Implementation Guide for HA Kubernetes in Active/Suspend mode
This white paper details a robust, scalable architecture that leverages the ReadWriteMany (RWX) capabilities of the Lightbits disaggregated, software-defi ned storage platform to establish a resilient Active/Passive High Availability (HA) framework for Kubernetes. By utilizing Lightbits’ native support for clustered NVMe® over TCP (NVMe/TCP) storage, this solution enables seamless volume failover and persistent data access across multiple Kubernetes nodes. The integration focuses on eliminating single points of failure at the storage layer, ensuring that mission-critical stateful applications can automatically recover on standby nodes without data loss or manual intervention. The paper outlines the implementation of Lightbits RWX volumes, the configuration of Kubernetes pod anti-affinity rules for HA orchestration, and the best practices for achieving rapid failover recovery times. This combined solution demonstrates how to achieve enterprise-grade reliability and performance for containerized workloads by leveraging Lightbits’ shared-storage efficiency within a dynamically orchestrated Kubernetes environment.