New(md.CORSMiddleware()))

    api.SetCache(router, a.RedisClient)
    api.SetRoutes(router, a.FireClient, a.FireAuth, a.RedisClient)

    err := router.Run(":" + a.ListenPort)
    log.Printf("Starting server on port: %s", a.ListenPort)
    if err != nil {
        return err
    }

    return nil
}
Enter fullscreen mode Exit fullscreen mode

Create Dockerfile

# Use the official Go image as the base image
FROM golang:1.18

WORKDIR /app

# Copy the Go module files
COPY go.mod go.sum ./

RUN go mod download

# Copy the rest of the application code
COPY . .

RUN go build -o main  ./cmd/main.go

CMD ["./main"]
Enter fullscreen mode Exit fullscreen mode

Set Up Env variables

Use shell script to automate setting env variables for GCP 

as env-variables.sh.

// env-variables.sh
#!/bin/bash

# Environment variables
export PROJECT_ID=recepies-6e7c0
export REGION=europe-west1

export REDIS_URL="rediss://default:AVrvA....-lemur-23279.u....:6379"
export FIREBASE_ACCOUNT_KEY="/app/config/account_key.json"
export CLIENT_URL="https://.....vercel.app/"
Enter fullscreen mode Exit fullscreen mode

Deployment script as deploy-with-yaml.sh.

#!/bin/bash

source env-variables.sh

#Comment if correctly deployed
docker build -t gcr.io/$PROJECT_ID/recipe-server:latest .
docker push gcr.io/$PROJECT_ID/recipe-server:latest

#Uncomment if json needs to be added to GCP 
# gcloud secrets create firebase-account-key --data-file=/mnt/c/own_dev/RecipesApp/server/config/account_key.json --project=recepies-6e7c0

#Add permission IAM
gcloud projects add-iam-policy-binding recepies-6e7c0 \
    --member="serviceAccount:service-988443547488@serverless-robot-prod.iam.gserviceaccount.com" \
    --role="roles/artifactregistry.reader"

gcloud run deploy recipe-service \
  --image gcr.io/$PROJECT_ID/recipe-server:latest \
  --region $REGION \
  --platform managed \
  --set-env-vars REDIS_URL=$REDIS_URL,CLIENT_URL=$CLIENT_URL,FIREBASE_ACCOUNT_KEY=$FIREBASE_ACCOUNT_KEY
Enter fullscreen mode Exit fullscreen mode

Deployment to your GCP Cloud Run

Run the deployment script

env-variables.sh

Common Issues and Troubleshooting

  • Permission Issues: Ensure the Cloud Run Service Agent has permission to read the image.
  • Environment Variables: Verify that all required environment variables are set correctly.
  • Port Configuration: Ensure the PORT environment variable is set correctly.

Image description

When all set up as needed you will see Image being build and pushed to your GCP project Artifact Registry. In the end i got this.

a9099c3159f5: Layer already exists
latest: digest: sha256:8c98063cd5b383df0b444c5747bb729ffd17014d42b049526b8760a4b09e5df1 size: 2846
Deploying container to Cloud Run service [recipe-service] in project [recepies-6e7c0] region [europe-west1]
✓ Deploying... Done.
  ✓ Creating Revision...
  ✓ Routing traffic...
Done.
Service [recipe-service] revision [recipe-service-00024-5mh] has been deployed and is serving 100 percent of traffic.
Service URL: https://recipe-service-819621241045.europe-west1.run.app
Enter fullscreen mode Exit fullscreen mode

There is a standart error that I came across multiple times 👇

Deploying container to Cloud Run service [recipe-service] in project [recepies-6e7c0] region [europe-west1] X Deploying… - Creating Revision… . Routing traffic… Deployment failed ERROR: 
(gcloud.run.deploy) Revision 'recipe-service-00005-b6h' 
is not ready and cannot serve traffic. Google Cloud Run Service Agent service-819621241045@serverless-robot-prod.iam.gserviceaccount.com must have permission to read the image, 
gcr.io/loyal-venture-436807-p7/recipe-server:latest. Ensure that the provided container image URL is correct and that the above account has permission to access the image. If you just enabled the Cloud Run API, the permissions might take a few minutes to propagate. Note that the image is from project [loyal-venture-436807-p7], which is not the same as this project [recepies-6e7c0]. Permission must be granted to the Google Cloud Run Service 
Agent service-819621241045@serverless-robot-prod.iam.gserviceaccount.com from this project. See https://cloud.google.com/run/docs/deploying#other-projects
Enter fullscreen mode Exit fullscreen mode

Often it states that PORT=8080 could not be set but main issue is something else like env variable not set or in my case firebase account_key.json incorrectly set for deployment.


When all is set you can test the connection and do requests.

I have my frontend deployed in Vercel and bellow you can see my Cloud Run Logs

Image description

Deploying a Go service to GCP Cloud Run can be streamlined with a few key configurations and automation scripts. 

Although there might be some common errors, such as permission issues or incorrect environment variables, understanding how to troubleshoot them through Cloud Run logs ensures a smooth deployment.
My repo you can find here.

Top comments (0)