What makes Don’t Make Me Think timeless?
Most books in the tech field age fast. However, when I recently re-read Don’t Make Me Think by Steve Krug, I was struck by how little it had aged. So, what makes it still relevant?
blank
Software developer, passionate about observability of distributed and transactional systems
Latest posts
SLO formulas implementation in PromQL step by step
The service level terminology provides a framework for quantifying the quality of a service's reliability. There are plenty of resources available on SLI, SLO, and SLA, most theorizing. This article proposes PromQL implementations for availability and latency SLOs.
Self-hosted observability for Ruby on Rails apps with Kamal and OpenTelemetry
Observability is becoming a standard. Cloud observability providers deliver a high-end solutions for the storage and visualization of the telemetry data. Yet, application owners may consider an entirely in-house approach. Here is how you can achieve it for a Ruby on Rails app.
Collecting Prometheus metrics from multi-process web servers, the Ruby case
In Prometheus, metrics collection must follow concrete rules. It is a challenge with multi-process web servers where each scrape might reach a different instance of the app which holds a local copy of the metric. In this article I describe a rebellious solution of which combines StatsD with Prometheus.
Simple Prometheus queries for metrics inspection
Cardinality is key. And it’s easy to get it out of control. Check out a proposal on how to inspect your metrics with simple Prometheus queries and tune your instrumentation.