advancedResource Management: Requests, Limits, QoS Classes & OOMKilled
Why can a container show low average CPU usage in monitoring while still suffering from CPU throttling?
Throttling happens in short bursts within each CFS scheduling period whenever the container's usage spikes above its per-period allotment — those brief throttled bursts get smoothed out and hidden by a longer-window average, which is why dedicated throttling metrics (not just average CPU%) are needed to actually see it.
Ready to master this question?
Generate a complete walkthrough — background, the full answer in plain language, a working code example explained line by line, a real-world scenario, common mistakes, and how this same question gets asked in different ways.
Sign in to generate a response