opentelemetry-instrumentation-exceptions

What it is and what it does

This package is an OpenTelemetry instrumentation plugin that automatically hooks into Python's exception handling to capture exceptions that would otherwise go unobserved. It intercepts uncaught exceptions at the process level, thread-level exceptions, and unhandled asyncio task exceptions, converting them into OpenTelemetry log records. This is useful when you already have OpenTelemetry infrastructure in place and want to ensure all exceptions—even those that crash threads or tasks silently—are visible in your observability system.

The instrumentation is minimal to set up: a single call to `UnhandledExceptionInstrumentor().instrument()` at startup. It requires Python 3.10 or later and depends on the core OpenTelemetry API and instrumentation packages. Since it is still in Beta status and was first released in mid-2026, it may have rough edges or API changes in future versions.

Use it for:

• Capture silent thread exceptions in multi-threaded applications that would otherwise be lost.
• Monitor asyncio task exceptions in async applications to detect tasks failing without explicit error handling.
• Centralize exception visibility in microservices using OpenTelemetry backends like Jaeger or Datadog.
• Audit unhandled exceptions in production without modifying application code.
• Detect and alert on process-level crashes or fatal exceptions in real time.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Automatically captures uncaught exceptions in your Python process, threads, and asyncio tasks and emits them as OpenTelemetry logs for observability.

Yes, if you already use OpenTelemetry and need to capture exceptions that would otherwise be invisible. The low install friction, active maintenance, and permissive license make it a straightforward addition to observability infrastructure. The Beta status and recent release date mean you should test it in a non-critical environment first and monitor for any edge cases in your specific exception patterns.

Install

opentelemetry-instrumentation-exceptions on PyPI

pip install opentelemetry-instrumentation-exceptions

uv add opentelemetry-instrumentation-exceptions

poetry add opentelemetry-instrumentation-exceptions

Installing opentelemetry-instrumentation-exceptions

Before you install

Low install friction with a pure-Python wheel and only three core OpenTelemetry dependencies. Actively maintained as of 2026-08-14 with recent releases; marked Beta status.

License in practice

Apache-2.0 permissive license allows commercial and private use with minimal restrictions.

Quickstart

pip install opentelemetry-instrumentation-exceptions

from opentelemetry.instrumentation.exceptions import UnhandledExceptionInstrumentor

UnhandledExceptionInstrumentor().instrument()

Requires Python 3.10 or later; requires an active OpenTelemetry API and exporter setup to actually receive the emitted logs.

Verify before relying

• Whether this package works with all asyncio patterns (e.g., task groups, context vars) or has known limitations.
• Performance overhead when instrumenting high-exception-rate applications.
• Whether exception context (stack traces, local variables) is preserved in the emitted logs.

Package facts

Evidence: opentelemetry_instrumentation_exceptions-0.65b0-py3-none-any.whl

Tags

More Monitoring packages