NatKube Podcast

This episode of NatKube Podcast, titled "Networking as Code – TheFirst Light" features a conversation between Álex, a Cloud and DevOps enthusiast, and Sam, a veteran Network Engineer who has embraced automation. The discussion centers on NatKube, a project designed to eliminate the common "fear" engineers feel when making manual configuration changes to core routers.Key Concepts and Technology• The Digital Twin: Sam describes NatKube as a "Digital Twin" for enterprise networks. This system allows engineers to simulate and test network behavior—such as checking if a BGP configuration will fail—before touching any physical equipment.• The Engine (Batfish): NatKube utilizes Batfish, a static analysis engine that parses configurations from vendors like Cisco and Juniper. Instead of deploying "blindly," the engine analyzes the logic of the configurations.• The Interface (Streamlit): To make Batfish more accessible to those who prefer a graphical interface over the command line, NatKube uses Streamlit, a Python-based framework. This provides a visual dashboard where users can upload configurations, verify connectivity, and perform differential analysis to see exactly how a proposed change will impact the network.Technical ImplementationThe project follows a NetDevOps philosophy, integrating several modern development tools:• Docker: The Batfish engine and the application run within containers for consistency and portability.• GitHub & SSH: The code is managed via GitHub, using SSH keys for secure access.• Open Source: The project is released under an MIT license to encourage community collaboration.• Functionality: The application includes features such as global route table viewing, logical traceroute (Reachability testing), node property analysis, and the ability to compare "Reference" vs. "Candidate" snapshots.ConclusionThe episode concludes with the idea that the future of network engineering lies in validating models rather than manually typing commands. By using tools like NatKube to ensure changes are safe before deployment, engineers can achieve the ultimate goal: "sleeping soundly on deployment Fridays".