Skip to main content

Service inventory: every service, RED at a glance

avuru obs team
Maintainers

The Services screen is live: every service seen in the selected window — auto-discovered zero-code on Kubernetes or sending OTLP directly — with its rate, error percentage and latency percentiles in one sortable table.

  • RED per service. Request rate, error % (badge when non-zero) and p50/p95/p99, computed from the same span data the trace explorer uses.
  • Drill-down. Click a row to open that service's traces, already filtered; or jump to the topology view of the same services on the service map.
  • Consistent filters. The global time range and the auxiliary-traffic toggle work here like everywhere else, and the state lives in the URL — a filtered view is shareable.