Skip to content

Latest commit

 

History

History
112 lines (87 loc) · 4.87 KB

README_en.md

File metadata and controls

112 lines (87 loc) · 4.87 KB
Raw

Instructions for using webhook auto deployment

Introduction

This document explains how to use this script to implement automatic deployment of multiple projects using Webhooks.

Installation

Install globally using npm i @brewer/webhook-manager -g.

Quick Start

1. Configure projects

Configure the projects that need to be automatically deployed in the config.json file.

  • port The port on which the service will run, default is 3200.

  • project The projects to be deployed. Each (project) needs to define the following attributes:

  • name - The name of the project, used for logging purposes.

  • path - The absolute path of the project deployment directory, which must contain deploy.sh.

  • command (optional) - The command to be executed after deployment is complete. For example:

{
  "port": 3200,
  "app1": { 
    "name": "My Project 1", 
    "path": "/var/www/project1",
    "command": "" 
  },
  "app2": {
    "name": "My Project 2",
    "path": "/var/www/project2",
    "command": "cd /var/www/project2 && git pull origin main && npm i && npm run build"
  }
}

To standardize and facilitate management, create a folder named webhookConfig (mkdir webhookConfig && cd webhookConfig) and create a config.json file in it.

2. Start the service

Run brewer-webhook in the webhookConfig folder to start the service. Logs will also be generated in this directory.

3. Check the service status

Access http://your-server-url:3200/webhook/health to check if the service is running properly. If it returns "health", the service is running properly. Deployment logs can be viewed in the ./logs/webhook.log file in the current directory.

4. Configure Webhooks according to the following steps

Configure a Webhook for each project in the code hosting platform (e.g. Github, Gitlab, Bitbucket). The Webhook address is http://your-server-url:3200/webhook/:project, where :project corresponds to the project name defined in config.json.

5. Check if the Webhook is set up correctly

Modify the project files, push the project, and check the requests in Github and the error logs in logs to resolve any issues.

At this point, we have completed the setup of a service that listens to multiple projects.

Example of deploy.sh

This file needs to be placed in the /var/www/project1 directory. If the command to be executed is relatively simple, it can be replaced with app2.command.

#! /bin/bash
WEB_PATH=/var/www/project1
echo "开始构建"
cd $WEB_PATH
echo "拉取main分支的代码"
git reset --hard origin/main
git clean -f
git pull
git checkout main
echo "检查权限"

# 建议手动进行安装 之后每一次只需要 npm run build即可
npm install 

npm run build
echo "Finished."

When pushing from Github to the server, the deploy.sh in the project(:project) directory will be executed.

This is a Bash script that automatically executes the deployment process of the project in the Web server. The following is a breakdown of the script's functionality:

Note

  • The project needs to be pulled into the server first (some may require permission configuration).
  • It is best to manage it inside the deploy.sh file.

webhookSetting Flow

  1. Settings
  2. Webhooks
  3. Add webhook
  4. Content type ===> json
  5. Payload URL ===> https://yoursite.com/webhook/ + app1
  6. Secret ===> keep empty
  7. SSL verification ===> if fail you can useDisable (not recommended)
  8. Just the push event.
  9. Active yes
  10. click Add webhook

Tool Introduction

This is a Node.js web server script used to listen for webhook requests and trigger deployment scripts. It uses the Express.js framework to handle HTTP requests, the body-parser middleware to parse JSON request bodies, the child_process module to generate shell commands, and the winston module for logging.

The script reads a JSON configuration file that defines the deployment settings for different projects. When a webhook request is received, the script checks the project name in the URL path based on the configuration file. If it finds a match, it generates a shell command to execute the deployment script for that project. If a post-deployment command is defined in the configuration, it also generates a shell command to execute it.

The script also provides a health endpoint at /webhook/health for checking the server status.

Overall, this script provides a simple way to automate the deployment process for multiple projects using webhooks.

Notes

  • This script requires a Node.js environment. If Node.js is not installed, please install it first.
  • Before starting the service, ensure that the configuration information in the config.json file is correct.
  • If the service cannot run normally, check the error information in the terminal output or log file.
  • Restart is required after modifying the configuration file!
  • It is recommended to use pm2 to start the project: pm2 start brewer-webhook --name webhooks.