Red Hat announced new capabilities and enhancements across its portfolio of open hybrid cloud solutions aimed at accelerating enterprise adoption of edge compute architectures through the Red Hat Edge initiative.
When it comes to food, we all know what's considered "good" and what's "bad".
For fried, battered, buttery or creamy: the category of cupcakes and such ... just say no. When fresh, grilled, steamed or baked: of course, you should eat your veggies and other fine foods ... bon appetite.
We can all understand this simple rule when eating. But for many, when it comes to software development, simple rules and advice from nutritional labels aren't always there for us. Simple rules like: use "good" open source components, not "bad" ones (i.e., vulnerable, outdated, risky licenses) when building and deploying your next app.
Evidence of Gluttony
Beyond the consumption of open source and third-party software components in development, the same guidance holds true for containers used by DevOps teams. Today, over 100,000 free apps are available on Docker Hub, and consumption of those applications into DevOps tool chains and infrastructure often goes unchecked.
The Downside of Efficiency
While sourcing components and containers can accelerate development efforts, many organizations do not apply sufficient scrutiny to assessing their quality or security. The 2016 Software Supply Chain report also indicates that 1 in 15 of the components used in applications contain at least one known security vulnerability. What this means is that not all components and containers are created equal. It also means that blind consumption practices can have both immediate and long-term consequences for the organization consuming the open source software components or containerized applications.
According to the 2017 DevSecOps Community Survey, conducted by Sonatype, 58% of organizations looking to improve the quality of components used in their development practices have implemented an open source governance practice. This is up only slightly from the 2014 survey which showed 57% of organizations with open source policies in place.
For those without a policy, there are no rules in place when it comes to "good" vs. "bad" consumption practices. Having policies in place can help to improve the quality of applications being produced, reduce vulnerabilities to attack, and eliminate use of components with risky license types.
At the time, over 2,700 survey participants of the 2014 survey indicated that more than 1 in 10 organizations (14%) had experienced or suspected a breach in their applications related to an open source software component. Because, that survey took place in the same month that the notorious OpenSSL software component (a.k.a"Heartbleed") vulnerability was announced, I felt like it might have inflated the number of breach related responses. To my surprise, the 2017 survey shows that 20% of organizations (1 in 5) had or suspected a breach related to open source software components in their applications.
In the past three years, as the rate of confirmed or suspected breaches related to vulnerable open source components increased by 50%, the number of organizations having rules in place to govern the use of secure components remained the same.
Across the DevOps community, we are seeing much more attention being paid to security concerns and more organizations finding news ways to automate practices early in the SDLC. Automating the evaluation of open source and third party components against rules designated in open source governance policies is one the approaches being adopted. While not the only answer to DevSecOps requirements, having an automated policy in place to enforce governance and guide remediation can pay off handsomely if it also helps prevent a breach.
When it comes to carrots and cupcakes, the right choices can offer great short- and long-term benefits. In the realm of components and containers, making "good" choices will also benefit your DevSecOps practice.
Derek Weeks is VP and DevOps Advocate at Sonatype.