Skip to main content

Creating a Application

1. Create an Application​

Click the "Deploy GitHub Code" button on the left. You will see the following screen:

create_app

The icon next to "Github" represents the Github username. Use the search bar to find repositories under that user.

Example:

  • Username: leapcell
  • Search query: blog

2. Provide Project Information for Packaging and Deployment​

See the Deployment Form section below for the meanings of the deployment form fields.

Click the "Next" button. You will see an information confirmation page. After confirming, click the "Deploy" button, and your service will start packaging and deploying.

3. Wait for Deployment​

See the Packaging and Deployment section below.

Packaging and Deployment​

Packaging involves bundling your code and dependencies into an image. Leapcell uses Kaniko for packaging, and the process takes place in a sandbox.

After packaging, the service is automatically deployed.

How to Package and Deploy​

Click the "Deploy" button in the upper right corner of the Service page. You will see the following screen:

deploy

Fields to fill include:

1. Runtime​

Choose your desired runtime. If you are unsure, you can select "Python Alpine," which is the minimal Python runtime.

Examples:

  • Python developer needing the requests library: Choose "Python Alpine."
  • AI developer needing Tensorflow and Pillow: Choose "Python Debian."
2. Build Command​

The Build Command is the packaging command. Essentially, you need to enter a bash script to package your service.

Examples:

  • Python developer needing the requests library:

    pip install requests

  • Web developer needing Flask and gunicorn:

    pip install -r requirements.txt

  • Next.js developer (using standalone mode):

    Add the deploy command to your package.json:

    "scripts": {
      "dev": "next dev",
      "build": "next build",
      "start": "next start",
      "lint": "next lint",
      "deploy": "next build && cp -r .next/static .next/standalone/.next/ && cp -r public .next/standalone/"
    },
    npm install

&& npm run deploy


- AI developer needing `Tensorflow` and `Pillow`:

```bash
apt-get update && apt-get install -y libsm6 libxext6 libxrender-dev
pip install tensorflow pillow
3. Start Command​

The Start Command is the command to start your service. Essentially, you need to enter a bash script to start your service.

This command executes each time the service starts. Therefore, it is best to keep it as fast as possible. Avoid starting multiple workers or processes, as this can lead to longer cold start times.

Examples:

  • Python web developer using Flask:

    gunicorn -w 1 -b :8080 app:app

  • Express.js developer:

    node app.js

  • Next.js developer:

    Because Next.js needs local storage for caching, it is recommended to use the standalone mode. Prepare a script run.sh:

    run.sh
    #!/bin/sh -x

    [ ! -d '/tmp/cache' ] && mkdir -p /tmp/cache

    node .next/standalone/server.js

    Start Command:

    ./run.sh
4. Serving Port​

The Serving Port is the port on which your service runs. Leapcell uses this port to confirm that your service has started successfully. If your service starts successfully but does not listen on this port, Leapcell will consider it a failure with a timeout.

It is generally recommended to use port 8080.

5. User Account Token Injection​

This option injects the following environment variables if selected:

Environment VariableDescription
LEAPCELL_API_KEYYour Rotation API Token
6. Root Directory (Optional)​

The Root Directory is the root directory of your code. If your code is in the root directory, you can leave this field empty. If your code is in a subdirectory, enter the name of the subdirectory (e.g., src).

7. Environment Variables​

Environment Variables allow you to set environment variables for your service. You can set them here, and they will override environment variables set in the Panel's Environment section. These variables are private and encrypted, making them suitable for sensitive information.

If left blank, default environment variables will be used.

8. Deploy​

Click the "Deploy" button, and your service will start packaging and deploying. You will see the following page:

deployments

The status indicates that packaging is in progress. Click the "Deployment Information" button to view real-time packaging logs:

build_log

After waiting for about a minute, you will see:

build_result

Click the arrow icon, and you will see information about your service.

9. Deployment Failure​

Packaging Failure

If your service fails to deploy, click the "Deployment Information" button to view real-time packaging logs. You can use the logs to determine the reason for deployment failure.

Startup Failure

If you deploy successfully but encounter failure when clicking the HTTP URL, refer to the section on Observability for debugging.