maple-python-style

You can establish OpenTelemetry tracing by calling init_observability() before starting FastAPI components. This provides module-scoped initialization for tracer and meter objects, reducing boilerplate while ensuring consistent observability across services.

Module-scoped OpenTelemetry initialization sets up tracer and meter objects at the module level to reduce duplication. It ensures consistent observability best practices across backend services by providing a single initialization point with an initialization guard to avoid reinitialization.

Yes, you can use decorators for bounded work to define clear span boundaries for worker tasks. This approach improves observability by automatically capturing tracing and metrics data for your background workers without manual span management.

Yes, the setup supports logging integration using OTLP exporters and LoggingInstrumentor for end-to-end visibility. It captures traces, metrics, and logs together, applying proper resource attributes across FastAPI, worker tasks, and CLI utilities.

You can configure inline endpoint and ingest key settings directly to simplify deployment. This approach avoids env-tangle in code paths while satisfying requirements for capturing traces, metrics, and logs with OTLP exporters.

OpenTelemetry initialization might run multiple times if not guarded properly. This setup includes an initialization guard specifically to avoid reinitialization, ensuring module-scoped tracer and meter objects are only configured once during application startup.