This section is still under construction. The goal of the labs is simple:
reproducible, hands-on exercises that teach observability by solving realistic performance problems—
using traces, metrics, and logs (tool-agnostic where possible).
What will be here soon
- Quickstart labs (10–20 min): setup + one focused RCA scenario
- Scenario labs: p99 spikes, downstream latency, retry storms, pool exhaustion, CPU throttling
- Kubernetes labs: HPA side-effects, noisy neighbors, pod restarts, resource limits
- OpenTelemetry labs: auto vs manual instrumentation, adding attributes, sampling basics
- RCA walkthroughs: metrics → traces → logs → causal chain + fix plan
Suggested starting points (already available)
- Root Cause Analysis Checklist
- Traces, Metrics, Logs
- p95 vs p99: Tail Latency
- Kubernetes signals for RCA
First lab I’m building
The first lab will be based on the OpenTelemetry Demo, with a guided exercise:
identify a p99 regression, locate the critical path in traces, confirm saturation signals in metrics,
and explain the failure mode via correlated logs.
Bookmark this page—new labs will appear here as they’re published.