Skip to content

Upstream Source

This page is part of Truthound Orchestration 3.x.

Source repository: seadonggyun4/truthound-orchestration Upstream docs path: docs/engines/error-reporting.md Edit upstream page: Edit in orchestration

Engine Error Reporting and Diagnostics

Engine failures become much easier to operate when teams distinguish among capability errors, source errors, execution failures, and serialization drift.

Who This Is For

  • operators reading engine-level failures in CI or production
  • contributors adding diagnostics to adapters or engines

When To Use It

Use this page when an engine-related failure does not clearly belong to the host itself.

Diagnostic Layers

Layer Typical Question
registry did we create the engine we expected?
versioning is this engine version compatible with the supported tuple?
capability does the engine support the requested operation?
execution did the engine fail while running the operation?
serialization did the result survive transport back into the host?

Practical Reading Order

  1. confirm engine selection and runtime context
  2. inspect the compatibility or preflight report
  3. inspect host-native logs and emitted result metadata
  4. compare the failure against the shared Failure Catalog

Production Pattern

  • keep engine diagnostics visible in host-native logs
  • treat preflight failures as configuration issues, not as runtime flukes
  • keep result consumers on the shared serializer contract