Skip to content

Conversation

@alexander-alderman-webb
Copy link
Contributor

@alexander-alderman-webb alexander-alderman-webb commented Jan 12, 2026

Description

  • Patch StreamableHTTPServerTransport.handle_request() to store isolation and current scopes on the Starlette request object's scope attribute.

  • Run MCP handlers in the context of the stored scopes, if these are available.

  • Modify use_isolation_scope() and use_scope() context managers to accept None. The MCP protocol allows for transport layers that are not HTTP-based, so the scopes may not be available.

The Python MCP SDK uses an in-memory queue of requests. The HTTP propagation context is not active in the MCP decorators we patch, since they run after requests are retrieved from the queue.

Issues

Closes #5272
Closes PY-2022
Closes TET-1688

Reminders

Comment on lines 1874 to 1902
def use_isolation_scope(isolation_scope: "Scope") -> "Generator[Scope, None, None]":
def use_isolation_scope(isolation_scope: "S") -> "Generator[S, None, None]":
"""
.. versionadded:: 2.0.0
Context manager that uses the given `isolation_scope` and runs the wrapped code in it.
The current scope is also forked to not bleed data into the existing current scope.
After the wrapped code is executed, the original scopes are restored.
Example Usage:
.. code-block:: python
import sentry_sdk
with sentry_sdk.isolation_scope() as scope:
scope.set_tag("color", "green")
sentry_sdk.capture_message("hello") # will include `color` tag.
sentry_sdk.capture_message("hello, again") # will NOT include `color` tag.
"""
if isolation_scope is None:
yield isolation_scope
return
Copy link
Contributor Author

@alexander-alderman-webb alexander-alderman-webb Jan 12, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure about accepting None in the context managers. It's convenient in the MCP integration since it helps avoid code duplication (or the need to break stuff into more functions).

We could also just check if the scopes are None, so I can adjust if needed.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I probably wouldn't allow None in the context managers. If they were just internal utils it'd probably be ok to update them to support this sort of internal usecase, but since they're part of the public API, we should keep them tidy.

You might be able to use nullcontext instead to get around some of the potential mess of having to conditionally wrap code in a with block.

@alexander-alderman-webb alexander-alderman-webb marked this pull request as ready for review January 12, 2026 14:43
@alexander-alderman-webb alexander-alderman-webb requested a review from a team as a code owner January 12, 2026 14:43
@github-actions
Copy link
Contributor

github-actions bot commented Jan 13, 2026

Semver Impact of This PR

🟢 Patch (bug fixes)

📋 Changelog Preview

This is how your changes will appear in the changelog.
Entries from this PR are highlighted with a left border (blockquote style).


New Features ✨

  • feat(ai): add parse_data_uri function to parse a data URI by constantinius in #5311
  • feat(asyncio): Add on-demand way to enable AsyncioIntegration by sentrivana in #5288
  • feat(openai-agents): Inject propagation headers for HostedMCPTool by alexander-alderman-webb in #5297
  • feat: Support array types for logs and metrics attributes by alexander-alderman-webb in #5314

Bug Fixes 🐛

Integrations

  • fix(integrations): pydantic-ai: properly format binary input message parts to be conformant with the gen_ai.request.messages structure by constantinius in #5251
  • fix(integrations): Anthropic: add content transformation for images and documents by constantinius in #5276
  • fix(integrations): langchain add multimodal content transformation functions for images, audio, and files by constantinius in #5278

Litellm

  • fix(litellm): fix gen_ai.request.messages to be as expected by constantinius in #5255
  • fix(litellm): Guard against module shadowing by alexander-alderman-webb in #5249

Other

  • fix(ai): redact message parts content of type blob by constantinius in #5243
  • fix(clickhouse): Guard against module shadowing by alexander-alderman-webb in #5250
  • fix(gql): Revert signature change of patched gql.Client.execute by alexander-alderman-webb in #5289
  • fix(grpc): Derive interception state from channel fields by alexander-alderman-webb in #5302
  • fix(mcp): Nest MCP spans under HTTP transactions by alexander-alderman-webb in #5292
  • fix(pure-eval): Guard against module shadowing by alexander-alderman-webb in #5252
  • fix(ray): Guard against module shadowing by alexander-alderman-webb in #5254
  • fix(threading): Handle channels shadowing by sentrivana in #5299
  • fix(typer): Guard against module shadowing by alexander-alderman-webb in #5253
  • fix: Stop suppressing exception chains in AI integrations by alexander-alderman-webb in #5309
  • fix: Send client reports for span recorder overflow by sentrivana in #5310

Documentation 📚

  • docs(metrics): Remove experimental notice by alexander-alderman-webb in #5304
  • docs: Update Python versions banner in README by sentrivana in #5287

Internal Changes 🔧

Release

  • ci(release): Bump Craft version to fix issues by BYK in #5305
  • ci(release): Switch from action-prepare-release to Craft by BYK in #5290

Other

  • chore(gen_ai): add auto-enablement for google genai by shellmayr in #5295
  • chore: Use pull_request_target for changelog preview by BYK in #5323
  • chore: add unlabeled trigger to changelog-preview by BYK in #5315
  • chore: Add type for metric units by sentrivana in #5312
  • ci: Update tox and handle generic classifiers by sentrivana in #5306

🤖 This preview updates automatically when you update the PR.

Copy link
Contributor

@sentrivana sentrivana left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

See comments, but other than that LGTM!

@alexander-alderman-webb alexander-alderman-webb marked this pull request as draft January 16, 2026 14:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

MCP tool calls break distributed tracing within same service

3 participants