Architecture decisions, written down.
Cloud architecture, Kubernetes, platform engineering and production security, written by the engineer doing the work, with the trade-offs stated.
Latest · Kubernetes
Building a Production-Grade Kubernetes MCP Server with Dynatrace Observability
I enabled AI assistants to access Kubernetes directly for network debugging and to correlate with Dynatrace. The solution uses TypeScript, avoids kubectl subprocesses, and self-monitors via OpenTelemetry.
Openshift
OpenShift Pod Stuck in Pending: How to Debug Taints, Tolerations, Affinity and Topology Spread
A Pod stays Pending when no node passes every required rule, and often no single rule is wrong. How the scheduler filters nodes, what taints, affinity and topology spread each guarantee, a worked example where correct rules combine badly, and a six-step workflow to find the blocking rule.
Openshift
OpenShift Networking Deep Dive: OVN-Kubernetes Architecture and Traffic Flow
Follow one request through an OpenShift cluster: how OVN-Kubernetes turns Kubernetes objects into a logical network, how OVS and Geneve move the packets, where Services, NetworkPolicy, the router and egress fit, and how to trace a failing connection layer by layer.
Reference architectures
Sitecore on Azure PaaS: How We Moved a National Tourism Platform Off VM Scale Sets
How a national tourism platform moved Sitecore, .NET and a React front end from VM Scale Sets to Azure PaaS behind Cloudflare. In one 24-hour window, Cloudflare served 99.2% of 55.8 million requests without reaching the origin.