Question 1

How do you diagnose a Node memory leak?

Accepted Answer

Heap snapshots over time, allocation timelines, and watch retained size of suspects (caches, closures, listeners). Reproduce with load. Compare snapshots. Check unbounded Maps, event listeners, and global arrays.

Question 2

process.memoryUsage() fields — what do they mean?

Accepted Answer

rss, heapTotal, heapUsed, external, arrayBuffers — different views of memory consumption. rss includes more than JS heap. external/arrayBuffers matter for Buffers. Don't alert only on heapUsed.

Question 3

Why can Buffer usage grow rss without huge heapUsed?

Accepted Answer

Buffer memory may be off-heap/external relative to V8 heap accounting. Track external/arrayBuffers. Stream instead of retaining buffers. Release references promptly.

Question 4

In-process caches — safe patterns?

Accepted Answer

Bounded LRU with TTL, memory budgets, and metrics; avoid unbounded Maps. Per-instance caches duplicate data horizontally. Consider Redis for shared cache. Stampede protection.

Question 5

Scenario: heap grows only in production under traffic.

Accepted Answer

Likely request-correlated retention: closures over req, orphaned timers, or growing module-level state. Add continuous profiling. Check ALS misuse. Verify framework middleware not buffering bodies.

Question 6

Lead: memory SLOs for Node services on K8s.

Accepted Answer

Set requests/limits from measured rss, alert on monotonic growth, OOMKill dashboards, and load-test leak checks in CI. Right-size before raising limits blindly. Treat OOMs as defects. Budget caches explicitly.

Memory Node.js Interview Questions

Other Node.js interview topics