Skip to content
Interview rounds

Java/Spring/JVM · Production incident

Up, idle, and answering nothing

A service stops answering while its CPU is close to zero. How do you find out what its request threads are stuck on?

BackendSenior12 min
Interview-style scenarioLast reviewed: Checked against official documentation

Wednesday 11:48 · booking-serviceIllustrative scenario, not a real incident

booking-service: no responses for 6 minutes, the gateway returns 504, CPU at 3%

booking-service is a Spring Boot service on Java 21 with embedded Tomcat. At 11:42 it stopped answering on every endpoint. The processes are up, and CPU has dropped from 40% to 3%.

Pods that stop answering are restarted automatically. A restarted pod serves traffic for a few minutes and then goes silent again. Nothing was deployed this week.

What the dashboards and logs show

Requests completed
0 a second on a silent pod (30 before)
CPU
3% (40% before)
Heap and GC
normal, no long pauses
Pod restarts
14 in 6 minutes, across 6 pods

Clarify · optional · scored under "Hypothesis before change"

Ask the interviewer

0 of 3 asked

Each question reveals a fact. What you ask before you act is part of your score.

First move · scored under "Mitigate before root cause"

Customers cannot book, and the pods keep being restarted. What do you do first?

Your rubric score, the model answer, the follow-up questions and the related lessons appear once you have answered the last step.