Skip to main content

% 16k.es

Getting started with Vagrant

/en/getting-started-with-vagrant/ map[name:16k.es]
Table of Contents

Getting started with Vagrant. Very first steps to install and run a vagrant image in a Mac

vagrant

Vagrant logo

# macOS install

Install Vagrant from the package in the official download page. There is a homebrew package but Hashicorp discourages installing from package managers.

Once installed it is added to the PATH variable:

% vagrant version
Installed Version: 2.2.9
Latest Version: 2.2.9

You're running an up-to-date version of Vagrant!

# Documentation

# Getting started

Vagrant can work with several providers, but VirtualBox is builtin. So make sure VirtualBox is installed as well.

## Up and running

This works out of the box:

$ vagrant init hashicorp/bionic64
$ vagrant up

VM can be terminated with:

$ vagrant destroy

## Project setup

First step is creating a Vagrantfile. This is:

  1. the root directory of the project
  2. describes the machines and resources you need

There is a command to setup the root directory of a project:

$ mkdir vagrant_getting_started
$ cd vagrant_getting_started
$ vagrant init hashicorp/bionic64

You can also run vagrant init in an existing directory to initialize it.

## Installing a box

You can add boxes with vagrant box add.

$ vagrant box add hashicorp/bionic64

This will download the box named “hashicorp/bionic64” from HashiCorp’s Vagrant Cloud box catalog, a place where you can find and host boxes.

## Using a box

Once you add a box, you have to configure it. So open the Vagrantfile and:

Vagrant.configure("2") do |config|
  config.vm.box = "hashicorp/bionic64"
end

You can specify a specific version:

Vagrant.configure("2") do |config|
  config.vm.box = "hashicorp/bionic64"
  config.vm.box_version = "1.1.0"
end

or an URL

Vagrant.configure("2") do |config|
  config.vm.box = "hashicorp/bionic64"
  config.vm.box_url = "https://vagrantcloud.com/hashicorp/bionic64"
end

## Up and ssh

You can start the environment and ssh to it:

$ vagrant up
$ vagrant ssh

vagrant destroy will stop and remove the box from Virtual box, but will not delete the files.

The latter is done by vagrant box remove.

## Synced folders

Vagrant shares the project directory with the /vagrant directory in the guest VM. So, after running:

$ vagrant up
$ vagrant ssh

you are inside the VM.Then you can see your project directory in /vagrant:

vagrant@vagrant:~$ pwd
/home/vagrant
vagrant@vagrant:~$ cd /vagrant/
vagrant@vagrant:/vagrant$ ls
Vagrantfile
vagrant@vagrant:/vagrant$ ls -la
total 8
drwxr-xr-x  1 vagrant vagrant  128 Jun 13 16:06 .
drwxr-xr-x 24 root    root    4096 Jun 13 16:44 ..
drwxr-xr-x  1 vagrant vagrant  128 Jun 13 16:06 .vagrant
-rw-r--r--  1 vagrant vagrant 3024 Jun 13 16:05 Vagrantfile

## Provisioning

There is an automatic provisioning feature builtin. For instance, you can create a script that will run after vagrant up to install software:

#!/usr/bin/env bash
 
apt-get update
apt-get install -y apache2
if ! [ -L /var/www ]; then
  rm -rf /var/www
  ln -fs /vagrant /var/www
fi

Then configure the Vagrantfile:

Vagrant.configure("2") do |config|
  config.vm.box = "hashicorp/bionic64"
  config.vm.provision :shell, path: "bootstrap.sh"
end

And create an index.html file to verify that it works:

<!DOCTYPE html>
<html>
  <body>
    <h1>Getting started with Vagrant!</h1>
  </body>
</html>

Everything will run when you run vagrant up.

If the environment is already running you can use vagrant reload --provision.

## Networking

Port forwarding:

Vagrant.configure("2") do |config|
  config.vm.box = "hashicorp/bionic64"
  config.vm.provision :shell, path: "bootstrap.sh"
  config.vm.network :forwarded_port, guest: 80, host: 4567
end

This will result in http://127.0.0.1:4567 forwarded to the web server in the guest VM.

See official documentation about networking.

## Teardown

You can suspend, halt, or destroy the guest machine:

  • vagrant suspend
  • vagrant halt
  • vagrant destroy

Finally you can delete from disk:

  • vagrant box remove NAME

## Rebuild

vagrant up will recreate the environment.

## Providers

You can use any other provider without changing the Vagrantfile:

$ vagrant up --provider=vmware_fusion

# References