Loading...

Getting started

Front End Hosting - Advanced provisions a Node.js application with the following environments:

  • Development (Dev)

  • Staging (Stage)

  • Production (Prod)

Viewing an application

  1. Sign in to the Cloud Platform user interface.
  2. In the top menu, click Develop.
  3. Select your organization.
  4. Select an application.
  5. Click View <application name>.

    New environments are provisioned with the default content page.

  6. To verify that a site works correctly, visit the default URL of a specific environment.

Setting up

  1. Add a public key to your Acquia profile.
  2. Clone the application’s repository.
  3. Copy your code to a new Git repository.
  1. Configure the package.json file.

    The package.json file must be located in the root folder of your project that contains the following commands. These are required to host a NodeJS site.

  2. Build command configuration.

    Set npm install && npm run build as the build command in the NodeJS Hosting build system to ensure a robust and efficient deployment process.

    "scripts": {
        ...
        "build": "npm install && next build",
        "start": "next start",
      },
     	 

Framework-based custom configuration

​Configure the start script in your package.json file in order to initiate the production server. Use the command that corresponds to your specific framework.

{
  "scripts": {
    "build": "npm install",
    "start": "node index.js"
  },
}

 

To serve production builds of frameworks like Angular, React, and Vue, Acquia recommends you to use either serve or http-server. These Node.js packages provide simple and efficient ways to host static files.

Installation

Install your preferred package using npm:

$ npm install serve

or

$ npm install http-server

Usage

After you build the project (typically into a dist directory), start the production server with the following commands:

serve -s dist

 

http-server dist

You can also include these commands in the start script of your package.json for easy deployment.

Implementing autoscaling

Autoscaling is vital to enhance the performance of your application and enable it to handle variable workloads. This feature dynamically adjusts the number of active instances of your environment based on its real-time performance metrics. The enablement process revolves around integrating the acquia-autoscaling javascript into your project that collects the necessary metrics.

To implement autoscaling in your application:

  1. Add the acquia-autoscaling.js script to the root directory of your project.

    The following code snippets show the content of the scripts:

    // ES5/CommonJS syntax
    // acquia-autoscaling.js
    const { collectDefaultMetrics, Registry } = require("prom-client");
    const http = require("http");
    const net = require("net");
    function startMetricsSvr() {
        // Create a new registry to track default metrics
        const register = new Registry();
       // Enable the collection of default metrics
        collectDefaultMetrics({ register });
       // Set up an HTTP server to expose the metrics
       const server = http.createServer(async (req, res) => {
            if (req.url === "/metrics") {
                try {
                    // Respond with metrics in Prometheus format
                    res.setHeader("Content-Type", register.contentType);
                    const metrics = await register.metrics();
                    res.end(metrics);
                } catch (err) {
                    res.writeHead(500);
                    res.end(err.message);
                }
            } else {
                res.writeHead(404);
                res.end("Not Found");
            }
        });
       const monitoringPort = 9100;
        server.listen(monitoringPort, () => {
            console.log(
                `Acquia Node.js Monitoring server listening on port ${monitoringPort}`
            );
        });
        // Prevent the Node.js process from exiting
        process.stdin.resume();
    }
    const originalListen = net.Server.prototype.listen;
    let injected = false;
    net.Server.prototype.listen = function (...args) {
        try {
            const port =
                typeof args[0] === "number"
                    ? args[0]
                    : typeof args[0] === "string" && !isNaN(parseInt(args[0]))
                    ? parseInt(args[0])
                    : args[0] && typeof args[0].port === "number"
                    ? args[0].port
                    : args[0] &&
                      typeof args[0].port === "string" &&
                      !isNaN(parseInt(args[0].port))
                    ? parseInt(args[0].port)
                    : null;
            if (port === 3000 && !injected) {
                injected = true;
                setTimeout(startMetricsSvr, 1000);
            }
        } catch (err) {
            console.error(err);
        }
       return originalListen.apply(this, args);
    };
  2. Edit the package.json file to add the NODE_OPTIONS='-r ./acquia-autoscaling.js' string before the original start command of your application:

    {
     "scripts": {
       "start": "NODE_OPTIONS='-r ./acquia-autoscaling.js' node app.js",
       "build": "npm install"
     },
    }
  3. Run the following NPM command in your Node.js code repository on your local machine to install the prom-client dependency:

    npm install prom-client

    This command adds the prom-client package to the dependencies of your application.

    Verify the functionality of your application to ensure that metrics are being correctly exposed and autoscaling is operational.

    After successful implementation and verification, your application can scale automatically to adapt to usage patterns and sustain optimal performance.

Environment variables configuration

For information about how to configure environment variables, visit Creating custom environment variables.

To use the latest environment variable values in your Node.js code, incorporate them during the Node.js build process. Build-time environment variables are injected during the build and remain fixed until redeployment. Runtime environment variables are read by the application at execution and update without redeployment.

Deployment workflow (without external CI/CD or Code Studio)

  1. Visit Code workflows with Cloud Platform.
  2. Deployment implicitly triggers the build.
  3. Set the post deployment cloud actions to clear Varnish cache after you deploy the branch through Cloud UI.

  4. After the Code Deployment Task for the deployed branch completes, the latest code changes are visible in the environment.

Subscription details

For information about how to access your subscription details, visit Viewing subscriptions.

Monitoring

To monitor instructions, visit Using Stack Metrics to monitor activity on your environments.

Build minutes

As part of the code build/deployment process, multiple commands run in the backend to produce a build artifact and deploy it to an environment. These include:

  • git commands to clone code from an application repository.
  • npm run-build commands to produce the deployable bundle of an environment branch.
  • The time taken to deploy the build to the specific environment.

Application deployment

The total build time is tracked as build minutes.

  • If a customer exceeds their allocated build minutes, they may continue to use Front End Hosting - Advanced without interruption. An Account Manager is notified, and the account holder is contacted to discuss upgrading their assigned entitlement for build minutes.
  • Build is implicitly triggered with each deployment. Build logs are visible in “Task Logs” in Cloud UI. A build is retriggered 2 times upon failure.
  • Post Deployment Cloud Actions can be enabled/disabled to clear Varnish/CDN cache.

Domain configuration

CDN configuration

Readiness probe path and authentication considerations

In Frontend Hosting – Advanced environments, the Kubernetes readiness probe uses the root path (/) to check application health. This probe periodically sends HTTP requests to the / endpoint to verify readiness.

Do not apply any authentication, such as Basic Auth, OAuth, JWT, or SSO, to the / endpoint. When you force authentication on this path, the Kubernetes readiness probe is prevented from working, which can result in 503 Service Unavailable errors and may cause the application to be marked as unhealthy.

If your site requires authentication, configure your authentication middleware to bypass checks for requests from the Kubernetes probe user agent, such as kube-probe/1.31+, on the / endpoint.

Follow this guidance to ensure reliable health checks and stable application operation within the Acquia environment.

Did not find what you were looking for?

If this content did not answer your questions, try searching or contacting our support team for further assistance.

Back to Section navigation