Our engineers have been systematically assessing all of the centralized components of our service in North America, including our configuration databases, the greylisting service, our various spam and virus filtering engines, network responsiveness between datacenters, and other elements that could be causing the very slow response times of our filters.
The likely culprit from our testing is the centralized asset storage servers, which are responding more slowly than usual by nearly an order of magnitude. The engineers are investigating this in detail to determine what is causing multiple asset stores in multiple datacenters to be so affected.
We do not yet have an estimated time of resolution but are hopeful that once the cause of the asset storage issue is identified the resolution will follow soon thereafter.
We apologize for the issue and will continue working on resolving this as quickly as possible.