How to Install Microsoft SQL Server on Docker

Introduction

Microsoft SQL Server is one of the most popular database systems in the world — used by businesses of all sizes to store and manage data. The great news is you don't need a Windows server or an expensive license to run it. You can run it for free on your Linux server using Docker in just a few minutes.

This guide walks you through every single step, with plain English explanations of every command.

What You'll Need

• A Linux server with Docker already installed (follow our Docker installation guide if you haven't done this yet)
• SSH access to your server
• Basic familiarity with the terminal

Key Terms Explained

Term	What it means
SQL Server	Microsoft's database management system — used to store, retrieve, and manage data
MSSQL	Short for Microsoft SQL Server
Database	An organized collection of data stored in tables (like spreadsheets)
Table	A structured set of data organized into rows and columns — like an Excel sheet
SA	"System Administrator" — the built-in admin account in SQL Server with full control
Port 1433	The default network port that SQL Server listens on for connections
sqlcmd	Microsoft's official command line tool for running SQL commands
EULA	"End User License Agreement" — the terms of use you must accept to use the software
PID	"Product ID" — tells SQL Server which edition to run (Express, Developer, etc.)
Docker Volume	A storage area managed by Docker that persists your data even if the container is deleted
chown	"Change Owner" — a Linux command to change who owns a file or folder

SQL Server Editions (Free Options)

You don't need to pay anything to follow this guide. Here are the free editions:

Edition	Best For	Limits
Express	Small apps, learning	10GB database size limit
Developer	Development and testing	Full features, not for production use

For this guide we'll use Express. Change it to Developer if you need full features for testing.

Step 1: Connect to Your Server via SSH

Open your terminal and connect to your server:

ssh root@your-server-ip

Breaking this down:

• ssh — Secure Shell, opens a secure remote connection to your server
• root — the admin username
• your-server-ip — replace with your actual server IP address

Type your password when prompted and press Enter. You won't see the password as you type — that's normal.

Step 2: Create a Folder for SQL Server

Keep things organized by creating a dedicated folder:

mkdir ~/mssql && cd ~/mssql

• mkdir — "make directory" — creates a new folder
• ~/mssql — creates it in your home directory (~ = home)
• && — runs the next command only if the first succeeds
• cd ~/mssql — moves you into the new folder

Step 3: Fix Folder Permissions

SQL Server 2022 runs as a non-root user inside the container (user ID 10001). We need to create the data folder and give it the correct ownership so SQL Server can write to it:

mkdir -p ./mssql-data
sudo chown -R 10001:10001 ./mssql-data

Breaking this down:

• mkdir -p ./mssql-data — creates the data folder (-p means don't error if it already exists)
• sudo — run as administrator
• chown — "change owner" — changes the ownership of a file or folder
• -R — applies the change recursively to everything inside the folder
• 10001:10001 — the user ID and group ID that SQL Server runs as inside Docker
• ./mssql-data — the folder we're changing ownership of

Why is this needed? SQL Server 2022 changed its default behavior to run as a non-root user for security reasons. If the data folder is owned by root, SQL Server can't write to it and will crash with a "Permission Denied" error.

Step 4: Create the Docker Compose File

nano docker-compose.yml

nano opens a terminal text editor. Type or paste the following:

services:
  mssql:
    image: mcr.microsoft.com/mssql/server:2022-latest
    restart: unless-stopped
    user: root
    ports:
      - "1433:1433"
    environment:
      - ACCEPT_EULA=Y
      - SA_PASSWORD=YourStrongPassword123!
      - MSSQL_PID=Express
    volumes:
      - mssql-data:/var/opt/mssql

volumes:
  mssql-data:

Breaking this down:

• image: mcr.microsoft.com/mssql/server:2022-latest — the official Microsoft SQL Server 2022 Docker image from Microsoft's container registry
• restart: unless-stopped — automatically restarts if the server reboots (unless you manually stop it)
• user: root — runs the container as root to avoid permission issues on startup
• ports: "1433:1433" — maps port 1433 on your server to port 1433 inside the container (the default SQL Server port)
• ACCEPT_EULA=Y — you must set this to Y to accept Microsoft's license agreement, otherwise SQL Server will refuse to start
• SA_PASSWORD — the password for the sa (System Administrator) account. Change this to your own strong password!
• MSSQL_PID=Express — sets the edition to Express (free). Valid values: Express, Developer, Standard, Enterprise
• volumes: mssql-data — uses a Docker-managed volume to store your databases permanently

Password Requirements: SQL Server enforces strong passwords. Your password must have at least 8 characters and include all of: an uppercase letter, a lowercase letter, a number, and a special character (like !, @, #). SQL Server will refuse to start if the password is too weak.

Save and exit nano: Press Ctrl + X, then Y to confirm, then Enter to keep the filename.

Step 5: Start SQL Server

docker compose up -d

• docker compose up — starts all containers defined in your compose file
• -d — "detached mode" — runs in the background so your terminal stays free

SQL Server takes about 30 seconds to fully initialize. Give it a moment before the next step.

Step 6: Verify It's Running

Check the container is up:

docker ps

You should see a container called mssql-mssql-1 with status Up.

Then check the logs to confirm SQL Server started successfully:

docker compose logs

Look for this line:

SQL Server is now ready for client connections.

If you see it — SQL Server is running!

