Code RoomCDN purge melts origin
MediumPrep Room Coding #2919

CDN purge melts origin

On-callStorage & CDNMid–Senior~30 min

After a content-management deploy, a job issued a global CDN purge of the entire site ('purge everything') to be safe. Minutes later, origin RPS spikes ~25x, origin latency and 5xx climb, and the CDN hit ratio drops near zero across all POPs at once as every edge has to re-fetch from a cold cache. Users see slow loads and intermittent errors. The purge was meant to clear a few changed pages, not the whole site. How do you triage the impact, get the site healthy, and prevent a blanket purge from melting origin again?

What a strong answer looks like

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.

0:00 of about 30 min
Which questions mattered is sealed until you submit. Telling you now would just be handing over the edge cases.