Decision Table

Map symptoms to measurements and fixes.

diagnosis

Detailed Description

A decision table turns symptoms into a debugging sequence. High p99, high CPU, slow DB, memory growth, and error spikes each suggest different first measurements.

This prevents random optimization. Start from the symptom, gather the right evidence, apply the smallest targeted fix, then verify with the same metric that exposed the issue.

Mental Model

high p99 -> event-loop lag?
high CPU -> flamegraph
slow DB -> query timing

When To Use

->You need a debugging order of operations
->Multiple bottlenecks appear together
->Incident response needs a checklist

Tools: Grafana, traces, logs, flamegraphs, slow query logs

diagnosis.js

Live Editor

Output

Run code to see output...