PostgreSQL Connection Pool Exhaustion
Response time degraded from 150ms to 8000ms. Error rate jumped to 22%. Can you find the bottleneck before the on-call engineer does?
Handpicked cases that cover the most common interview topics
Response time degraded from 150ms to 8000ms. Error rate jumped to 22%. Can you find the bottleneck before the on-call engineer does?
The app runs fine at 200 RPS but crashes at 600 RPS. Heap dumps collected. Old Gen filling suspiciously. What's leaking?
One backend node is receiving 80% of traffic. Others are mostly idle. Classic or weighted? Let's investigate.
From database bottlenecks to infrastructure misconfiguration
Connection pools, slow queries, deadlocks, replication lag
14 cases →Memory leaks, thread starvation, GC pauses, CPU saturation
18 cases →Network bottlenecks, disk I/O, load balancer issues
9 cases →Alert tuning, metrics interpretation, SLO/SLA analysis
6 cases →The same process as a real performance debugging session
Pick by difficulty (Junior → Senior) or technology stack. Filter by time available.
Reveal clues one by one — graphs, logs, configs, query results. Think like an on-call engineer.
Write your root cause analysis and proposed solution. No time limit.
See the expert solution, common mistakes, and what interviewers actually look for.