Question
Users report prices flapping: a product shows $19.99 on one page load and $24.99 on the next, inconsistently, after a pricing update was pushed at 13:00. Dashboards: the DB has the correct new price ($24.99) everywhere; a Redis cache fronts the product read path with a 1-hour TTL; cache hit rate is 92%; there are 8 app instances, each with a small in-process LRU cache in FRONT of Redis (a recent 'optimization' added two weeks ago) holding entries for 10 minutes; no TTL was reduced for the price push. How do you triage the flapping, serve correct prices now, and prevent stale prices on the next update?
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.