Securing Microsoft SQL Server

Here is a quick check list for System Engineers and Data Base Administrators to lock down MS SQL Server and secure the database server:

Communication: If possible and if budget permits, do not expose the SQL Server directly to the Internet. Allow the communication from a web server to the SQL database server over an internal IP address.

Strong Passwords: Ensure that the server uses a strong password for the “sa” account. Use a combination of letters, numbers and symbols. This make it difficult for Brute Force hacks to guess the password.

Authentication: Keep this to Windows Authentication if possible and avoid setting SQL to run under Mixed Mode. Windows Authentication will require SQL server to check the Active Directory and moreover the logins will not be stored in the SQL server.

Service Packs and Updates: Make sure that the SQL Server is always patched with the latest Service Pack and Security Updates. This ensures that the latest security vulnerabilities are addressed and blocked.

Service Accounts: Run both MSSQLSERVER and SQLSERVERAGENT under the Domain User permissions. Do not run them under any Administrator account. This ensures very less damage in case the user account or the server is compromised.

Block Ports: If there is no need to expose the MS SQL Server to the Internet, block port 1433 and 1434 at the firewall. This means that the SQL Server will not be accessible from the Internet, thus making it hard for outside attackers to reach to the server. It also prevents worms and viruses.

Backups: Encrypt and compress the backup files. Put a password on the backups and store them to a safe location. Do not keep them on the server.

2 comments
  1. Loving your work, but I would also change the default ports in config manager / SAC, and then possibly block those assigned ports from internet access if required.
    Andrew (deepfat on twitter)

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

You May Also Like