Trigger Cloud Run functions using Cloud Tasks


This tutorial shows you how to use Cloud Tasks within an App Engine application to trigger a Cloud Run function and send a scheduled email.

Objectives

  • Understand the code in each of the components.
  • Create a SendGrid account.
  • Download the source code.
  • Deploy a Cloud Run function to receive Cloud Tasks requests and send an email via the SendGrid API.
  • Create a Cloud Tasks queue.
  • Create a service account to authenticate your Cloud Tasks requests.
  • Deploy the client code that allows a user to send an email.

Costs

Cloud Tasks, Cloud Run functions, and App Engine have a free tier, so as long as you are running the tutorial within the free tier of the given products it should not result in additional costs. For more information, see Pricing.

Before you begin

  1. Select or create a Google Cloud project.

    Go to the App Engine page

  2. Initialize an App Engine application in your project:

    1. On the Welcome to App Engine page click Create Application.

    2. Select a region for your application. This location will serve as the LOCATION_ID parameter for your Cloud Tasks requests, so make a note of it. Note that two locations, called europe-west and us-central in App Engine commands, are called, respectively, europe-west1 and us-central1 in Cloud Tasks commands.

    3. Select Node.js for the language and Standard for the environment.

    4. If the Enable billing popup appears, select your billing account. If you do not currently have a billing account, click Create billing account and follow the wizard.

    5. On the Get started page, click Next. You'll take care of this later.

  3. Enable the Cloud Run functions and Cloud Tasks APIs.

    Enable the apis

  4. Install and initialize the gcloud CLI.

Understanding the code

This section walks you through the app's code and explains how it works.

Creating the task

The index page is served using handlers in the app.yaml. The variables needed for task creation are passed in as environment variables.

runtime: nodejs16

env_variables:
  QUEUE_NAME: "my-queue"
  QUEUE_LOCATION: "us-central1"
  FUNCTION_URL: "https://<region>-<project_id>.cloudfunctions.net/sendEmail"
  SERVICE_ACCOUNT_EMAIL: "<member>@<project_id>.iam.gserviceaccount.com"

# Handlers for serving the index page.
handlers:
  - url: /static
    static_dir: static
  - url: /
    static_files: index.html
    upload: index.html

This code creates the endpoint /send-email. This endpoint handles form submissions from the index page and passes that data to the task creation code.

app.post('/send-email', (req, res) => {
  // Set the task payload to the form submission.
  const {to_name, from_name, to_email, date} = req.body;
  const payload = {to_name, from_name, to_email};

  createHttpTaskWithToken(
    process.env.GOOGLE_CLOUD_PROJECT,
    QUEUE_NAME,
    QUEUE_LOCATION,
    FUNCTION_URL,
    SERVICE_ACCOUNT_EMAIL,
    payload,
    date
  );

  res.status(202).send('📫 Your postcard is in the mail! 💌');
});

This code actually creates the task and sends it on to the Cloud Tasks queue. The code builds the task by:

  • Specifying the target type as HTTP Request.

  • Specifying the HTTP method to be used and the URL of the target.

  • Setting the Content-Type header to application/json so downstream applications can parse the structured payload.

  • Adding a service account email so that Cloud Tasks can provide credentials to the request target, which requires authentication. The service account is created separately.

  • Checking to make sure the user input for date is within the 30 days maximum and adding it to the request as field scheduleTime.

const MAX_SCHEDULE_LIMIT = 30 * 60 * 60 * 24; // Represents 30 days in seconds.

const createHttpTaskWithToken = async function (
  project = 'my-project-id', // Your GCP Project id
  queue = 'my-queue', // Name of your Queue
  location = 'us-central1', // The GCP region of your queue
  url = 'https://example.com/taskhandler', // The full url path that the request will be sent to
  email = '<member>@<project-id>.iam.gserviceaccount.com', // Cloud IAM service account
  payload = 'Hello, World!', // The task HTTP request body
  date = new Date() // Intended date to schedule task
) {
  // Imports the Google Cloud Tasks library.
  const {v2beta3} = require('@google-cloud/tasks');

  // Instantiates a client.
  const client = new v2beta3.CloudTasksClient();

