End user authentication for Cloud Run tutorial

Stay organized with collections Save and categorize content based on your preferences.

Setting up gcloud defaults

To configure gcloud with defaults for your Cloud Run service:

1. Set your default project:

   gcloudconfigsetprojectPROJECT_ID

   Replace PROJECT_ID with the name of the project you created for this tutorial.

2. Configure gcloud for your chosen region:

   gcloudconfigsetrun/regionREGION

   Replace REGION with the supported Cloud Run region of your choice.

Retrieving the code sample

To retrieve the code sample for use:

1. Clone the sample app repository to your local machine:

   Node.js

   gitclonehttps://github.com/GoogleCloudPlatform/nodejs-docs-samples.git

   Alternatively, you can download the sample as a zip file and extract it.

   Python

   gitclonehttps://github.com/GoogleCloudPlatform/python-docs-samples.git

   Alternatively, you can download the sample as a zip file and extract it.

   Java

   gitclonehttps://github.com/GoogleCloudPlatform/java-docs-samples.git

   Alternatively, you can download the sample as a zip file and extract it.

2. Change to the directory that contains the Cloud Run sample code:

   Node.js

   cdnodejs-docs-samples/run/idp-sql/

   Python

   cdpython-docs-samples/run/idp-sql/

   Java

   cdjava-docs-samples/run/idp-sql/

Visualizing the architecture

1. An end-user makes the first request to the Cloud Run server.

2. The client loads in the browser.

3. The user provides login credentials through the Google sign-in dialog from Identity Platform. An alert welcomes the signed-in user.

4. Control is redirected back to the server. The end-user votes using the client, which fetches an ID token from Identity Platform and adds it to the vote request header.

5. When the server receives the request, it verifies the Identity Platform ID token, confirming that the end-user is appropriately authenticated. Then server sends the vote to Cloud SQL, using its own credentials.

Understanding the core code

The sample is implemented as client and server, as described next.

Integrating with Identity Platform: client-side code

This sample uses Firebase SDKs to integrate with Identity Platform in order to sign-in and manage users. To connect to Identity Platform, the client-side JavaScript holds the reference to the project's credentials as a config object and imports the necessary Firebase JavaScript SDKs:

constconfig={apiKey:'API_KEY',authDomain:'PROJECT_ID.firebaseapp.com',};

<!-- Firebase App (the core Firebase SDK) is always required and must be listed first--> <script src="https://www.gstatic.com/firebasejs/7.18/firebase-app.js"></script> <!-- Add Firebase Auth service--> <script src="https://www.gstatic.com/firebasejs/7.18/firebase-auth.js"></script>

The Firebase JavaScript SDK handles the sign-in flow by prompting the end-user to sign-in to their Google Account via a popup window. It then redirects them back to the service.

functionsignIn(){constprovider=newfirebase.auth.GoogleAuthProvider();provider.addScope('https://www.googleapis.com/auth/userinfo.email');firebase.auth().signInWithPopup(provider).then(result=>{// Returns the signed in user along with the provider's credentialconsole.log(`${result.user.displayName} logged in.`);window.alert(`Welcome ${result.user.displayName}!`);}).catch(err=>{console.log(`Error during sign in: ${err.message}`);window.alert('Sign in failed. Retry or check your browser logs.');});}

When a user successfully signs in, the client uses Firebase methods to mint an ID token. The client adds the ID token to the Authorization header of its request to the server.

asyncfunctionvote(team){if(firebase.auth().currentUser){// Retrieve JWT to identify the user to the Identity Platform service.// Returns the current token if it has not expired. Otherwise, this will// refresh the token and return a new one.try{consttoken=awaitfirebase.auth().currentUser.getIdToken();constresponse=awaitfetch('/',{method:'POST',headers:{'Content-Type':'application/x-www-form-urlencoded',Authorization:`Bearer ${token}`,},body:'team='+team,// send application data (vote)});if(response.ok){consttext=awaitresponse.text();window.alert(text);window.location.reload();}}catch(err){console.log(`Error when submitting vote: ${err}`);window.alert('Something went wrong... Please try again!');}}else{window.alert('User not signed in.');}}

Integrating with Identity Platform: server-side code

The server uses the Firebase Admin SDK to verify the user ID token sent from the client. If the provided ID token has the correct format, is not expired, and is properly signed, the method returns the decoded ID token. The server extracts the Identity Platform uid for that user.

