Question 1

Describe the Node.js event loop phases at a high level.

Accepted Answer

Timers â†’ pending callbacks â†’ idle/prepare â†’ poll â†’ check â†’ close callbacks, with process.nextTick and microtasks between. I/O callbacks mostly run in poll; setImmediate in check; setTimeout in timers. Microtasks (promises) and nextTick queues drain between phases and can starve I/O if abused. Interview depth often hinges on ordering, not memorizing every internal name.

Question 2

What is the difference between process.nextTick and Promise.then / queueMicrotask?

Accepted Answer

nextTick runs before other microtasks and can starve the event loop if recursively scheduled; Promise microtasks are the standard ES queue. Prefer promises/queueMicrotask for app logic. nextTick is for ensuring callbacks run before I/O continues — use sparingly. Recursive nextTick is a classic hang.

Question 3

setTimeout(fn, 0) vs setImmediate(fn) — which runs first?

Accepted Answer

It depends on context: inside an I/O callback, setImmediate usually runs first; in main module, ordering can vary. Don't rely on either for precise sequencing — use explicit control flow. This question tests whether candidates know 'it depends' rather than a false absolute.

Question 4

How can heavy synchronous work break the event loop?

Accepted Answer

Long CPU loops block the single JS thread, delaying timers, I/O, and health checks — appearing as latency spikes. Offload to worker_threads, child processes, or external services. Profile with clinic/0x or CPU profiles. Keep handlers short; batch and yield with setImmediate for cooperative multitasking when needed.

Question 5

Scenario: API p99 latency jumps but CPU is low. Event loop related?

Accepted Answer

Possibly event-loop delay from sync code bursts, UV thread-pool saturation, or lock contention — measure event loop lag. Instrument monitorEventLoopDelay / async hooks carefully. Check DNS, fs, and crypto pool sizes. Correlate with GC pauses. Low CPU with high latency often means waiting, not compute.

Question 6

Lead: how do you teach teams to reason about event-loop safety?

Accepted Answer

Ban sync fs in request paths, lint for known blockers, SLOs on event-loop delay, and design reviews for CPU-heavy features. Provide worker patterns and circuit breakers. Make 'don't block the loop' a shared language in postmortems. Automate detection in staging load tests.

Event Loop Node.js Interview Questions

Other Node.js interview topics