  // Construct the fully qualified queue name.
  const parent = client.queuePath(project, location, queue);

  // Convert message to buffer.
  const convertedPayload = JSON.stringify(payload);
  const body = Buffer.from(convertedPayload).toString('base64');

  const task = {
    httpRequest: {
      httpMethod: 'POST',
      url,
      oidcToken: {
        serviceAccountEmail: email,
        audience: url,
      },
      headers: {
        'Content-Type': 'application/json',
      },
      body,
    },
  };

  const convertedDate = new Date(date);
  const currentDate = new Date();

  // Schedule time can not be in the past.
  if (convertedDate < currentDate) {
    console.error('Scheduled date in the past.');
  } else if (convertedDate > currentDate) {
    const date_diff_in_seconds = (convertedDate - currentDate) / 1000;
    // Restrict schedule time to the 30 day maximum.
    if (date_diff_in_seconds > MAX_SCHEDULE_LIMIT) {
      console.error('Schedule time is over 30 day maximum.');
    }
    // Construct future date in Unix time.
    const date_in_seconds =
      Math.min(date_diff_in_seconds, MAX_SCHEDULE_LIMIT) + Date.now() / 1000;
    // Add schedule time to request in Unix time using Timestamp structure.
    // https://googleapis.dev/nodejs/tasks/latest/google.protobuf.html#.Timestamp
    task.scheduleTime = {
      seconds: date_in_seconds,
    };
  }

  try {
    // Send create task request.
    const [response] = await client.createTask({parent, task});
    console.log(`Created task ${response.name}`);
    return response.name;
  } catch (error) {
    // Construct error for Stackdriver Error Reporting
    console.error(Error(error.message));
  }
};

module.exports = createHttpTaskWithToken;

Creating the email

This code creates the Cloud Run function that is the target for the Cloud Tasks request. It uses the request body to construct an email and send it via the SendGrid API.

const sendgrid = require('@sendgrid/mail');

/**
 * Responds to an HTTP request from Cloud Tasks and sends an email using data
 * from the request body.
 *
 * @param {object} req Cloud Function request context.
 * @param {object} req.body The request payload.
 * @param {string} req.body.to_email Email address of the recipient.
 * @param {string} req.body.to_name Name of the recipient.
 * @param {string} req.body.from_name Name of the sender.
 * @param {object} res Cloud Function response context.
 */
exports.sendEmail = async (req, res) => {
  // Get the SendGrid API key from the environment variable.
  const key = process.env.SENDGRID_API_KEY;
  if (!key) {
    const error = new Error(
      'SENDGRID_API_KEY was not provided as environment variable.'
    );
    error.code = 401;
    throw error;
  }
  sendgrid.setApiKey(key);

  // Get the body from the Cloud Task request.
  const {to_email, to_name, from_name} = req.body;
  if (!to_email) {
    const error = new Error('Email address not provided.');
    error.code = 400;
    throw error;
  } else if (!to_name) {
    const error = new Error('Recipient name not provided.');
    error.code = 400;
    throw error;
  } else if (!from_name) {
    const error = new Error('Sender name not provided.');
    error.code = 400;
    throw error;
  }

  // Construct the email request.
  const msg = {
    to: to_email,
    from: 'postcard@example.com',
    subject: 'A Postcard Just for You!',
    html: postcardHTML(to_name, from_name),
  };

  try {
    await sendgrid.send(msg);
    // Send OK to Cloud Task queue to delete task.
    res.status(200).send('Postcard Sent!');
  } catch (error) {
    // Any status code other than 2xx or 503 will trigger the task to retry.
    res.status(error.code).send(error.message);
  }
};

Preparing the application

Setting up SendGrid

  1. Create a SendGrid account.

  2. Create a SendGrid API key:

    1. Log in to your SendGrid account.

    2. In the left hand navigation open Settings and click API Keys.

    3. Click Create API Key and select restricted access. Under the Mail Send header, select Full Access.

    4. Copy the API Key when it is displayed (you will only see this once, make sure you paste it somewhere so you can use it later on).

