API recovered automatically after AWS triggered a server shutdown.

This is caused by our production environment currently being configured with just a single instance running of the back-end. In the future, we'll have multiple instances running, so that a restart of a single instance does not cause downtime.

API went down.