Continuous Cluster Discovery in Kubernetes

CybusMQ continuous cluster discovery in Kubernetes prevents split-brain scenarios through periodic polling of the Kubernetes API.

CybusMQ implements continuous cluster discovery for Kubernetes deployments through periodic polling of the Kubernetes API. This prevents split-brain scenarios that could occur with one-time discovery approaches.

How It Works

A discovery process periodically queries the Kubernetes API to discover available broker pods. The broker compares the current cluster state with discovered nodes and automatically establishes or removes connections as needed to maintain the correct cluster topology.

This continuous approach eliminates edge cases where nodes might not know about each other, preventing split-brain scenarios.