We have resolved the issue and continue to monitor the situation.
We have identified the issue, one of our time-series databases has a degraded performance, and are working on a fix.
We are investigating increased latencies with the API.
