Fixing log errors at the end of tests

[galvana]

Description Of Changes

During test teardown (and interpreter shutdown), Python's asyncio emits a late DEBUG-level log message ("Using selector: EpollSelector") via the standard logging module. Because the InterceptHandler routes all standard library logs through Loguru, this message reaches Loguru's stdout sink after pytest has already closed sys.stdout. Loguru's catch=True catches the resulting ValueError but prints a noisy diagnostic block at the end of every test run:

--- Logging error in Loguru Handler #5 ---
Record was: {'elapsed': ..., 'message': 'Using selector: EpollSelector', ...}
Traceback (most recent call last):
File ".../loguru/handler.py", line 206, in emit
self.sink.write(str_record)
File ".../loguru/simple_sinks.py", line 16, in write
self.stream.write(message)
ValueError: I/O operation on closed file.
--- End of logging error ---

This is harmless (no logs are lost, no tests are affected) but it's confusing and noisy.

The fix replaces the raw sys.stdout file object with a callable sink function (_safe_stdout_sink) that wraps sys.stdout.write() in a try/except ValueError. This is idiomatic for Loguru — callable sinks are a first-class feature and the codebase already uses one for the RedisSink. Callable sinks bypass Loguru's SimpleSink wrapper entirely, so the error is caught by our function before Loguru's error reporting ever sees it.

Code Changes

• Added _safe_stdout_sink callable sink function that writes to sys.stdout and silently drops ValueError from closed file descriptors
• Updated setup() to use _safe_stdout_sink as the default sink instead of sys.stdout directly

Steps to Confirm

1. Run the test suite (nox -s pytest or pytest --no-cov inside the container)
2. Verify the --- Logging error in Loguru Handler #5 --- diagnostic no longer appears at the end of the test output
3. Verify normal log output still appears correctly during tests (formatting, colors, levels)

Pre-Merge Checklist

• Issue requirements met
• All CI pipelines succeeded
• CHANGELOG.md updated
  • Add a db-migration This indicates that a change includes a database migration label to the entry if your change includes a DB migration
  • Add a high-risk This issue suggests changes that have a high-probability of breaking existing code label to the entry if your change includes a high-risk change (i.e. potential for performance impact or unexpected regression) that should be flagged
  • Updates unreleased work already in Changelog, no new entry necessary
• UX feedback:
  • All UX related changes have been reviewed by a designer
  • No UX review needed
• Followup issues:
  • Followup issues created
  • No followup issues
• Database migrations:
  • Ensure that your downrev is up to date with the latest revision on main
  • Ensure that your downgrade() migration is correct and works
    • If a downgrade migration is not possible for this change, please call this out in the PR description!
  • No migrations
• Documentation:
  • Documentation complete, PR opened in fidesdocs
  • Documentation issue created in fidesdocs
  • If there are any new client scopes created as part of the pull request, remember to update public-facing documentation that references our scope registry
  • No documentation updates required

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
fides-plus-nightly	Ready	Preview, Comment	Feb 6, 2026 10:41pm

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
fides-privacy-center	Ignored		Feb 6, 2026 10:41pm

[greptile-apps]

Greptile Overview

Greptile Summary

This PR fixes noisy logging errors that appeared at the end of test runs when asyncio emitted late log messages after pytest closed sys.stdout. The solution replaces the raw sys.stdout file object with a callable sink function (_safe_stdout_sink) that wraps sys.stdout.write() in a try/except to silently catch ValueError exceptions from closed file descriptors.

• Added _safe_stdout_sink function that safely writes to stdout and catches ValueError when the stream is closed
• Updated setup() to use _safe_stdout_sink instead of sys.stdout directly for the default sink
• This is idiomatic for Loguru - callable sinks are first-class and the codebase already uses one for RedisSink

Confidence Score: 5/5

• This PR is safe to merge with minimal risk
• The change is well-contained, addresses a specific issue with clear cause and effect, follows Loguru's idiomatic patterns (callable sinks), and only modifies the logging setup without changing any business logic. The implementation is simple and defensive - catching only the specific ValueError that occurs when writing to closed stdout.
• No files require special attention

Important Files Changed

Filename	Overview
src/fides/api/util/logger.py	Added _safe_stdout_sink callable to gracefully handle closed stdout during test teardown, preventing noisy error diagnostics

[greptile-apps]

_{1 file reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}

[thabofletcher]

🚀 🚀 🚀