Due to a failed deployment, high error rates were observed for some customers on our Europe endpoint causing a partial outage that lasted 3 minutes. Our team rolled back the deployment immediately. We're investigating why this deployment failed and will put in place processes to avoid this in the future.