Create Your First Azure SQL Database

You've built applications, configured identity, and deployed containers — but where does your data actually live? Without a managed, scalable, and secure database, your Azure workloads are just stateless shells. This lesson walks you through creating your first Azure SQL Database, from provisioning to connecting with a real client, so you can start persisting data the way production systems do.

The problem this lesson solves

Every serious application needs a place to store data — user profiles, orders, logs, or telemetry. You could host a database on a VM, but then you are responsible for backups, patching, high availability, and security updates. That's a distraction from your actual product.

Azure SQL Database solves this by offering a fully managed relational database service. You don't manage the underlying operating system or infrastructure. Microsoft handles patching, backups, and replication. That's a massive win for developers and DevOps engineers who want to focus on application logic.

The pain point: many developers waste hours trying to connect to a database that doesn't exist yet, or they provision the wrong tier and overpay. This lesson eliminates that confusion by giving you a repeatable path to a working database and a verified connection.

Core concept / mental model

Think of Azure SQL Database as a car with a driver — you control the destination and the comfort, but a professional handles the engine mechanics. You choose the performance tier, configure networking, and manage your schema. Azure handles updates, patches, and automatic backups.

Here's the mental model:

• Logical server — a management container that holds your databases, similar to an on-premises SQL Server instance. It has a DNS name, firewall rules, and admin credentials.
• Single database — the actual database inside the logical server, with its own performance characteristics and storage.
• Connection string — the address and credentials your application uses to talk to the database.

In simpler terms: the logical server is the building, the database is an apartment inside, and the connection string is the key that lets your app in.

Pro tip: Don't confuse the logical server with a VM. You never RDP into it. You interact through the Azure portal, CLI, or direct database connections.

How it works step by step

The process of creating your first Azure SQL Database follows a logical order. Get these steps right and you'll avoid most common pitfalls.

Step 1: Create a resource group

A resource group is a logical container for related Azure resources. Put your server, database, and any other related resources in the same group for easier management and cost tracking.

Step 2: Provision the logical server

When you create a database, you'll also create a logical server (or reuse an existing one). The server needs:

• A globally unique DNS name
• Admin credentials (login and password)
• A region (choose one close to your users)

Step 3: Create the database

Choose your performance tier — Basic for learning, Standard for low-cost production, Premium for high throughput. Name your database and select a collation (default is fine for most cases).

Step 4: Configure networking

By default, Azure SQL Database blocks all external traffic. You must configure a firewall rule to allow your client IP address to connect. For production, use a Virtual Network rule or a private endpoint.

Step 5: Get the connection string

Azure generates connection strings for common clients (ADO.NET, JDBC, Python, PHP). Copy the one for your language and use it in your app.

Step 6: Connect from your code

Use your language's SQL client library to connect and run queries. We'll do this hands-on in the next section.

Pro tip: Always set a server-level firewall rule from your IP, not 0.0.0.0/0. The latter exposes your database to the entire internet.

Hands-on walkthrough

Let's create your first Azure SQL Database and connect to it. You can use the Azure portal or the CLI — the CLI is more reproducible, so we'll use it here.

Prerequisites

• An Azure subscription (free tier works)
• Azure CLI installed and logged in

1. Log in and set your subscription

az login
az account show --output table

If you have multiple subscriptions, select one:

az account set --subscription "your-subscription-id"

2. Create a resource group

az group create --name rg-sqldb-tutorial --location eastus

Expected output:

{
  "id": "/subscriptions/.../resourceGroups/rg-sqldb-tutorial",
  "location": "eastus",
  "name": "rg-sqldb-tutorial",
  "provisioningState": "Succeeded"
}

3. Create the logical server

az sql server create \
  --name my-sqldb-server-2025 \
  --resource-group rg-sqldb-tutorial \
  --location eastus \
  --admin-user sqladmin \
  --admin-password 'YourStrongPass!1'

Security note: Store the admin password in Azure Key Vault or use Azure AD authentication in production. Never hardcode passwords in code.

Expected output: JSON with fullyQualifiedDomainName like my-sqldb-server-2025.database.windows.net.

4. Create the database

az sql db create \
  --resource-group rg-sqldb-tutorial \
  --server my-sqldb-server-2025 \
  --name TodoDB \
  --edition Basic \
  --capacity 5

The Basic tier is perfect for learning and light workloads. Later you can scale up.

