We have narrowed our investigation and found the root cause of today's service outage. We are currently working on hardening this mechanism at our load-balancer level to prevent future occurrences.

We are currently investigating a brief outage that was returning HTTP 520s for a couple of minutes. All services are now healthy, but we are continuing to investigate root cause.