Incident · Easy · DNS/HTTP
Healthy servers, no visitors
- SEV1
- 15:45 UTC
- UP TO ~1 HOUR
90s
The situation
Some sites behind a large CDN go down for up to an hour. The origin servers are fine and the CDN's edge machines are up. Visitors just cannot find them. A software configuration update went out at 15:45 UTC.
The evidence
Logs and signalsReconstructed
15:45edge DNS: software configuration update applied15:46resolvers: no answer for customer hostnames15:47origin health: OK · edge health: OK15:50inbound traffic volume: falling, not rising
Your call
What is the most likely cause?
Charts and logs are reconstructions. Aim for 90 seconds; going over costs nothing.