HTTP, TCP, UDP or Ping: Which Check Type Should You Actually Use?
Each check type proves something different and costs something different. A decision guide, plus the common mistake of monitoring a port when you meant to monitor a service.
Your Health Check Endpoint Is Probably Lying to You
A /health route that always returns 200 will keep your uptime chart green through a total outage. How to build one that actually means something.
Monitoring Dependencies You Do Not Control
When your payment provider goes down, your users blame you. How to detect vendor outages before support does, and what to do with the information.
Monitoring a New Service on Launch Day: A Checklist
The monitoring you set up in the first week decides how the first outage goes. A concrete checklist, ordered by what actually protects you.
How Often Should You Check? Choosing a Monitoring Interval That Makes Sense
One minute is not always better than five. How to derive your interval from detection budget, retry behaviour and what the check actually costs your target.