Describe a time when a critical production bug caused a service outage. How did you diagnose, fix, and prevent recurrence?

Use the CIRCLES framework: Context, Impact, Root cause, Corrective action, Lessons learned, Evaluation, Summary. 1) Set context and scope. 2) Quantify impact (SLA breach, user count). 3) Conduct root cause analysis (logs, stack traces, hypothesis testing). 4) Implement corrective action (code fix, regression tests). 5) Document lessons and update runbooks. 6) Evaluate effectiveness (monitoring, post‑mortem review). 7) Summarize outcomes and next steps. 120‑150 words, no narrative.

During a peak‑traffic period, our user‑profile service crashed, causing a 99.9% SLA violation and impacting 12,000 active users. I immediately assembled the incident response team, isolated the failure to a race condition in the Redis cache during high concurrency. We rolled back the last deployment, patched the cache logic, and deployed a hotfix within 45 minutes. Post‑mortem included root‑cause analysis, updated monitoring alerts, and a new automated test for cache consistency. I also introduced a canary deployment strategy to catch similar issues early. The fix reduced recurrence by 90% and improved our mean time to recovery from 30 to 12 minutes. 175‑200 words.