constfirebase=require('firebase-admin');// Initialize Firebase Admin SDKfirebase.initializeApp();// Extract and verify Id Token from headerconstauthenticateJWT=(req,res,next)=>{constauthHeader=req.headers.authorization;if(authHeader){consttoken=authHeader.split(' ')[1];// If the provided ID token has the correct format, is not expired, and is// properly signed, the method returns the decoded ID tokenfirebase.auth().verifyIdToken(token).then(decodedToken=>{constuid=decodedToken.uid;req.uid=uid;next();}).catch(err=>{req.logger.error(`Error with authentication: ${err}`);returnres.sendStatus(403);});}else{returnres.sendStatus(401);}};

defjwt_authenticated(func:Callable[...,int])-> Callable[...,int]:"""Use the Firebase Admin SDK to parse Authorization header to verify the user ID token. The server extracts the Identity Platform uid for that user. """@wraps(func)defdecorated_function(*args:a,**kwargs:a)-> a:header=request.headers.get("Authorization",None)ifheader:token=header.split(" ")[1]try:decoded_token=firebase_admin.auth.verify_id_token(token)exceptExceptionase:logger.exception(e)returnResponse(status=403,response=f"Error with authentication: {e}")else:returnResponse(status=401)request.uid=decoded_token["uid"]returnfunc(*args,**kwargs)returndecorated_function

/** Extract and verify Id Token from header */privateStringauthenticateJwt(Map<String,String>headers){StringauthHeader=(headers.get("authorization")!=null)?headers.get("authorization"):headers.get("Authorization");if(authHeader!=null){StringidToken=authHeader.split(" ")[1];// If the provided ID token has the correct format, is not expired, and is// properly signed, the method returns the decoded ID tokentry{FirebaseTokendecodedToken=FirebaseAuth.getInstance().verifyIdToken(idToken);Stringuid=decodedToken.getUid();returnuid;}catch(FirebaseAuthExceptione){logger.error("Error with authentication: "+e.toString());thrownewResponseStatusException(HttpStatus.FORBIDDEN,"",e);}}else{logger.error("Error no authorization header");thrownewResponseStatusException(HttpStatus.UNAUTHORIZED);}}

Connecting the server to Cloud SQL

The server connects to the Cloud SQL instance Unix domain socket using the format: /cloudsql/CLOUD_SQL_CONNECTION_NAME.

/** * Connect to the Cloud SQL instance through UNIX Sockets * * @param {object} credConfig The Cloud SQL connection configuration from Secret Manager * @returns {object} Knex's PostgreSQL client */constconnectWithUnixSockets=asynccredConfig=>{constdbSocketPath=process.env.DB_SOCKET_PATH||'/cloudsql';// Establish a connection to the databasereturnKnex({client:'pg',connection:{user:credConfig.DB_USER,// e.g. 'my-user'password:credConfig.DB_PASSWORD,// e.g. 'my-user-password'database:credConfig.DB_NAME,// e.g. 'my-database'host:`${dbSocketPath}/${credConfig.CLOUD_SQL_CONNECTION_NAME}`,},...config,});};

definit_unix_connection_engine(db_config:dict[str,int])-> sqlalchemy.engine.base.Engine:"""Initializes a Unix socket connection pool for a Cloud SQL instance of PostgreSQL. Args: db_config: a dictionary with connection pool config Returns: A SQLAlchemy Engine instance. """creds=credentials.get_cred_config()db_user=creds["DB_USER"]db_pass=creds["DB_PASSWORD"]db_name=creds["DB_NAME"]db_socket_dir=creds.get("DB_SOCKET_DIR","/cloudsql")cloud_sql_connection_name=creds["CLOUD_SQL_CONNECTION_NAME"]pool=sqlalchemy.create_engine(# Equivalent URL:# postgres+pg8000://<db_user>:<db_pass>@/<db_name># ?unix_sock=<socket_path>/<cloud_sql_instance_name>/.s.PGSQL.5432sqlalchemy.engine.url.URL.create(drivername="postgresql+pg8000",username=db_user,# e.g. "my-database-user"password=db_pass,# e.g. "my-database-password"database=db_name,# e.g. "my-database-name"query={"unix_sock":f"{db_socket_dir}/{cloud_sql_connection_name}/.s.PGSQL.5432"# e.g. "/cloudsql", "<PROJECT-NAME>:<INSTANCE-REGION>:<INSTANCE-NAME>"},),**db_config,)pool.dialect.description_encoding=Nonelogger.info("Database engine initialized from unix connection")returnpool

