Skip to content
Multi-Site Kubernetes BGP Anycast Architecture

Multi-Site Kubernetes Cluster with BGP Anycast

The engineering question How do services stay reachable when a site or path fails? I built and operate the routing and cluster integration in my own lab, using BGP, VPNs and Cilium. The result is a working environment for observing routing and service-discovery behaviour, with the architecture described below. It does not establish an availability guarantee; the next useful work is repeatable recovery measurement. The June 2026 inventory counted 13 Kubernetes nodes across two sites: 7 in the Netherlands and 6 in Greece. Earlier references to 7 described the Netherlands cluster alone. A Netherlands snapshot from 28 October 2026 still records 7; a comparable current whole-estate count has not been verified. Historical counts are not live capacity. ...

December 16, 2025 · Updated October 29, 2026 · 3 min · Kyriakos Papadopoulos
GitOps Homelab Architecture

GitOps Homelab: Full-Stack Infrastructure as Code

Everything in the homelab — network devices, VMs, Kubernetes workloads, Docker services, backups — is managed through GitLab CI/CD, Atlantis, and Argo CD. The aim is to make intended configuration reviewable and repeatable; manual recovery remains an operational concern. I own the architecture, integration and operation of this lab. Separating platform provisioning from application delivery makes changes easier to review and repeat. It does not make recovery automatic; configuration drift, provider failures and restoration still need checking. The next step is continued reconciliation and recovery validation, not a customer launch. ...

November 1, 2025 · Updated October 29, 2026 · 3 min · Kyriakos Papadopoulos