PM2 deployment

pm2.md

1. Preparing the server

Install git

sudo apt install git-all

Generate Server's SSH public key

ssh-keygen -t rsa -b 4096 -C "deploy"
cat ~/.ssh/id_rsa.pub

Then add this to Deploy keys on Github repo

Add your local SSH public key to server

Insert your local SSH public key to .ssh/authorized_keys on server. If server don't has this file, let create one :

cd
mkdir ~/.ssh
touch ~/.ssh/authorized_keys
chmod 700 ~/.ssh
chmod 600 ~/.ssh/authorized_keys

Install Nodejs and pm2

In local machine and server install nodejs and pm2

Install Nodejs

sudo apt-get install nodejs npm

Install pm2

sudo npm install pm2 -g

CLI autocompletion

pm2 completion install

About pm2, follow the instructions here

1. Deploy project(local machine)

Config Ecosystem file

Generate an ecosystem.config.js template with:

pm2 init

module.exports = {
  apps : [{
    // Name of app
    name: 'API',
    // Script for pm2 run forever
    // If use static website, remove it
    script: 'app.js',

    // Options reference: https://pm2.io/doc/en/runtime/reference/ecosystem-file/

    // Args for script for pm2 run forever
    // If use static website, remove it
    args: 'one two',
    // Current directory on server
    cwd: "/var/www/production/current",
    // Config out file for web errors
    error_file: "/var/www/production/logs/web.err.log",
    // Config out file for web logs
    out_file: "/var/www/production/logs/web.out.log",
    // Number of instances to be started in cluster mode
    instances: 1,
    // Enable or disable auto restart after process failure
    autorestart: true,
    // Enable or disable the watch mode
    watch: false,
    // Restart the app if an amount of memory is exceeded (format: /0-9?/ K for KB, ‘M’ for MB, ‘G’ for GB, default to B)
    max_memory_restart: '1G',
    env: {
      NODE_ENV: 'development'
    },
    // ^env_\S*$ => Specify environment variables to be injected when using –env
    env_production: {
      NODE_ENV: 'production'
    }
  }],

  deploy : {
    production : {
      // SSH user
      user : 'node',
      // SSH host
      host : '212.83.163.175',
      // GIT remote/branch
      ref  : 'origin/master',
      // GIT remote
      repo : 'git@github.com:repo.git',
      // Fetch all branches or fast
      fetch: 'all'
      // Path in the server
      path : '/var/www/production',
      // Command run after pull source code
      'post-deploy' : 'npm install && pm2 reload ecosystem.config.js --env production'
    }
  }
};

Edit ecosystem.config.js for your project.

You can reference my config files

• For Vue static app with multiple environments on this link
• OR for nuxt app on this link

Setup and Deploy

Setup

Make your first deploy and populate the distant path with:

pm2 deploy production setup

Deploy

pm2 deploy production

Config Nginx on server

Install NGINX

sudo apt update
sudo apt install nginx

Then you will need to add some configurations for our app. Create a config file at /etc/nginx/conf.d/my_site.conf:

server {
  listen 80;
  server_name 212.83.163.175;
  root /var/www/production/current/dist;
        access_log      /var/www/production/logs/nginx_access.log;
        error_log       /var/www/production/logs/nginx_error.log;

  try_files $uri $uri/ /index.html;
}

"/var/www/production/current" is your cwd in above ecosystem config This is simple config for nginx.

You can reference my config for Vue static app or Nuxt on this link.

Restart Nginx

sudo service nginx restart

Create symbolic link (optional)

You can create symbolic link for cwd Take a look at the nginx config above, the doc root points to /var/www/production/current/dist, if you wanna shortcut this to /home/deploy/www directory, you need to create a symbolic link from the build directory to /home/deploy/www. The build will be placed at /var/www/production/current/dist, so run the command below to create a symbolic link

ln -s /var/www/production/current/dist /home/deploy/www

and replace /home/deploy/www to root key in nginx config file