# Uncomment and add env vars for local development# spring.datasource.username=${DB_USER}# spring.datasource.password=${DB_PASSWORD}# spring.cloud.gcp.sql.database-name=${DB_NAME}# spring.cloud.gcp.sql.instance-connection-name=${CLOUD_SQL_CONNECTION_NAME} 

privatefinalJdbcTemplatejdbcTemplate;publicVoteController(JdbcTemplatejdbcTemplate){this.jdbcTemplate=jdbcTemplate;}

Handling sensitive configuration with Secret Manager

Secret Manager provides centralized and secure storage of sensitive data such as Cloud SQL configuration. The server injects the Cloud SQL credentials from Secret Manager at runtime via an environment variable. Learn more about Using secrets with Cloud Run.

// CLOUD_SQL_CREDENTIALS_SECRET is the resource ID of the secret, passed in by environment variable.// Format: projects/PROJECT_ID/secrets/SECRET_ID/versions/VERSIONconst{CLOUD_SQL_CREDENTIALS_SECRET}=process.env;if(CLOUD_SQL_CREDENTIALS_SECRET){try{// Parse the secret that has been added as a JSON string// to retrieve database credentialsreturnJSON.parse(CLOUD_SQL_CREDENTIALS_SECRET.toString('utf8'));}catch(err){throwError(`Unable to parse secret from Secret Manager. Make sure that the secret is JSON formatted: ${err}`);}}

defget_cred_config()-> dict[str,str]:"""Retrieve Cloud SQL credentials stored in Secret Manager or default to environment variables. Returns: A dictionary with Cloud SQL credential values """secret=os.environ.get("CLOUD_SQL_CREDENTIALS_SECRET")ifsecret:returnjson.loads(secret)

/** Retrieve config from Secret Manager */publicstaticHashMap<String,Object>getConfig(){Stringsecret=System.getenv("CLOUD_SQL_CREDENTIALS_SECRET");if(secret==null){thrownewIllegalStateException("\"CLOUD_SQL_CREDENTIALS_SECRET\" is required.");}try{HashMap<String,Object>config=newGson().fromJson(secret,HashMap.class);returnconfig;}catch(JsonSyntaxExceptione){logger.error("Unable to parse secret from Secret Manager. Make sure that it is JSON formatted: "+e);thrownewRuntimeException("Unable to parse secret from Secret Manager. Make sure that it is JSON formatted.");}}

Set up Identity Platform

Identity Platform requires manual setup in the Google Cloud console.

1. In the Google Cloud console, enable the Identity Platform API:

   Enable the API

2. Configure your project:

   1. In a new window, go to the Google Auth Platform > Overview page.

      Go to Overview

   2. Click Get Started and follow the project configuration setup.

   3. In the App information dialog:

      1. Supply the application name.
      2. Select one of the displayed user support emails.

   4. In the Audience dialog, select External.

   5. In the Contact information dialog, enter a contact email.

   6. Agree to the user data policy, then click Create.

3. Create and obtain your OAuth Client ID and Secret:

   1. In the Google Cloud console, go to the APIs & Services > Credentials page.

      Go to Credentials

   2. At the top of the page, click Create Credentials and select OAuth client ID.

   3. From Application Type, select Web Application and supply the name.

   4. Click Create

   5. The client_id and client_secret values will be used in the next step.

4. Configure Google as a provider:

   1. In the Google Cloud console, go to the Identity Providers page.

      Go to Identity Providers

   2. Click Add A Provider.

   3. Select Google from the list.

   4. In the Web SDK Configuration settings, enter the client_id and client_secret values from the previous step.

   5. Under Configure your application, click Setup Details.

5. Copy the configuration to your application:

   • Copy the apiKey and authDomain values into the sample's static/config.js to initialize the Identity Platform Client SDK.

Deploying the service

Follow the steps to complete infrastructure provisioning and deployment:

1. Create a Cloud SQL instance with postgreSQL database using the console or CLI:

   gcloudsqlinstancescreateCLOUD_SQL_INSTANCE_NAME\--database-version=POSTGRES_16\--region=CLOUD_SQL_REGION\--cpu=2\--memory=7680MB\--root-password=DB_PASSWORD

2. Add your Cloud SQL credential values to postgres-secrets.json:

   Node.js

   {"CLOUD_SQL_CONNECTION_NAME":"PROJECT_ID:REGION:INSTANCE","DB_NAME":"postgres","DB_USER":"postgres","DB_PASSWORD":"PASSWORD_SECRET"}

