Category: Continuous Delivery

DevSecOps is everywhere. More and more organizations realize that the traditional separation between DevOps and security is wasteful and dangerous. As a result, they are rushing to integrate security into all their software development processes. 

Adding security to containerized applications, cloud resources and other parts of modern DevOps infrastructure is already well understood. But what about databases? Databases were always—and remain—a critical system component and a prime target for attackers. Here, I’ll review the basics of modern database security and show you how to create a continuous security process for databases in a DevSecOps organization.

What is DevSecOps?

DevSecOps incorporates security practices into the DevOps methodology. DevSecOps promotes a ‘security-as-code’ approach via the partnership between release engineers and security teams. Similar to DevOps, the DevSecOps approach focuses on creating innovative solutions for complex software development processes. Teams complete this within the framework of an agile approach.  

DevSecOps aims to close the gap between security and IT teams while ensuring the speedy and safe delivery of code. Organizations replace the siloed development method with a new approach that promotes the shared responsibility of security tasks during the delivery process. 

In DevSecOps, two supposedly opposite aims—delivery speed and secure code—are united into one effective process. 

Database Security Basics

The term “database security” is an umbrella term used to describe all of the tools created to develop and preserve database availability, integrity and confidentiality. A key responsibility for any DBA is to ensure that the information stored in databases is always secure.

The goal of database security is to protect:

• Database management systems
• Applications that access database information
• Datastores within databases
• Hardware servers
• Virtual database servers, such as cloud-based servers
• Network access points

It is critical to strike a balance between security and usability. While you must prioritize data security, information needs to be accessible to offer value to the organization. The more access you provide to the information retained in a database, the more vulnerable the information is. However, an overly secure database is neither practical nor functional.   

A good way to achieve a balance between accessibility and security is to create meaningful security protocols, create tiered levels of information access and invest in a robust database security and activity monitoring tooling. 

Common Threats and Challenges

Here are some of the most common types of database attacks:

• Insider threats—These involve privileged users misusing their access credentials. An insider could be a negligent employee who unwittingly exposes the database to attack, a malicious insider who abuses the database on purpose or an outside infiltrator who steals credentials to breach the database. The risk of a breach increases as more employees hold privileged access credentials.
• Injection attacks—These involve inserting arbitrary attack strings into the database queries that HTTP headers or web applications serve. These attacks can be SQL or NoSQL. Secure coding practices and regular vulnerability tests should help protect you against injection attacks.
• Database software vulnerability exploits—Database software providers regularly release security patches for vulnerabilities in their software. However, if you don’t apply the patches quickly, you prolong your exposure and run the risk that hackers will exploit vulnerabilities in your database. 
• Backup attacks—If you don’t protect your backup data with sufficient controls, you expose your database to an attack on your backups. 

The following challenges can exacerbate these threats:

• Large volumes of data—Organizations must adequately manage increasingly large volumes of data using scalable tools and practices.
• Sprawling infrastructure sprawl—Organizations are increasingly shifting to hybrid and multi-cloud environments, making their networks more complex and security harder to manage. 
• Evolving regulations—With the global regulatory landscape changing and growing more complex, it is increasingly challenging to maintain compliance.
• Skill shortage—The demand for cybersecurity personnel far outstrips supply, and this shortage is only likely to increase in the future. 

4 Techniques for Continuous Database Security With DevSecOps

A DevSecOps approach calls for continuous database security during all stages of database development and usage. Let’s explore a few techniques for achieving continuous security.

Data Masking

Data masking technology lets you create a structurally similar, anonymized version of an organization’s data. You can use this data for user training or software testing. The aim is to safeguard the real data while using a functional substitute for use cases in which the actual data is not needed.  

In data masking, you keep the format of the data the same. You only change the values of the data. You can alter the data in various ways including character shuffling, encryption and word or character substitution. Whichever method you choose, you have to modify the value in some way that makes reverse engineering or detection impossible. 

Use data masking when presenting data to users (whether inside or outside the organization) who don’t need to access the data as part of their role. You should apply data masking to all sensitive information before displaying it in public forums. 

Use Web Application and Database Firewalls

Protect your database server from security threats using a firewall that stops access to traffic as a default. The only traffic permitted should come from particular web servers or applications that require access to the information. The firewall must also prevent your database from establishing outbound connections unless they are specifically needed. 

Protecting the database using a firewall is not enough. You should also deploy a web application firewall (WAF). That’s because SQL injection attacks, which target web applications, can be employed to delete or exfiltrate data from the database. A database firewall may not stop this from occurring if attacks are initiated from an application that is a permitted source of traffic. However, a web application firewall likely will stop attacks of this nature because it can analyze traffic at the application layer and identify malicious query patterns.

Cloud Backup

Database backup solutions help organizations safeguard their data with backup database copies in the event of user error, physical hardware failure, data corruption or ransomware attacks. Organizations can ensure their information is always accessible by using database backup tools, even if their central database fails or is compromised. 

Cloud backup services allow you to send a database copy to an off-site, secondary location for safekeeping if there is a disaster or failure. A third-party service provider typically hosts the secondary data storage systems and servers. This provider bills the backup customer based on capacity used or storage space, number of users, data transmission bandwidth, number of times the information is accessed or the number of servers. 

Using cloud data backup can help you improve data protection without overwhelming IT staff. This labor-saving advantage could be significant and may even offset some of the added costs connected with cloud backup, including data transmission costs. 

Secure Database User Access

You must strive to limit the number of individuals with access to the database to the extreme minimum. Administrators should only have minimal privileges—those they require to carry out their job and only when they require access. 

For small organizations, this might be difficult to achieve because employees have multiple roles, and there may be only one or two database administrators. However, you should at least manage permissions using roles or groups rather than granting access directly.  

If you have a large organization, you must think about automating access management via access management software. Access management software can give authorized users a provisional password with the privileges they require every time they want access to the database. This software also logs the activities conducted during that time and stops administrators from sharing their passwords. 

It is critical to lock down any form of password sharing—this is a common practice by administrators and database users, but it makes accountability and effective database security nearly impossible. 

In addition, you should follow standard account security procedures: 

• Enforce the use of strong passwords
• Store, encrypt and salt password hashes
• Lock accounts after three or four login retries
• Deactivate accounts when employees leave or switch to different roles

DevSecOps and Database Security

In this article, I explained the basics of DevSecOps and database security, covered the primary threats facing database systems in a modern IT environment and provided four ways you can implement continuous security for databases:

• Data masking—All modern RDBMS systems offer data masking mechanisms that can reduce the impact of compromised accounts.
• Using WAFs and database firewalls—Firewalls provide excellent runtime protection and visibility for application-layer attacks.
• Cloud backup—Backing up databases in the cloud is a continuous process that can enhance availability and reduce the impact of ransomware attacks.
• Secure database user access—Using least-privilege principles can dramatically reduce the threat surface of a database.

I hope this will be useful as you extend your DevSecOps practices to critical database systems in your environment.