STAREAST 2022 - Testing in DevOps

Even if your code coverage is 80%, how much confidence do you have in your unit tests? Are they catching the edge cases? Can you fearlessly refactor? Add new features? Perhaps you are writing code that another team will rely on (or worse, you have to rely on!) Or maybe you have to untangle some ugly spaghetti code and want some idea of what it does before you start chopping it up. A mutation testing framework like PIT from pitest.org tests your JUnit tests and shows you what you are and aren’t testing. If your unit tests all pass, mutation tests make changes to a copy of your source code...

In today's modern digital world, where Customer Experience is the key - Application performance – is in demand and a crucial success factor like never before. Traditional transaction-based performance measures are not sufficient to ensure a swift digital customer experience. Performance measuring of customer journeys has become vital as it builds confidence and reliability in the application/system under test. There are multiple challenges in measuring customer journeys' performance - diverse operating systems, devices, and geolocations etc., will account for hundreds of scenarios to be...

Synergies among AI and software testing include both leveraging AI for test automation, and devising methods for testing AI systems. Researchers, software vendors, and practitioners are already using AI to develop a new wave of test automation tools. These tools automatically explore, model, reason, learn, and test software applications. However, as with most AI and ML-based technologies, these tools raise concerns around whether or not we really trust and depend on them. ML-based systems tend to be highly dynamic and adaptable at runtime, thereby requiring extensive testing both offline,...

With the cloud, infrastructure such as networking, security, virtual machines, and much more is entirely software code now. Instead of manually creating cloud environments for applications and making changes, administrators write code to do that work. When changes are needed, that code is enhanced and re-run. Just like application code, infrastructure code needs to be tested. After all, code that introduces defects into existing environments can put developers, testers, and even end-users out of service. That said, most organizations do not adequately test infrastructure code. Anybody in...

Building a mature AppSec program is critical to the success of any product in managing the most vulnerable areas of the application. How can we ensure that the DevSecOps pipeline implemented is working effectively? The paper answers them by highlighting the key measures for every CISO to monitor and track the effectiveness of the AppSec maturity. The effective outcomes were measured by tracking 6 key metrics to validate if DevSecOps is successfully implemented. When done right, DevSecOps goes well beyond “shifting security left” to “shifting security everywhere” ensuring application is...

The role of a tester is bigger and expandable beyond time-boxed sprints. As much as the Test First mindset starts from sprint planning, it is important for testers to be tech-savvy to support the tested code successfully released into production. And behold, close out with Smoke testing. This session focuses on the role of a tester in and beyond sprint:

Two things that the audience can take away from this session:1...