   Python

   {"CLOUD_SQL_CONNECTION_NAME":"PROJECT_ID:REGION:INSTANCE","DB_NAME":"postgres","DB_USER":"postgres","DB_PASSWORD":"PASSWORD_SECRET"}

   Java

   {"spring.cloud.gcp.sql.instance-connection-name":"PROJECT_ID:REGION:INSTANCE","spring.cloud.gcp.sql.database-name":"postgres","spring.datasource.username":"postgres","spring.datasource.password":"PASSWORD_SECRET"}

3. Create a versioned secret using the console or CLI:

   gcloudsecretscreateidp-sql-secrets\--replication-policy="automatic"\--data-file=postgres-secrets.json

4. Create a service account for the server using the console or CLI:

   gcloudiamservice-accountscreateidp-sql-identity

5. Grant roles for Secret Manager and Cloud SQL access using the console or CLI:

   1. Allow the service account associated with the server to access the created secret:

      gcloudsecretsadd-iam-policy-bindingidp-sql-secrets\--memberserviceAccount:idp-sql-identity@PROJECT_ID.iam.gserviceaccount.com\--roleroles/secretmanager.secretAccessor

   2. Allow the service account associated with the server to access Cloud SQL:

      gcloudprojectsadd-iam-policy-bindingPROJECT_ID\--memberserviceAccount:idp-sql-identity@PROJECT_ID.iam.gserviceaccount.com\--roleroles/cloudsql.client

6. Create an Artifact Registry:

   gcloudartifactsrepositoriescreateREPOSITORY\--repository-formatdocker\--locationREGION

   • REPOSITORY is the name of the repository. For each repository location in a project, repository names must be unique.

7. Build the container image using Cloud Build:

   Node.js

   gcloudbuildssubmit--tagREGION-docker.pkg.dev/PROJECT_ID/REPOSITORY/idp-sql

   Python

   gcloudbuildssubmit--tagREGION-docker.pkg.dev/PROJECT_ID/REPOSITORY/idp-sql

   Java

   This sample uses Jib to build Docker images using common Java tools. Jib optimizes container builds without the need for a Dockerfile or having Docker installed. Learn more about building Java containers with Jib.

   1. Use the gcloud credential helper to authorize Docker to push to your Artifact Registry.

      gcloudauthconfigure-docker

   2. Use the Jib Maven Plugin to build and push the container to Artifact Registry.

      mvncompilejib:build-Dimage=REGION-docker.pkg.dev/PROJECT_ID/REPOSITORY/idp-sql

8. Deploy the container image to Cloud Run using the console or CLI. Note that the server is deployed to allow unauthenticated access. This is so that the user can load the client and begin the process. The server verifies the ID token added to the vote request manually, authenticating the end-user.

   gcloudrundeployidp-sql\--imageREGION-docker.pkg.dev/PROJECT_ID/REPOSITORY/idp-sql\--allow-unauthenticated\--service-accountidp-sql-identity@PROJECT_ID.iam.gserviceaccount.com\--add-cloudsql-instancesPROJECT_ID:REGION:CLOUD_SQL_INSTANCE_NAME\--update-secretsCLOUD_SQL_CREDENTIALS_SECRET=idp-sql-secrets:latest

   Also note the flags, --service-account, --add-cloudsql-instances, and --update-secrets, which specify the service identity, the Cloud SQL instance connection, and the secret name with version as an environment variable, respectively.

Finishing touches

Identity Platform requires that you authorize the Cloud Run service URL as an allowed redirect after it has signed in the user:

1. Edit the Google provider by clicking the pen icon in the Identity Providers page.

2. Click Add Domain under Authorized Domains on the right panel, and enter the Cloud Run service URL.

   You can locate the service URL in the logs after the build or deployment or you can find it anytime using:

   gcloudrunservicesdescribeidp-sql--format'value(status.url)'

3. Go to the APIs & Services > Credentials page

   1. Click the pencil icon beside your OAuth Client ID to edit it and under the Authorized redirect URIs click theAdd URI button.

   2. In the field copy and paste the following URL and click the Save button at the bottom of the page.

   https://PROJECT_ID.firebaseapp.com/__/auth/handler

Trying it out

To try out the complete service:

1. Navigate your browser to the URL provided by the deployment step above.

2. Click the Sign in with Google button and go through the authentication flow.

3. Add your vote!

   It should look like this:

   

If you choose to continue developing these services, remember that they have restricted Identity and Access Management (IAM) access to the rest of Google Cloud and will need to be given additional IAM roles to access many other services.