Code RoomRedis single core saturated
MediumPrep Room Coding #2962

Redis single core saturated

On-callConcurrencyDistributed systemsMid–Senior~35 min

Every service that touches the shared Redis suddenly sees latency spike together at 11:20 — sessions, rate-limiting, and feature flags all slow at once. Redis dashboards: the single Redis primary's CPU is pinned at 100% on one core (the box has spare cores), `instantaneous_ops_per_sec` is flat (not rising) while the slowlog fills up, and clients pile up in the connected-clients count. A new analytics feature deployed at 11:15 runs periodic `KEYS pattern:*` and large `SMEMBERS` calls against the same instance. Triage and mitigate.

What a strong answer looks like

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.

0:00 of about 35 min
Which questions mattered is sealed until you submit. Telling you now would just be handing over the edge cases.