5 best practices for strengthening your software supply chain with DevSecOps
Connecting state and local government leaders
Because every application is mission critical, strengthening DevSecOps processes provides needed transparency into an agency’s full portfolio of applications.
As recent software supply chain vulnerabilities like Log4j and the SolarWinds have shown, applications built with pre-existing code can create a huge liability for government agencies if an exploited weakness disrupts operations. Developers need easy-to-use, scalable solutions that can help them secure their software supply chain so attackers cannot remotely inject and execute malware or malicious code.
DevSecOps is an approach to software development in which security is embedded into every step of the process. There are a number of steps federal IT leaders can take to strengthen their DevSecOps processes that can help reduce risk without stifling innovation.
1. Take a holistic approach – security is everyone’s responsibility
The basic premise of DevSecOps is to marry software development operations with security to help ensure that individuals of all abilities and across all technology disciplines within an organization have a better understanding and prioritization of security when it comes to software. You’ll often hear the term “shift left” used to describe DevSecOps because it helps ensure security best practices are built into applications from the start rather than being bolted on haphazardly after a software package is finalized. Taking a holistic approach -- making every member of the process be accountable – ultimately enhances the security of the software supply chain system.
2. Automate secure software distribution
There’s no question software is running the world. It’s becoming increasingly important that organizations have reliable, scalable solutions to help them securely develop, deploy and update their software regularly – hence the reason DevSecOps has become an imperative. The industry is heading toward a future in which updates will be constant. Effectively, software will become “liquid” and “software pipes” will constantly stream updates into systems and devices without human intervention.
Agencies need transparency into the software supply chain. Some government and industry leaders suggest improving security with a software bill of materials, which is akin to the ingredients label on a package of food. An SBOM can provide a chain of custody, detailing all the people who touched a particular piece of software or anything related to the end product – from development to when runtime and distribution all the way to the edge.
3. Test and verify
Initial security tests might give an app a green light, but that approval is only as good as the testing behind it. Weak testing can create a false sense of security about software with lingering vulnerabilities.
Security testing should include red teaming for almost everything – rather than reserving it for critical systems. For example, an Army base’s weakest point might be its payroll system because it may not go through the same stringent security testing as a weapons system. However, that could become an entry point for cybercriminals who then access an entire network and sensitive data.
4. Leverage secure container repositories
The Department of Defense’s Platform One initiative gives developers access to a central repository of tested and validated software components that meet DOD’s container hardening specifications, allowing faster authorization for use. Thanks to the Platform One Iron Bank registry, developers can now rapidly and securely customize mission-critical applications using certified components. Iron Bank-certified vendors are granted a continuous authority to operate so they can push validated code into production quickly, which helps all organizations ‘shift left’ to bake security into every stage of development and to seamlessly deploy updates across geographies, from ground to cloud, to any device throughout the software supply chain with ease and peace of mind.
Another useful resource is the CVE program, which identifies, defines and catalogs cybersecurity vulnerabilities. Public and private tech organizations publish consistent descriptions of vulnerabilities for public consumption. Security researchers, IT professionals and developers worldwide use these records as a point of reference when addressing critical software vulnerabilities.
5. Implement zero trust
Developers can improve supply software chain security by building zero trust into every step. If developers build in the need for constant verification, organizations can maintain security even when parts of a network become disconnected.
Strengthening DevSecOps processes provides needed transparency into an organization’s full portfolio of applications, which is key given that in today’s environment, from a security perspective – every application is “mission critical.”
Looking toward the future
As we witnessed with the recent Log4j attack, deteriorations in the software supply chain can quickly wreak havoc on enterprise and government systems alike – and at a great cost. In 2022, open-source supply chain security will take center stage as organizations rally to democratize security testing and SBOM and increasingly leverage tools to create a full chain of custody for software releases. These practices will help keep systems running smoothly and securely and – most importantly – ensure federal agencies and military organizations are able to meet their missions.
Asaf Karas is chief technology officer of security at JFrog.