January 2020 Summaries

Chaos Engineering is increasingly being integrated into CI/CD pipelines to enhance the reliability and resilience of software systems by introducing controlled chaos experiments early in the development process. Traditionally conducted in QA or stage environments, chaos testing is now being considered for earlier stages to identify and address potential issues before they manifest in production. The article discusses the typical CI/CD process, which involves developers making incremental code changes that trigger automated build processes, including various testing stages. It highlights the benefits of incorporating chaos experiments, such as automated canary deployments, into CI/CD pipelines to proactively identify system weaknesses and ensure reliability without causing production downtime. The use of tools like the Gremlin API facilitates the integration of chaos experiments into CI/CD pipelines by allowing teams to script and automate chaos tests, ultimately aiming to improve deployment velocity by reducing the need for rollbacks due to unforeseen failures. The article emphasizes starting with small, manageable experiments and gradually increasing complexity as confidence grows, advocating for continuous improvement in pipeline reliability.

In 2019, the field of Chaos Engineering matured significantly, shifting focus from understanding its necessity to effectively implementing it. This evolution was marked by the launch of Gremlin Free, a tool that enhances the capabilities of Netflix's Chaos Monkey by offering features such as broader cloud compatibility and additional attack modes like CPU stress tests. Furthermore, Gremlin introduced Scenarios, which are guided experiments for simulating real-world outages, and expanded Chaos Engineering practices to Kubernetes environments. The year also saw Gremlin achieve SOC II Type 2 certification, emphasizing its commitment to security. The company engaged extensively with its community, hosting Chaos Conf 2019, bootcamps, and talks worldwide, while growing its Chaos Engineering Community Slack to over 4,000 members.

The podcast episode of "Break Things on Purpose" features Kelsey Hightower, Principal Developer Advocate at Google, discussing various aspects of Chaos Engineering, Kubernetes, and the importance of empathy in tech culture. Hightower delves into his career journey from Puppet Labs to CoreOS and Google, highlighting how these experiences shaped his understanding of distributed systems and configuration management. The conversation covers Kubernetes' role in infrastructure management, its evolution, and the significance of declarative configurations. Hightower emphasizes the importance of empathy sessions for developers to understand user challenges and the need for inclusivity and cooperation within the tech community. He also shares insights on serverless computing as a North Star for infrastructure simplicity and reflects on the potential of Kubernetes as a control plane for building robust infrastructure tools. The podcast underscores the value of learning from failures, with parallels drawn to crash tests in the automotive industry, and advocates for a culture of continuous improvement and inclusivity in the tech world.