Code Room
On-callHardoc-g012
Subject Cache incidentsLevel Senior–Staff~35 minCommon in Reliability & on-call · Distributed systems interviewsIndustries Technology, Software development

Question

A celebrity posted at 20:00 and one piece of content is going viral. Your Redis Cluster spreads keys across 12 shards, but one shard's CPU is at 100% while the other 11 sit at 20%, that shard's network egress is saturated, and reads of the viral post's cached payload are timing out even though the data is cached and small. Overall cluster memory is fine. How do you triage and mitigate this in real time?

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.

Diagram & narrate the incident
Loading whiteboard…
Run or narrate your approach, then ask the coach.