This incident has been resolved.

This issue has been identified as a route leak (https://www.thousandeyes.com/learning/glossary/bgp-route-leak) on a global scale that affected CloudFlare, AWS and other world-wide service providers.

Cloudflare status: https://www.cloudflarestatus.com/incidents/46z55mdhg0t5
AWS status: https://status.aws.amazon.com/

Although the situation is not fully resolved, our infrastructure and the services it relies on are now functioning well, though we are unable to rule out further service interruptions. We are continuing to monitor the situation.

We are continuing to work on a fix for this issue.

The incident was acknowledged by CloudFlare; you can follow along for updates here: https://www.cloudflarestatus.com/incidents/46z55mdhg0t5

Network performance issues at CloudFlare are causing degraded performance or outages on all regions, affecting customers using CloudFlare.

The issue seems to be caused by an unconfirmed outage of CloudFlare.

We are currently investigating this issue.