We experienced two short periods of elevated latency and request timeouts on the Broker API: - 02:35 – 02:40 AM EDT - 02:59 – 03:04 AM EDT
Affected requests were mostly account and position reads, which can be safely retried, along with a small number of order submissions that timed out without returning a confirmation.
Root cause: Our scheduled overnight batch processing ran with higher-than-normal data volume, which temporarily reduced available capacity and slowed dependent API calls.
Status: Fully resolved as of 03:04 AM EDT. The service recovered once batch processing completed, and metrics have been normal since.