// the find
iam-veeramalla/observability-zero-to-hero
Repo for learning observability
A 7-day hands-on tutorial series for building the standard Kubernetes observability stack from scratch: Prometheus/Grafana for metrics, EFK for logs, Jaeger and OpenTelemetry for tracing. It's aimed at developers or ops engineers who want a guided, day-by-day path to understanding observability rather than a library or tool to adopt into a project.
The progression is well thought out — it goes from raw metrics and PromQL, to custom app instrumentation with real alerting, to logging, to tracing, ending with OpenTelemetry tying it all together, which mirrors how teams actually layer this stuff in practice. Each day ships working artifacts, not just slides: Helm values files, Alertmanager configs, Kubernetes manifests, and small Node.js and Go services with actual instrumentation code (prom-client, tracing.js, OTel collector configs) you can deploy and poke at yourself.
It's a tutorial, not something you integrate — there's no package, no reusable code, and its value ends once you've worked through the 7 days. Some steps are EKS-specific (ALB controller setup in day 2/3), so anyone on a different cluster has to translate those parts. There's no CI, no tests, and nothing indicating the manifests are kept in sync with newer Prometheus/OTel Collector releases, which move fast enough that Helm values and chart versions here could silently rot. Navigation is also a bit rough — inconsistent readme.md/README.md casing and no top-level index linking the days together.