🔒 Paid

Memory Leak

A recent deploy introduced a memory leak, and usage on the API service has begun climbing steadily. GC pauses are growing longer, error rates are rising, and left untreated the process will eventually be OOM-killed. You cannot ship a code fix live — diagnose how to contain the leak's impact before it takes the service down, and make sure your fix actually survives the leak recurring.

Difficulty
Intermediate
Expected Time
~15 min
Category
Infrastructure & Capacity

Skills

  • Diagnosing memory leaks and GC pressure
  • Containment vs. root-cause fixes under time pressure
  • Process supervision and auto-restart policies
  • Recognizing plausible-but-incomplete mitigations

System Architecture

  • API Service
  • Request Handler
  • In-Memory Cache
  • Process Supervisor
  • Downstream Database

This is a Pro scenario. Upgrade to Pro to unlock it.

Upgrade to ProBack to Incidents