Expected output: JSON with status: Online after a short wait.

5. Configure the firewall

Get your public IP and allow it:

MY_IP=$(curl -s ifconfig.me)
az sql server firewall-rule create \
  --resource-group rg-sqldb-tutorial \
  --server my-sqldb-server-2025 \
  --name AllowMyIP \
  --start-ip-address $MY_IP \
  --end-ip-address $MY_IP

6. Connect with Python

Install the pyodbc driver (or use pymssql). Here's how to connect and run a query:

import pyodbc

server = 'my-sqldb-server-2025.database.windows.net'
database = 'TodoDB'
username = 'sqladmin'
password = 'YourStrongPass!1'
driver = '{ODBC Driver 18 for SQL Server}'

# Connection string
conn_str = f'DRIVER={driver};SERVER={server};DATABASE={database};UID={username};PWD={password};Encrypt=yes;TrustServerCertificate=no;Connection Timeout=30'

conn = pyodbc.connect(conn_str)
cursor = conn.cursor()

# Create a table
cursor.execute('''
CREATE TABLE Todos (
    Id INT PRIMARY KEY IDENTITY,
    Title NVARCHAR(100) NOT NULL,
    IsCompleted BIT NOT NULL DEFAULT 0
)
''')
conn.commit()

# Insert a row
cursor.execute("INSERT INTO Todos (Title) VALUES ('Learn Azure SQL')")
conn.commit()

# Query
cursor.execute('SELECT Id, Title, IsCompleted FROM Todos')
for row in cursor.fetchall():
    print(row)

conn.close()

Expected output:

(1, 'Learn Azure SQL', False)

This proves your database is live and accessible.

7. Query via the CLI (optional)

You can also run SQL directly using sqlcmd:

sqlcmd -S my-sqldb-server-2025.database.windows.net -U sqladmin -P 'YourStrongPass!1' -d TodoDB -Q "SELECT @@VERSION"

Compare options / when to choose what

For your first database, single database on Basic tier is the right choice. It's cheap, fully managed, and enough for learning. Move to Managed Instance only if you need features like SQL Agent or cross-database queries.

Pro tip: Use the serverless compute tier if your workload is intermittent — you pay per second of usage, not for idle time.

Troubleshooting & edge cases

"Cannot connect to the server"

This is the most common issue. Check:

• Firewall rules — your client IP might not be allowed. Re-run the firewall step or use Azure portal's "Add client IP" button.
• TLS/Encryption — modern drivers default to Encrypt=yes. Make sure you use the latest ODBC driver.
• Credentials — you must use the server admin login, not your subscription credentials.

"Login failed for user 'sqladmin'"

Double-check the username and password. Also, if you enabled Azure AD-only authentication, the SQL admin login won't work.

Database creation fails

Your region might have capacity issues or the database name might conflict. Try a different region or a more unique name.

Performance is slow

The Basic tier has limited IOPS. If your queries time out, consider scaling to a higher tier or optimizing your queries with indexes.

Edge case: connecting from Azure Functions

If your code runs inside Azure (like an Azure Function), you don't need to allow your IP. Instead, enable the "Allow Azure services" firewall rule or better, use a private endpoint.

What you learned & what's next

You now know how to create your first Azure SQL Database from scratch, connect to it using Python, and avoid the most common pitfalls. You understand the mental model of logical servers, databases, and connection strings, and you can compare Azure SQL options to pick the right one for your needs.

In the next lesson, we'll explore securing your database with Azure Active Directory authentication and managed identities — crucial for production deployments. You'll then move on to integrating your database with other Azure services like App Service and Functions.

Key takeaways from this lesson:

• Azure SQL Database is fully managed — no patching, backups, or OS maintenance.
• A logical server contains one or more databases; the server is the management boundary.
• The Basic tier is ideal for learning and small workloads.
• Always configure firewall rules carefully — never expose your database to the whole internet.
• Use the connection string generated by Azure with your language's SQL client.
• The CLI makes provisioning reproducible — script it for production.

You're now ready to build data-driven applications on Azure. Go create your database again, this time from memory, and then explore the next lesson on securing it.