Advanced Operations
Multi-cluster overview
For a platform administrator managing several registered clusters, KubeOpera API's federation routes (/api/v1/federation/overview, /health, /cost, /anomalies, /optimizer) aggregate health, cost, and anomaly data across every registered cluster into one rolled-up view, rather than requiring a click through each cluster individually. The platform-admin dashboard's own overview page is built on exactly these routes — the same data is also available as an MCP tool (get_multi_cluster_overview) if you want it outside the dashboard.
GitOps
Every tenant application, and the platform's own backend services, are deployed through FluxCD reading from Git rather than through direct kubectl apply — this is covered in depth in the Setup Guide, and per-application deployment mechanics (how a KubeOperaApp resource becomes a running app inside a tenant's own vCluster) are covered on the App Controller page. If you're trying to understand why a change you made in the dashboard took a few minutes to actually apply, or why a manual kubectl edit on a live resource got silently reverted, both are consequences of this GitOps model — the dashboard's config change reaches Git first, and whatever's in Git is what Flux keeps the cluster converged to.