Vagrant Learning

Using Vagrant is similar to using other CLI-based tools, such as git. In your shell (I'm using PowerShell on the fancy Windows Terminal; all these commands will also work with bash), make a new directory for a Vagrant project:

# 
mkdir vagrant_learning
cd vagrant_learning

If you cloned this directory, you won't need to do this. Just cd into this directory.

Now, we'll use vagrant to initialize our new directory with a box (UNIX machine image).

# Init with 
vagrant init hashicorp/precise64

You can choose other boxes to init your project with from the Vagrant Cloud.

Run vagrant up to run your machine!

Working with your (virtual) machine

Now, you should be able to access your system by typing.

vagrant ssh

The files (from your host machine) should be available in the virtual machine's /vagrant/ directory. If you downloaded this git repository, you'll find a directory ruby-scripts/ there. Try running ruby foo.rb. Wow! What a convoluted "Hello, World!" tutorial!

If you make changes on your host machine, you'll need to use vagrant reload to see those changes on your virtual machine (if you're SSH'd into the virtual machine, you can press Ctrl + d to exit).

Adding more software, poking around

To install/try out things within your machine, you can use the logged in vagrant account which has sudo privileges.

Just do stuff like sudo apt-get install vim or whatever to get the tools you need.

Halting, Suspending, Destroying

Some other commands you may find useful are vagrant halt, vagrant suspend / vagrant resume, and vagrant destroy.

Doing vagrant suspend will save the state of your VM; running vagrant resume allows you to pick up right where you left off.

vagrant halt stops the machine; you'll have to use vagrant up to start the (same) machine up again.

Finally vagrant destroy will destroy the current VM. If you were trying out stuff in SSH, using sudo to install new software, and so on... these changes will all be lost with your machine. Running vagrant up will create a new machine to replace the one you destroyed.

More info

Run vagrant --help to see what other commands are available.

Provisioning

Provisioning is how you set up your system after loading some box ("image"), such as the official Ubuntu image provided by HashiCorp (creators of Vagrant).

At the bottom of the Vagrantfile you will see a shell script for doing stuff like installing new software. I followed the example setup from here.

This code makes use of a Ruby Heredoc to "embed" a shell script within a Ruby file (the Vagrant file uses Ruby's syntax).

config.vm.provision "shell", inline: <<-SHELL, privileged: true

echo "--> I'm a script from the Vagrantfile."

# Try installing something
apt-get update

# Tools for working within the virtual machine
apt-get -y install vim tmux 
SHELL

Spinning up an NGINX web server

NGINX is a small, quick web server. I looked at this guide on how to quickly set up an nginx server.

In the default Vagrantfile, uncomment out the line that looks like this.

config.vm.network "forwarded_port", guest: 80, host: 8080, host_ip: "127.0.0.1"

This will allow port 80 on the guest system (the VM you're running with Vagrant) to be accessible at localhost:8080 from your host machine.

Now, go down to the provisioning Heredoc string we just made. Add in these lines.

apt-get install -y nginx
service nginx start

Leaving all the default options, the /usr/share/nginx/www directory will serve up static assets (like HTML files, CSS, etc.).

If you try editing /usr/share/nginx/www/index.html you will see your changes reflected at http://localhost:8080 on your host machine.

More Resources

Aside from the official docs, I looked at

Getting Started with Vagrant on Windows (Zack Wallace, 01/12/2017)

captainalan/vagrant-learning