Latency symptoms have cleared up. We are monitoring to identify cause if it begins again.
We are observing significant latency in TCP connections to the service. Engineers are working to evaluate the cause.
Investigating degraded performance from a number of gatways.