Your question is Debugging a Production Issue. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
Tell me how you approach debugging a production issue in a live environment.
Describe how you establish impact, preserve evidence, narrow the fault domain, and choose a safe mitigation without making the incident worse. Explain how you use metrics, logs, traces, recent changes, reproduction, rollback, and post-fix validation.
This checks whether you debug from observable evidence rather than guesswork, distinguish mitigation from root-cause analysis, and protect customer data and system stability. Scope your answer to a live service or mobile-backed API environment, including the trade-offs of debugging directly in production.