Distributed Tracing
Instrument two Node.js services with OpenTelemetry, run Jaeger v2 as the trace backend, read a multi-service trace as a Gantt chart, and configure head-based sampling so tracing scales to real production traffic.
Prerequisites: Metrics and Dashboards with Prometheus.
A single user request in a microservices system might touch five services before a response comes back. Logs and metrics on any one service only show that service's slice; distributed tracing stitches every slice — called a span — into one tree (a trace) so you can see exactly where time was spent and where a failure originated.
You'll use OpenTelemetry for instrumentation (the vendor-neutral standard every major backend now speaks) and Jaeger as the trace backend. Jaeger v2 is the current supported line — it's built directly on the OpenTelemetry Collector framework and ingests OTLP natively, with no separate "enable OTLP" flag required. Jaeger v1 reached end-of-life on 2025-12-31, so v2 is the version to target.
Test what you just learned
Self-testing is one of the best ways to retain new skills. Unlock project quizzes to check your understanding.
Log in to unlock0 / 7 complete