Automating User Data Configuration Using the CLI | Engineering Notes

# Day 23: Automating User Data Configuration Using Azure CLI

🎯 Objective

Create an Azure Virtual Machine using Azure CLI and automatically configure it at launch using a User Data / cloud-init script.

In this example, we will:

Create an Ubuntu VM using Azure CLI

Automatically install Nginx

Start and enable the Nginx service

Open port 80 for HTTP traffic

Verify the VM, resources, and Network Security Group (NSG) rules

👉 This eliminates the need to SSH into the VM and manually install Nginx after deployment.

🧠 Concept

Azure CLI allows us to create and configure Azure resources directly from the terminal.

Instead of manually creating a VM through the Azure Portal, we can use:

bash

az vm create

We can also provide a startup script using:

bash

--custom-data nginx.sh

The VM's cloud-init process reads the script during the initial boot and executes the configuration automatically.

Flow

plain text

Azure CLI
    ↓
az vm create
    ↓
Azure VM Created
    ↓
cloud-init reads nginx.sh
    ↓
Install Nginx
    ↓
Start Nginx
    ↓
VM ready to serve HTTP traffic

⚙️ Step 1: Check Azure Account

Before creating resources, verify that Azure CLI is authenticated with the correct account.

bash

az account show

This displays information such as:

Subscription

Tenant

Current Azure account

Subscription state

📦 Step 2: List Resource Groups

Check the available resource groups:

bash

az group list -o table

The -o table option displays the result in an easier-to-read table format.

🔧 Step 3: Store Resource Group Name

Instead of repeatedly typing the resource group name, store it in a shell variable:

bash

rg_name="{resource-group-name}"

Example:

bash

rg_name="devops-rg"

Now we can reference it using:

bash

$rg_name

📝 Step 4: Create the User Data Script

Create a shell script:

bash

vim nginx.sh

Add:

bash

#!/bin/bash

apt update -y
apt install -y nginx

systemctl start nginx
systemctl enable nginx

Save and exit the file.

What This Script Does

plain text

apt update
    ↓
Update package repository

apt install nginx
    ↓
Install Nginx

systemctl start nginx
    ↓
Start Nginx immediately

systemctl enable nginx
    ↓
Automatically start Nginx after reboot

🚀 Step 5: Create the Azure VM

Create the VM and pass the Nginx script as custom data:

bash

az vm create \
  --resource-group "$rg_name" \
  --name devops-vm \
  --image Ubuntu2404 \
  --location eastus \
  --size Standard_B1s \
  --admin-username azureuser \
  --generate-ssh-keys \
  --custom-data nginx.sh \
  --os-disk-size-gb 30 \
  --storage-sku Standard_LRS

Important Options

The important part for automation is:

bash

--custom-data nginx.sh

This allows the VM to configure itself during the initial boot.

📋 Step 6: Check Created Resources

List all resources created inside the resource group:

bash

az resource list \
  --resource-group "$rg_name" \
  -o table

You should see resources related to the VM, such as:

Virtual Machine

Network Interface

Network Security Group

Public IP

Virtual Network

Disk

🌐 Step 7: Open HTTP Port 80

Nginx listens on TCP port 80 by default.

Allow incoming HTTP traffic:

bash

az vm open-port \
  --resource-group "$rg_name" \
  --name devops-vm \
  --port 80

This creates or updates an inbound rule in the VM's Network Security Group (NSG).

Without this rule, Nginx may be running correctly but won't be accessible from the internet.

🔐 Step 8: Verify NSG Rules

Check the Network Security Group rules:

bash

az network nsg rule list \
  --resource-group "$rg_name" \
  --nsg-name devops-vmNSG \
  -o table

Look for an inbound rule allowing:

plain text

TCP → Port 80 → Allow

> The actual NSG name can differ depending on how the VM/network resources were created. Check your resource list if devops-vmNSG is not the generated name.

🔍 Step 9: Check VM Details

Display detailed VM information:

bash

az vm show \
  --resource-group "$rg_name" \
  --name devops-vm \
  --show-details

This provides information such as:

VM status

Public IP address

Private IP address

VM size

Operating system

Resource group

For a cleaner output, you can also use:

bash

az vm show \
  --resource-group "$rg_name" \
  --name devops-vm \
  --show-details \
  -o table

🌍 Step 10: Test Nginx

Get the public IP address:

bash

az vm show \
  --resource-group "$rg_name" \
  --name devops-vm \
  --show-details \
  --query publicIps \
  -o tsv

Then test it:

bash

curl http://<PUBLIC_IP>

Or open it in a browser:

plain text

http://<PUBLIC_IP>

If the configuration was successful, the default Welcome to nginx! page should appear.

🔍 Troubleshooting User Data

If Nginx was not installed correctly, SSH into the VM:

bash

ssh azureuser@<PUBLIC_IP>

Check Nginx:

bash

sudo systemctl status nginx

Check cloud-init status:

bash

cloud-init status

Check the cloud-init output:

bash

sudo cat /var/log/cloud-init-output.log

These logs are especially useful when the nginx.sh script fails during the first boot.

⚠️ Important Notes

#!/bin/bash should be the first line of the script.

Ubuntu uses the apt package manager.

-custom-data passes the initialization script to the VM.

The configuration is processed by cloud-init during VM initialization.

Opening port 80 is required for external HTTP access.

systemctl enable nginx ensures Nginx starts automatically after VM reboot.

A running Nginx service does not automatically mean it is publicly accessible; Azure networking and NSG rules must also allow the traffic.

🧪 Validation Checklist

[ ] Azure CLI authenticated

[ ] Correct resource group selected

[ ] nginx.sh created

[ ] VM created successfully

[ ] Custom data passed to VM

[ ] Nginx installed automatically

[ ] Nginx service running

[ ] Nginx enabled at startup

[ ] Port 80 opened

[ ] NSG inbound rule verified

[ ] Public IP retrieved

[ ] Nginx page accessible from browser

🔗 Reference

Microsoft Azure CLI — VM Commands:

https://learn.microsoft.com/en-us/cli/azure/vm?view=azure-cli-latest

📌 Summary

Azure CLI can automate both infrastructure provisioning and initial server configuration.

Instead of:

plain text

Create VM
→ SSH into VM
→ Update packages
→ Install Nginx
→ Start Nginx
→ Configure networking

we can automate most of the process:

plain text

nginx.sh
        ↓
az vm create --custom-data
        ↓
Azure creates VM
        ↓
cloud-init executes script
        ↓
Nginx installed and started
        ↓
Port 80 opened through NSG
        ↓
Web server ready

This approach makes VM deployment faster, repeatable, and less dependent on manual configuration.