common end of the road realization, monitoring usually starts simple, then slowly grows into dashboards, exporters, alerts, databases, and tooling that mostly exists to support more tooling. Meanwhile, when something actually breaks, people still SSH into the box because it’s the fastest way to understand reality and because of habits...doesn’t mean monitoring is useless, it just means many setups drift past the point of practical value.

for a medium sized enviro, most useful monitoring is usually, knowing something is broken, having enough context to start quickly, avoiding noise. beyond that, it’s easy to build a system that feels impressive but mostly creates maintenance work.

The important distinction is whether the monitoring helps you make decisions faster, or whether it became another thing you have to operate. A lot of homelabs and smaller infrastructures cross that line without noticing.