systematic-debugging

To find the root cause of flaky test failures, you must complete a Phase 1 investigation to gather evidence before proposing any fixes. This structured approach uses pattern analysis and single-hypothesis minimal testing to locate the actual source of the timing or race condition.

The best way to debug multi-component build or CI failures is to follow a phase-driven workflow that enforces root-cause-first investigation. This process uses backward tracing when the bug appears deep in the call stack, ensuring you verify the root cause before implementing any fix.

To stop applying patches and verify a root cause for unexpected production bugs, enforce the Iron Law of no fixes until Phase 1 evidence is gathered. You then use pattern analysis and minimal single-hypothesis testing to confirm the exact origin of the unexpected behavior.

When debugging fixes keep failing under time pressure, you should use explicit stop conditions to prevent rationalizations. This pressure-resistant process requires halting repeated fix attempts, stepping back to complete root-cause investigation, and applying stop rules after consecutive failures.

Yes, you can use systematic hypothesis testing for deep call-stack problems in integration tests. The workflow provides specific guidance for multi-component systems, including backward tracing from deep runtime failures to isolate and verify the root cause before fixing.

You need pattern analysis before fixing unexpected runtime behavior because it identifies failure trends across single-system and multi-component layers. This analysis is required during the phase-driven workflow to form a single, testable hypothesis that confirms the root cause and prevents recurring bugs.