Common Pitfalls in GitOps Workflows

GitOps describes the automated synchronisation of the state of cloud-native applications against a version-controlled reference state. As such, it takes a Git repository as the single source of truth, so that development and operations teams are working with the same set of tools. When implementing a GitOps workflow, however, you are met with several questions like:

How should I structure my repository for managing multiple instances of my applications?
How can I manage privileges for different teams?
How do I deal with sensitive data such as passwords?
In this talk, I will show that there is no one-size-fits-all answer to these questions. Instead, I will discuss multiple scenarios, so that you can choose which solution works best with your specific workflow.

Pascal Fries is an IT Consultant at ATIX AG in Munich, Germany. He supports customers in the automation of container platforms such as Docker and Kubernetes, including the management of these platforms themselves, as well as the development and delivery of applications to them. In particular, he specialises in DevOps practices and cloud native CI/CD. He also offers training, workshops, and webinars on Kubernetes and GitOps.

This post is also available in: German