investigate

To find the root cause of an intermittent failure, gather exact error evidence and reproduction conditions first, then classify symptoms against common patterns like race conditions or state corruption before testing explicit hypotheses.

The best way to debug configuration drift and schema mismatches is through evidence-first investigation, mapping the observed symptoms to common failure modes and testing explicit hypotheses with minimal scope before applying fixes.

Troubleshoot dependency conflicts by extracting recent changes and exact reproduction conditions, classifying the failure pattern, and verifying a minimal-scope fix only after confirming the root cause through targeted hypothesis testing.

During incident response debugging, you should stop forming new hypotheses after repeated incorrect tests and request more context, ensuring fixes are applied only after the root cause is explicitly confirmed.

Prevent regressions when fixing routing mismatches by applying the minimal fix immediately after root cause confirmation, then creating and running an explicit regression test to verify the resolution.

Yes, evidence-based investigation applies to dependency conflicts and state corruption by enforcing pattern-based classification and safe hypothesis testing, ensuring fixes target the verified root cause rather than guesses.