Many product-company interviews ask you to walk through a real production outage or latency spike. Beyond checking CPU/memory and slow query logs, what metrics—like connection pool exhaustion or garbage collection pauses—do you check first?
Log in to comment