Question
At 14:02 your origin fleet's request rate jumps 40x and origin CPU and DB connections saturate; the site slows to a crawl. The CDN dashboard shows cache hit ratio collapsed from 96% to 31% across all POPs simultaneously, and origin 5xx is climbing. Twenty minutes earlier, someone shipped a config change that appended a cache-busting query param to asset URLs, and separately a marketing email blast just went out driving a traffic spike. Multiple POPs are now requesting the same uncached objects from origin at once. How do you triage, stop the origin from melting, and fix the root cause?
Stop the bleeding first (mitigate), then form hypotheses from real signals. Separate root cause from symptom, communicate status as you go, and close with what prevents a repeat.