Question
A compliance team's 'export my data' feature starts timing out and erroring for a wave of users this week. The export service fetches objects from storage and zips them; its dashboard shows request latency for certain object GETs jumping from milliseconds to many seconds or outright errors, and the error bodies say the object is in an archive/cold tier and not immediately retrievable. A lifecycle policy added last month auto-transitions objects older than 90 days to a cheaper archive class to cut cost, and these users have old data. How do you triage and unblock these exports, then design the retrieval path so it stops timing out?
Stop the bleeding first (mitigate), then form hypotheses from real signals. Separate root cause from symptom, communicate status as you go, and close with what prevents a repeat.