Concepts
The mental model behind everything else. If you've never run OpenTelemetry before, read these in order — they assume nothing and explain the reasoning behind the defaults you'll see in the product.
Spans and tracesThe two primitives every other concept builds on.Release attributionHow GreenSlope ties an incident back to the commit that caused it.Change eventsDeploys, feature flips, env-var edits — every state change on one timeline.SLOs and burn ratesObjectives, error budgets, and the multi-window alerts that wake you up.AlertsRouting, acknowledgements, and the Doctor page that triages them.Tenants and tiersThe account model and what each tier limit actually means in practice.
Read these in order if you're new
If you've never run OpenTelemetry before, the fastest path to feeling at home in the product is:
- Spans and traces — what a span is, what a trace is, why we sample, why we keep error traces always.
- Release attribution — the one piece of glue that turns an error into a "this release broke."
- SLOs and burn rates — why we page on burn rate, not raw error counts.
- Alerts — what arrives in Slack, how to acknowledge, and how the Doctor page triages.
The other two — Change events and Tenants and tiers — you can read on demand when the product surfaces them.