Chapter 03 / 06

Signal Correlation

Logs, metrics, and traces in context

Step 1See the baseline latency

The latency chart shows P99 hovering in the normal range before the incident window. Watch what happens when the incident starts — the line jumps dramatically in one bucket.

Executive viewLearn how three different data types tell one coherent story — reducing guesswork in incidents.

3 data types, 1 story

Modern observability combines metrics, logs, and traces into one coherent picture. Without correlation, teams spend hours manually stitching these signals together. With Tracefox, they arrive at root cause in minutes.

📈
Metrics
Latency spike detected in payment-service
📋
Logs
Error logs confirm root cause
🔍
Traces
Exact request traced to payment-service

payment-service — latency timeline

P99 P50
4.2s2.1s0
T-12T-8T-4T+0T+4T+8T+12T+16
Incident window — payment-service degradation

payment-service — error logs

3 errors
09:12:07.342ERRORConnectionPoolTimeoutException: Timeout waiting for connection from pool after 30000ms (pool.size=10, pool.active=10)
09:12:09.104WARNConnection pool utilization at 100% — all 10 connections active. Requests will queue.
09:12:11.887ERRORConnectionPoolTimeoutException: Timeout waiting for connection from pool after 30000ms (pool.size=10, pool.active=10)
09:12:14.233ERRORConnectionPoolTimeoutException: Timeout waiting for connection from pool after 30000ms (pool.size=10, pool.active=10)
09:12:18.009WARNSlow query detected: db.payments INSERT 284ms (threshold: 200ms)
09:12:22.441INFOHikariCP pool stats: total=10, active=10, idle=0, waiting=47