As previously mentioned partial failure is hard to deal with.
Reliable Insights
A blog on monitoring, scale and operational Sanity
Your HTTP router is usually the best place to measure your application latency.
Have you ever found yourself having to keep on updating and tweaking certain regexes in PromQL?
Services are not distinguished by their metric names in Prometheus.
While each application is different, a rough idea of how many metric there should be would be useful.
In this blogpost we'll run you through a quick 'hello world' example instrumenting a Rails application with the Prometheus ruby client.
Every now and then someone asks what metric type a increasing timestamp should be. Let's take a look.
We're often asked how to call set() on a Counter. So how do you do that?
