Running small infrastructure without surprises

Field notes on hosting a handful of servers: monitoring that actually pages, backups you have restored at least once, and networking that survives a provider maintenance window.

Read the notes

Monitoring

Uptime checks, latency budgets and the three alerts that are worth waking up for.

Backups

Nightly snapshots, off-site copies and a quarterly restore drill with a written checklist.

Networking

Firewall defaults, TLS everywhere, and how to keep a status page reachable when the rest is not.

Recent notes


Restore drills: what we learned from a failed one

Alert fatigue and the case for fewer dashboards

Choosing a congestion control for a busy edge

About

A small engineering notebook, updated when there is something worth writing down. No newsletter, no tracking.