Multi-region, multi-server architecture across Australian data centres. No single point of failure.
We built a 10-server high-availability WordPress cluster across three Australian cities. We hardened it with CIS benchmarks, AppArmor, auditd, and a Wazuh SIEM. Then we asked the obvious question: does it actually hold up? So we red-teamed ourselves — running…
Every server starts at roughly 49% compliance against the CIS Ubuntu Linux benchmark. That means half the security controls an industry-standard framework recommends are missing out of the box. We took nine production servers — six web nodes and three…
A ten-server cluster spread across three cities is only as good as your ability to see what is happening inside it. After moving our database layer into a private VPC, we had strong network isolation — but no centralised visibility.…
Our nine-server HA WordPress cluster was fast, redundant, and self-healing — but every database query was crossing the public internet. MariaDB replication between Sydney, Brisbane, and Melbourne. HyperDB read/write splitting. Page view counter increments on every visit. All of it…
How two open-source MCP servers give Claude the ability to provision, configure, harden, and manage BinaryLane cloud infrastructure autonomously — and how that scaled from a single server to a multi-city HA cluster.
Why BinaryLane uses local RAID10 storage instead of shared SANs or Ceph clusters, and why true high availability must be built at the application layer — not hidden inside a storage fabric.
Claims of high availability mean nothing without proof. We deployed a test endpoint across all six web servers, hammered it with 500 concurrent requests, and then killed servers mid-test — stopping nginx, shutting down entire regions, and pulling database replicas…
How we built a multi-city high availability WordPress stack across Sydney, Melbourne and Brisbane on BinaryLane — 6 web servers behind an anycast load balancer, 3 database nodes with HyperDB read/write splitting, live migration between cities, partner server redundancy, and zero single points of failure. All for ~$59/month, built entirely by Claude AI.
Curious how this HA stack was built entirely by AI? Book a time and I'll walk you through the architecture, the tools, and the process.