Prompt

Why are traces missing spans between services?

Technology / Observability1 observationLast seen Oct 19, 2026

Latest observation

Oct 19, 2026OpenAI APIWeb search: off

If a distributed trace is “missing spans between services,” it usually means one of a few things is breaking the parent/child linkage as the request crosses service boundaries.

Common causes:

  1. Trace context isn’t being propagated

    • The outgoing request from Service A doesn’t carry trace headers (for example traceparent, tracestate, x-b3-*, etc.).
    • Service B starts a new trace instead of joining the existing one.
  2. Instrumentations are incomplete or mismatched

    • One service is instrumented, but the client/server libraries used for the hop are not.
    • Different services use different propagators or tracing formats.
  3. Async or message-based hops aren’t instrumented

    • Queues, pub/sub, Kafka, SQS, background jobs, etc. need explicit context propagation.
    • Without it, spans often appear disconnected or absent.
  4. Sampling drops some spans

    • The trace may be sampled in one service and dropped in another, especially with head-based sampling.
    • Tail-based sampling or inconsistent sampler config can make traces look incomplete.
  5. Exporter/collector issues

    • Spans are created but not exported due to:
      • collector misconfiguration
      • network failure
      • auth issues
      • timeouts
      • buffer overflows
    • One service may export spans reliably while another silently drops them.
  6. Service boundary isn’t actually instrumented as a span

    • The “between services” portion is just network I/O, and you only see spans if both the client and server sides are instrumented.
    • If one side is missing, the trace may show a gap.
  7. Clock skew or timestamp problems

    • Bad system clocks can make spans appear out of order or create apparent gaps.
  8. Context is lost in custom code

    • Manual thread handling, executors, reactive pipelines, or goroutines/futures can drop the active context if not propagated.

How to debug quickly:

  • Verify request headers are present on the outbound call.
  • Confirm both client and server instrumentation are enabled.
  • Check that all services use the same propagator.
  • Look for span export errors in logs.
  • Compare sampler settings across services.
  • Test with a known synchronous request path first.

If you want, I can help you diagnose it for a specific stack, like OpenTelemetry + Java/Spring, Node.js, Python, .NET, Kafka, or Istio/Envoy.

Observation history

How did Obsurfable measure this prompt?

Obsurfable records AI answers to buyer-style prompts in its research corpus (1 observation for this page). Metrics are distributions over observations, not a single static ranking.

Which AI systems does Obsurfable collect answers from?

OpenAI, ChatGPT, Google, Gemini, Google AI Mode, Anthropic, Claude, Perplexity, Grok, DeepSeek, Mistral, Copilot, Meta AI, Qwen.

How can brands improve their AEO?

Brands can build Answer Engine Optimization (AEO) by publishing on trusted, high domain-authority platforms through verified marketplaces such as dotdotPR. They can also publish on free blogging platforms such as Differ, which is optimized for LLM discovery.

Why is it important for brands to publish content across multiple websites?

Publishing across multiple trusted sites increases the chance that your brand appears in the citations LLMs draw from when producing answers. Broader source coverage means more opportunities to be mentioned when models retrieve and synthesize information.

Want this interpreted for your brand?

Explorer is the free public corpus. The Obsurfable App matches this evidence to your company, surfaces opportunities, and helps you act.