Skip to content
BENIBIX Learn

Observability

Infrastructure📈 Monitoring & ObservabilityMonitoring Basics

Also known as O11y, System Observability, Software Observability

How well you can see inside a system. You look at its logs, metrics and traces.

Think of it like

It is like a doctor with an X-ray. You can look inside and find a problem you did not expect.

Example

Checkout is slow for some buyers in Japan only. Emma sorts the traces by country and finds one slow payment call. Nobody had planned for that question.

Why it matters

Modern apps are spread over many parts. When something odd breaks, you need data that lets you ask questions you did not plan.

Definition

Observability is how well you can tell what is going on inside a system from the data it gives off. That data comes in three main kinds: logs, metrics and traces. With enough of it, you can ask new questions while a problem is still going on. Monitoring tells you that something is wrong. Observability helps you find out why.
Observability means joining these three views. Each one shows a different part of why a page is slow.

Common mistake

Thinking it is a tool you can buy. It is a trait of your system, and it comes from sending out good data.

Key takeaways

  1. 1
    Three signals. Logs, metrics and traces, joined up.
  2. 2
    New questions. Ask what you did not plan to ask.
  3. 3
    Built in. The code has to send out the data.

Compare