Advanced Monitoring

Beyond the dashboards, KubeOpera exposes its observability data through four services you can query directly — from scripts and CI jobs.

Service	Gives you	Backed by
log-gateway	Logs for a service or namespace, plus extracted patterns.	Loki
tracing-gateway	Trace detail, slow and failing traces, and the service map.	Jaeger
apm-gateway	Request rate, p99 latency and error rate per service.	Prometheus
slo-manager	SLO compliance, error budget and burn rate.	Prometheus

Each takes simple parameters — a service, a namespace, a time range — and translates them into the right LogQL, PromQL or Jaeger query, so you don't need to know those languages.

Example: a deploy gate in CI

Block a production deploy if the service's error budget is nearly spent:

BUDGET=$(curl -s -H "Authorization: Bearer $KUBEOPERA_TOKEN" \
  "https://kubeopera.example.com/api/slo/api/v1/slos/checkout-availability/status" \
  | jq -r .error_budget_remaining)

if (( $(echo "$BUDGET < 0.1" | bc -l) )); then
  echo "Error budget below 10% — holding the deploy."
  exit 1
fi

Sending your own traces

KubeOpera runs an OpenTelemetry collector that forwards to Jaeger. Point your apps' OTLP exporter at it (see Monitoring your applications) and their traces appear in tracing-gateway, the Traces page and AI investigations automatically.

Next steps

• Dashboards reference — the same data in the UI.