Step 7: Open the Firewall Port

sudo ufw allow 1433

• ufw — "Uncomplicated Firewall" — Linux's built-in firewall tool
• allow 1433 — opens port 1433 so external tools can connect to SQL Server

Cloud server users (AWS, DigitalOcean, Linode, etc.): You also need to open port 1433 in your cloud provider's dashboard under "Security Groups" or "Firewall" settings. This is a separate firewall from the Linux one — both need to be configured.

Step 8: Test the Connection from Inside the Container

Let's verify SQL Server is working by connecting to it directly:

docker exec -it mssql-mssql-1 /opt/mssql-tools18/bin/sqlcmd -S localhost -U sa -P 'YourStrongPassword123!' -No

Breaking this down:

• docker exec — runs a command inside a running container
• -it — "interactive terminal" — lets you type and see output
• mssql-mssql-1 — the name of your container (verify with docker ps)
• /opt/mssql-tools18/bin/sqlcmd — the path to Microsoft's SQL command line tool inside the container
• -S localhost — connect to the server at localhost (inside the container)
• -U sa — username is sa (System Administrator)
• -P 'YourStrongPassword123!' — your password (keep the single quotes)
• -No — disables SSL certificate validation, required for SQL Server 2022

You'll see a 1> prompt — you're connected! Run a quick test:

SELECT @@VERSION
GO

• SELECT @@VERSION — displays the SQL Server version information
• GO — sends the command to SQL Server for execution (always needed after SQL statements in sqlcmd)

Type exit to disconnect.

Step 9: Create Your First Database

Still inside sqlcmd (or reconnect using the command from Step 8), run these SQL commands:

-- Create a new database
CREATE DATABASE myapp;
GO

-- Switch to using it
USE myapp;
GO

-- Create a users table
CREATE TABLE users (
    id INT PRIMARY KEY IDENTITY,
    name NVARCHAR(100),
    email NVARCHAR(100),
    created_at DATETIME DEFAULT GETDATE()
);
GO

-- Insert a test record
INSERT INTO users (name, email) VALUES ('John Doe', 'john@example.com');
GO

-- Read the data back
SELECT * FROM users;
GO

Breaking this down:

• CREATE DATABASE — creates a new database
• USE myapp — switches to that database so following commands run inside it
• CREATE TABLE — creates a new table with defined columns
• INT PRIMARY KEY IDENTITY — an auto-incrementing number that uniquely identifies each row
• NVARCHAR(100) — a text field that supports Unicode characters, up to 100 characters
• DATETIME DEFAULT GETDATE() — a date/time field that automatically saves the current time
• INSERT INTO — adds a new row of data
• SELECT * — retrieves all rows and columns from the table

Step 10: Connect Using a GUI Tool (Recommended)

The command line is useful but a visual tool makes working with databases much easier. Azure Data Studio is Microsoft's free, cross-platform database GUI.

Download Azure Data Studio: azure.microsoft.com/products/data-studio

Connection details:

• Server: your-server-public-ip
• Port: 1433
• Authentication type: SQL Login
• Username: sa
• Password: YourStrongPassword123!
• Trust server certificate: Enable this

Alternatively, DBeaver is another free GUI that works great with SQL Server and many other databases: dbeaver.io

Troubleshooting Common Issues

Error: Permission Denied / .system could not be created

This is a folder permissions issue. Run:

docker compose down
mkdir -p ./mssql-data
sudo chown -R 10001:10001 ./mssql-data
docker compose up -d

Error: The provided PID is invalid

Your MSSQL_PID value is wrong. Make sure it's exactly Express or Developer — no extra characters, correct capitalization.

Error: Password validation failed

Your SA_PASSWORD doesn't meet the requirements. It must include uppercase, lowercase, a number, and a special character.

Can't connect from outside the server

• Check Linux firewall: sudo ufw status — port 1433 should show as ALLOW
• Check your cloud provider's security group has port 1433 open
• Make sure the container is running: docker ps

Container name not found

Run docker ps and copy the exact container name shown — it may differ slightly.

Connection refused from GUI tool

Make sure "Trust server certificate" is enabled in your connection settings.

Useful Commands Reference

# Start SQL Server
cd ~/mssql && docker compose up -d

# Stop SQL Server
cd ~/mssql && docker compose down

# Restart SQL Server
cd ~/mssql && docker compose restart

# View logs
cd ~/mssql && docker compose logs

# Connect via terminal
docker exec -it mssql-mssql-1 /opt/mssql-tools18/bin/sqlcmd -S localhost -U sa -P 'YourPassword' -No

# Check running containers
docker ps

# Check firewall status
sudo ufw status

What's Next?

Now that SQL Server is running, here are some things you can explore:

• Connect your app — link SQL Server to a Node.js, .NET, Python, or PHP application
• Create more databases — one database per project is a common practice
• Set up regular backups — use SQL Server's built-in backup commands to protect your data
• Create additional users — avoid using the sa account for your apps; create a dedicated user with limited permissions instead
• Explore SQL Server Management Studio (SSMS) — Microsoft's full-featured Windows GUI for managing SQL Server

Summary

Here's the full process in a nutshell:

1. Connect via SSH: ssh root@your-ip
2. Create the project folder: mkdir ~/mssql && cd ~/mssql
3. Fix folder permissions with chown
4. Write your docker-compose.yml with your SA password and edition
5. Launch: docker compose up -d
6. Confirm it's ready in the logs
7. Open port 1433 in the firewall
8. Connect with sqlcmd and run a test query
9. Create a database and table
10. Connect using Azure Data Studio or DBeaver

You now have Microsoft SQL Server running on your Linux server inside Docker — completely free and ready to use!