[Nir-Az]

Summary

• Prevent re-entrant signal handling: a _cleanup_in_progress guard ensures a second Ctrl+C (or SIGTERM) during cleanup force-exits immediately via os._exit(1) instead of recursing into _cleanup_devices again
• Use os._exit(1) instead of sys.exit(1) after cleanup to avoid hanging on BrainStem USB hub threads that block SystemExit

Test plan

• On Windows: run pytest with an Acroname hub, press Ctrl+C — verify ports disable and process exits cleanly
• On Linux: same test with Ctrl+C
• Jenkins abort (SIGTERM): verify graceful shutdown without hang
• Press Ctrl+C twice rapidly — verify second signal force-exits immediately

🤖 Generated with Claude Code

[Copilot]

Pull request overview

This PR updates the Python unit-test signal handling utilities to avoid re-entrant cleanup and to prevent hangs on Ctrl+C/SIGTERM by forcing a hard process exit once cleanup begins.

Changes:

• Add a module-level _cleanup_in_progress guard to prevent re-entrant signal handling and force-exit on a second signal during cleanup.
• Replace sys.exit(1) with os._exit(1) to avoid hangs caused by threads blocking normal interpreter shutdown.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

sys is no longer used in this module after switching from sys.exit to os._exit. Please remove the unused import (or, if you keep os._exit, use sys to flush stdio before exiting).

[Copilot]

os._exit() bypasses Python exception propagation and stdio flushing. If signal_handler() raises or logs buffered output, the process may continue unexpectedly or drop the last log lines. Consider wrapping signal_handler() in a try/finally (or try/except + finally) that always flushes output and then calls os._exit(1) so abort is guaranteed even if cleanup fails.

[AviaAv]

LGTM