This incident has been resolved.
We're investigating a 2 minute outage caused by a long running query on our database.