Downloading the source code

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

    git clone https://github.com/GoogleCloudPlatform/nodejs-docs-samples.git
    
  2. Change to the directory that contains the sample code:

    cd cloud-tasks/
    

Deploying the Cloud Run function

  1. Navigate to the function/ directory:

    cd function/
    
  2. Deploy the function:

    gcloud functions deploy sendEmail --runtime nodejs14 --trigger-http \
      --no-allow-unauthenticated \
      --set-env-vars SENDGRID_API_KEY=SENDGRID_API_KEY \

    Replace SENDGRID_API_KEY with your API key.

    This command uses flags:

    • --trigger-http to specify the Cloud Run functions trigger type.

    • --no-allow-unauthenticated to specify the function invocation requires authentication.

    • --set-env-var to set your SendGrid credentials

  3. Set access control for the function to only allow authenticated users.

    1. Select the sendEmail function in the Cloud Run functions UI.

    2. If you don't see permissions info for sendEmail, click SHOW INFO PANEL in the upper right hand corner.

    3. Click the Add principals button above.

    4. Set New principals to allAuthenticatedUsers.

    5. Set the Role.

      • First generation (1st gen) functions: set the role to Cloud Function Invoker
      • Second generation (2nd gen) functions: set the role to Cloud Run Invoker
    6. Click SAVE.

Creating a Cloud Tasks queue

  1. Create a queue using the following gcloud command:

    gcloud tasks queues create my-queue --location=LOCATION

    Replace LOCATION with your preferred location for the queue, for example, us-west2. If you do not specify the location, the gcloud CLI picks the default.

  2. Verify that it was created successfully:

    gcloud tasks queues describe my-queue --location=LOCATION

    Replace LOCATION with the location of the queue.

Creating a service account

The Cloud Tasks request must provide credentials in the Authorization header for the Cloud Run function to authenticate the request. This service account allows Cloud Tasks to create and add an OIDC token for that purpose.

  1. In the Service accounts UI, click +CREATE SERVICE ACCOUNT.

  2. Add a service account name(friendly display name) and select create.

  3. Set the Role and click Continue.

    • First generation (1st gen) functions: set the role to Cloud Function Invoker
    • Second generation (2nd gen) functions: set the role to Cloud Run Invoker
  4. Select Done.

Deploying the endpoint and the task creator to App Engine

  1. Navigate to app/ directory:

    cd ../app/
    
  2. Update variables in the app.yaml with your values:

    env_variables:
      QUEUE_NAME: "my-queue"
      QUEUE_LOCATION: "us-central1"
      FUNCTION_URL: "https://<region>-<project_id>.cloudfunctions.net/sendEmail"
      SERVICE_ACCOUNT_EMAIL: "<member>@<project_id>.iam.gserviceaccount.com"

    To find your queue location, use the following command:

    gcloud tasks queues describe my-queue --location=LOCATION

    Replace LOCATION with the location of the queue.

    To find your function URL, use the following command:

    gcloud functions describe sendEmail
  3. Deploy the application to the App Engine standard environment, using the following command:

    gcloud app deploy
  4. Open the application to send a postcard as an email:

    gcloud app browse

Clean up

After you finish the tutorial, you can clean up the resources that you created so that they stop using quota and incurring charges. The following sections describe how to delete or turn off these resources.

Deleting Resources

You can clean up the resources that you created on Google Cloud so they won't take up quota and you won't be billed for them in the future. The following sections describe how to delete or turn off these resources.

Delete the Cloud Run function

  1. Go to the Cloud Run functions page in the Google Cloud console.

    Go to the Cloud Run functions page.

  2. Click the checkboxes next to your functions.

  3. Click the Delete button at the top of the page and confirm your delete.

Delete the Cloud Tasks queue

  1. Open the Cloud Tasks queues page in the console.

    Go to the Cloud Tasks queues page

  2. Select the name of the queue you want to delete and click Delete queue.

  3. Confirm the action.

Deleting the project

The easiest way to eliminate billing is to delete the project that you created for the tutorial.

To delete the project:

  1. In the Google Cloud console, go to the Manage resources page.

    Go to Manage resources

  2. In the project list, select the project that you want to delete, and then click Delete.
  3. In the dialog, type the project ID, and then click Shut down to delete the project.

What's next