TELEMETRY EVIDENCE
Jaeger Trace request causality across services.
Jaeger already holds the distributed traces your teams use for request flow. AutoObserve connects trace evidence to metrics, logs and change context — without replacing your existing tracing backend.
- Evidence
- Traces · Spans
- Protocol
- OTLP / Jaeger gRPC
- Enables
- Request flow
- Status
- Preview
Jaeger
Traces
AutoObserve
Request flow
Investigation
WHY THIS INTEGRATION MATTERS
Request flow is evidence — not a separate tool.
When checkout-api degrades, traces explain which downstream calls slowed first. AutoObserve connects Jaeger trace evidence to the same investigation window as metrics and deployments.
Request causality
Follow span paths from gateway to database without switching tools.
Coexistence
Keep Jaeger UI — AutoObserve subscribes to trace evidence in parallel.
Cross-signal correlation
Align trace latency with metrics spikes and deployment events.
Service identity
Resolve spans to the same service.name used across telemetry.
EVIDENCE RECEIVED
Traces and spans from Jaeger
Distributed trace evidence becomes request-flow context inside AutoObserve — linked to service identity and change events.
- Traces
Export
Forward OTLP or Jaeger-native spans to AutoObserve.
Identity
Align service.name and version with metrics and logs.
Connect
Join span latency to deployment and metric evidence.
Investigate
Use trace paths inside evidence-backed RCA.
HOW AUTOOBSERVE USES THIS EVIDENCE
Evidence becomes connected request-flow evidence.
Jaeger is not replaced — it is connected. AutoObserve treats traces as one evidence class in a multi-signal investigation.
Provides
- Traces
AutoObserve creates
- Request flow
- Span causality
- Cross-signal correlation
- Investigation context
Jaeger is request flow. AutoObserve is connected understanding.
COEXIST WITH JAEGER
Add AutoObserve without replacing your tracing stack.
Collector forwarding, dual-export and sidecar paths keep Jaeger authoritative while AutoObserve builds connected evidence.
Forward spans from your Collector to AutoObserve via OTLP.
Applications
OpenTelemetry Collector
AutoObserve
QUICK START
Connect Jaeger trace evidence
Configure OTLP export from your Collector, then verify checkout-api spans appear in connected evidence.
Configure OTLP traces
Add an AutoObserve OTLP endpoint to your Collector trace pipeline.
Align service identity
Ensure service.name matches metrics and log resource attributes.
Verify request flow
Confirm checkout-api spans resolve during a test investigation window.
WHAT SUCCESS LOOKS LIKE
Traces connected. Request flow ready.
Success means spans correlate with service identity and deployment context — not merely that OTLP export is configured.
Spans ingesting
checkout-api traces arrive within expected export intervals.
Identity aligned
service.name matches metrics and log identity.
Request paths visible
Downstream dependencies appear in span hierarchy.
REAL INCIDENT
Same checkout spike — trace path in focus.
The canonical checkout-api degradation with the trace span highlighted — the request-flow signal that explains downstream causality.
Metric
14:31:00
Checkout latency +184%
Walk through the evidence chain · 1/6
Evidence step 1 of 6
TROUBLESHOOTING
Common Jaeger integration failures
Problem
No traces arriving
Likely cause
OTLP endpoint, auth token or sampler dropping spans.
Resolution
Verify Collector exporter config and sampling rules for production services.
Verify
checkout-api spans visible within one export batch.
Problem
Traces present but service identity missing
Likely cause
service.name does not match OpenTelemetry resource attributes.
Resolution
Align SDK resource attributes with metrics and log conventions.
Verify
Spans resolve to checkout-api in production context.
Problem
Incomplete request paths
Likely cause
Aggressive head sampling hides downstream spans.
Resolution
Review tail sampling or error-biased sampling for critical paths.
Verify
Payment dependency spans appear during degradation window.
PRODUCTION INTELLIGENCE
How Jaeger contributes to production understanding
Trace evidence enables request-flow investigation inside connected production intelligence.
| Evidence | Enables |
|---|---|
| Traces | Request flow |
| Span latency | Downstream causality |
| Service graph | Dependency reasoning |
DESIGN PARTNERS
Validate this evidence source with your production stack.
Design partners connect real evidence sources and validate investigation workflows before GA — with direct engineering access.