Skip to content
All outages

Incident · Medium · DNS/HTTP · Infra

Every core at 100%

  • SEV1
  • 13:42 UTC
  • GLOBAL
90s

The situation

Within a minute, every HTTP-serving machine in your edge network pins its CPU. Visitors get 502s and traffic falls by about 80%. Your own internal access tool is down too, so engineers struggle to log in.

The evidence

Edge CPU vs requests servedillustrative
  • CPU %
  • requests
Logs and signalsReconstructed
13:42waf: new managed rule deployed (mode=simulate)
13:42config store: pushed to all locations in seconds
13:43edge: CPU 100% on all HTTP cores
13:45edge: 502 Bad Gateway ↑
-- inbound packet volume: normal

Your call

What is eating the CPU?

Charts and logs are reconstructions. Aim for 90 seconds; going over costs nothing.

Last updated: Official postmortem