If you need to run a Node.js script as a service and have it automatically start during system boots, then forever-service is the way to go. Just execute the following commands:

npm install -g forever-service
forever-service install test

By running these commands, you will set up app.js in the current directory as a service using forever. This service will restart itself every time the system reboots. Additionally, it attempts a graceful stop when manually stopped. The script also includes provisions for logrotate.

You can find the Github repository at: https://github.com/zapty/forever-service

Currently, forever-service supports Amazon Linux, CentOS, and Redhat. Support for other Linux distributions, Mac, and Windows is currently in progress.

Please note that I am the developer of forever-service.

I chose to run my Amazon Linux instance on Ubuntu and utilized systemd for setup.

The first step is creating a <servicename>.service file, in my case it was cloudyleela.service.

sudo nano /lib/systemd/system/cloudyleela.service

Within this file, input the following:

[Unit]
Description=cloudy leela
Documentation=http://documentation.domain.com
After=network.target

[Service]
Type=simple
TimeoutSec=0
User=ubuntu
ExecStart=/usr/bin/node /home/ubuntu/server.js
Restart=on-failure

[Install]
WantedBy=multi-user.target

This configuration launches the node application with a set restart policy in case of failure. Amazon instances come without default user passwords.

Once you reload the file from disk, you can start your service. Enabling it will activate the service, automatically starting it during boot-up.

ubuntu@ip-172-31-21-195:~$ sudo systemctl daemon-reload
ubuntu@ip-172-31-21-195:~$ sudo systemctl start cloudyleela
ubuntu@ip-172-31-21-195:~$ sudo systemctl enable cloudyleela
Created symlink /etc/systemd/system/multi-user.target.wants/cloudyleela.service → /lib/systemd/system/cloudyleela.service.
ubuntu@ip-172-31-21-195:~$

Find an excellent systemd tutorial for node.js here.

If running a webserver:

You might encounter issues running the webserver on port 80. An easy workaround is running it on a different port (e.g. 4200) and redirecting traffic from that port to port 80 using this command:

sudo iptables -t nat -A PREROUTING -i -p tcp --dport 80 -j REDIRECT --to-port 4200

Unfortunately, this method is not persistent and needs to be repeated after server restart. A better practice is incorporating this command into the service script:

1. ExecStartPre for adding port forwarding
2. ExecStopPost for removing port forwarding
3. PermissionStartOnly for executing with sudo permissions

Example:

[Service]
...
PermissionsStartOnly=true
ExecStartPre=/sbin/iptables -t nat -A PREROUTING -p tcp --dport 80 -j REDIRECT --to-port 4200
ExecStopPost=/sbin/iptables -t nat -D PREROUTING -p tcp --dport 80 -j REDIRECT --to-port 4200

Don't forget to reload and restart your service:

[ec2-user@ip-172-31-39-212 system]$ sudo systemctl daemon-reload
[ec2-user@ip-172-31-39-212 system]$ sudo systemctl stop cloudyleela
[ec2-user@ip-172-31-39-212 system]$ sudo systemctl start cloudyleela
[ec2-user@ip-172-31-39-212 system]$

For microservices (update on Dec 2020)

While the previous solution offers flexibility, setting it up takes time. With multiple applications, this process becomes cumbersome. PM2 simplifies this by managing all applications under one service.

PM2 can also install its own service during initial setup.

npm install pm2 -g

Adding new services is straightforward:

pm2 start index.js --name "foo"` 

To ensure automatic startup on reboot, save the setup with:

pm2 save

Use pm2 list to view all running node applications.

Additionally, PM2 provides an online dashboard for remote monitoring but some features may require licensing.

If you're looking for a quick fix, you can simply start your app from /etc/rc.local by adding your command there.

For a more elegant solution, consider packaging your application into an rpm file, creating a startup script to place in /etc/rc.d so that you can use chkconfig on your app, and then installing the rpm on your instance.

You may find this or this helpful (or simply search 'creating rpm packages' on Google).

If you need to automate the start and stop of your application, consider creating a script that follows chkconfig's conventions and placing it in /etc/init.d. This will allow you to set your script to start when other services are started.

You can refer to an existing script in /etc/init.d as a template. This article outlines the requirements including:

• An executable script with the necessary shell identifier (e.g., #!/bin/bash)
• A comment in the format # chkconfig: where levels 345 indicate startup priorities, and startprio and stopprio define the order for starting and stopping services respectively. You can use an existing service as a reference for these values.

Once your script is prepared, you can run the following commands:

chkconfig --add yourscript 
chkconfig yourscript on 

This should configure your script correctly. In some distributions, you may need to manually create a symbolic link to /etc/init.d/rc.d, although AWS distros often handle this step automatically when enabling the script.