The incident has been resolved. All calls received during the outage will be processed once the current queue is cleared.
An additional four-minute planned downtime of the async API was required to prevent further issues, from 11:39–11:43 UTC (07:39–07:43 EDT, 04:39–04:43 PDT)