Archive

April 2026

2 posts

I got paged for a DNS outage that was a laptop on hotel wifi

A short post about the incident that taught me to check the physical layer first, and the monitoring rule I added afterwards.

1 min readincident · dns

Backups: the checks that matter are the ones nobody runs

A restoration drill is the only backup that has ever saved anyone. How to run one in an afternoon, and the three failures it will find.

2 min readbackups · infrastructure