Gateway Gus saysObservability rests on three pillars: logs, metrics, and traces.
To understand a running system you need logs (discrete events — what happened), metrics (numeric time-series — how much/how fast), and traces (the path of one request across services — where time went). Together they answer "is it healthy, and if not, why?" Modern stacks (OpenTelemetry) unify how you collect all three.
Power-ups you unlock
Logs: discrete events (what happened)
Metrics: numeric time-series (how much/fast)
Traces: one request across services (where)
OpenTelemetry unifies collection
Timeout Titan attacks — common mistakes
Relying on logs alone (no metrics/traces)
Unstructured logs that can’t be queried
No correlation id linking the three
Boss battleMatch each pillar to the question it best answers.
Example code
<!doctype html><html><head><meta charset="utf-8"></head>
<body style="background:#06040d;color:#e6e0ff;font-family:monospace;padding:20px"><pre>logs → what happened?
metrics → how much / how fast?
traces → where did the time go?</pre></body></html>