Our downstream database provider had a problem for 10 minutes between 9:23 and 9:33 PM UTC. That led to an increased API latency and some requests timing out. The incident has been resolved but we're continuing monitoring system health and investigating the problem with our database provider.