We are going to upgrade our main database and cache cluster instances with higher cpu and memory in order to accomodate capacity from new customers. This change will cause downtime of some minutes.