We identified multiple queries running more slowly than expected in our primary database, causing degraded performance. We are working on a fix.