The root cause has been identified. A background service caused extreme database load by performing a task improperly, which resulted in our web instances being taken out of the pool. The cause has been corrected, and service is stable.
We’ve found the source of the issue and are working to identify the root cause. Service has been fully restored.
We’re currently investigating an issue which has caused all web nodes to be taken offline.