I think it would be ideal if we discuss strategies to find these kind of database failures in the logs or even share tips on better logging design to make these kind of database failures surface to instance operators/admins.