We've changed network traffic in our Europe region to a different transit path, and are seeing recovered performance. We're allowing some jobs to run in this region again, while we monitor to verify full recovery.