At ~7:01 am and ~11:04 am eastern, we saw brief spikes in latency leading to increased error rates, due to a database having to be restarted by our host. We are working with our host to gain more information here on the root cause, and have put in place preventative measures that we hope will guard against this moving forward