Building an Inventory Management Database with the Kubernetes API Server and Custom Resources
Blog post from New Relic
Inventory management is a crucial yet challenging task for New Relic's Container Fabric team, which manages a rapidly growing infrastructure of about 1,000 machines across multiple regions. Initially relying on Mesosphere DC/OS for container orchestration, the team faced increasing complexity in inventory management, prompting them to explore various solutions. They ultimately adopted the Kubernetes API server for a more efficient and scalable inventory management system, leveraging its custom resource definitions to maintain an up-to-date picture of their infrastructure. This new approach reduces operational bottlenecks by eliminating the need for VPN connections and enabling dynamic configuration changes, thereby enhancing their capability to operate at scale. The solution also lays the groundwork for future applications, such as automated monitoring and change orchestration, as the team plans to transition fully from DC/OS to Kubernetes.