User data are scripts that use the cloud-config format, designed to be run by the cloud-init process. These scripts are generally used for the initial configuration of a server and run on the first boot. Deploying a server with user data allows you to run arbitrary commands and change several aspects of the server during provisioning. Here are a few examples of what you can do with user data scripts:
  • Change user passwords
  • Server hardening
  • Notify people or systems that a new device has been deployed
  • Install packages and dependencies

​
Compatibility

Here are the operating systems that currently support setting user data and the directive you should use when writing your scripts.
DistroOSDirectiveLanguage
Linux- Ubuntu 16.04 LTS and newer
- CentOS 7.4 and newer
- Rocky Linux
- Debian 10 and newer
- RedHat Enterprise Linux		#cloud-configYAML
Windows- All Windows images# ps1_sysnativePowerShell
Flatcar- Flatcar# !/bin/bashBash

​
File format

​
UNIX-like systems

If you’re using a UNIX-like OS like Linux, your scripts must be written using the YAML file format, which uses whitespace and new lines to delimit lists, associative arrays, and values. Here’s an example of a user data script that uses the {{ PASSWORD }} variable to change the root password and allow for root logins. 
#cloud-config
user: root
chpasswd: { expire: false }
password: { { PASSWORD } }
disable_root: false
ssh_pwauth: true
runcmd:
  - echo "PermitRootLogin yes" >> /etc/ssh/sshd_config
  - systemctl restart ssh
These files are generally fairly intuitive and easy to understand, but there are a few things you need to keep in mind when using them. Here is a good getting started guide that can help.
Always validate your YAML code. Google ‘YAML validator’ if you don’t have a linter on your IDE, but remember never to add sensitive data to these services.

​
Windows systems

For Windows, you’ll have to use the cloud-baseinit convention. This is just a fancy way of saying that your user data will have to be written as if you were using Powershell. Here’s a straightforward example: 
#ps1_sysnative
mkdir c:/test
New-Item c:/test/test.txt
Set-Content c:/test/test.txt 'Hello World'

​
Creating a user data

Before creating a user data script, make sure you’re using the correct directive and programming language for the operating system you’ll be using it with.

​
From scratch

1

Create user data from scratch

Go to the dashboard, navigate to Settings > User data in the sidebar, click Create user data… and choose From scratch.

​
From a template

You can start with user data scripts by choosing one of the provided templates. From package updates to notifying Slack on deploys, templates are available for different use cases.
1

Create user data from template

Go to the dashboard, navigate to Settings > User data in the sidebar, click Create user data… and choose From template.
2

Select and customize template

Choose the template you want and click Use this template. You can then edit its content.

​
Variables

To make development easier, variables are provided that can be used in your user data script so you can access data that is only available during or right after a deployment. Variables are wrapped in double curly brackets ({{ VAR }}). Here are the available variables that can be used by any user data script.
VariableValue
{{ HOSTNAME }}The normalized hostname.
{{ PASSWORD }}Root password of the deployed server.
{{ USER_DISTRO }}Distro username. Examples: ubuntu, centos, Admin.
{{ MAC_ADDRESS }}MAC Address.
{{ PUBLIC_IPV4 }}Public IPv4 address.
{{ SSH_KEYS }}An array of the SSH keys you chose to deploy your server with, if any.
{{ SERIAL }}Windows serial (only available for Windows systems).

​
Examples