It might have been a result of Monday’s maintenance. It’s the second issue we’ve experienced since then but have a workaround here: