Building observability from scratch, three times over 🏗️

Starting with Nagios on a single box is fine until your infrastructure starts growing. Rebuilding an observability stack three times taught us exactly where the friction points are when scaling to thousands of VMs.

Here is the breakdown of that evolution:

• Move from isolated tools to a unified stack for health checks and metrics and logs
• Focus on migration patterns to keep transitions smooth as you scale
• Learn what actually matters when building from scratch

Scaling infrastructure is a constant grind. These insights come from real migrations and the patterns that made them work.

Take a look at the full story of how this stack evolved:

Building observability from scratch, three times over - Upsun Developer

A look back at how Platform.sh / Upsun's observability evolved from Nagios on a single box to a unified stack across thousands of VMs, and the migration patterns that made each transition smooth.

developer.upsun.com