We periodically gets the critical alarms that usually only occur once after an upgrade of the controller. By supports own admission (see case 00149356) this alarm is mostly ****
"In most cases we consider this error benign, because it usually seems to be triggered by the spooler being very busy processing messages, and so the request from the controller to the spooler fails intermittently. (This is also why you tend to see it right after a robot upgrade - everything's pretty busy because the robot's catching up after the restart.) The only time this error really indicates a problem is when it occurs over and over in a controller.log and the robot itself fails to start -- usually that indicates some problem with antivirus or a restrictive firewall, etc that is preventing the probes from communicating with each other."
I.E if there really is a persistent / actionable issue you will get a robot inactive anyway. It is possible to enter a key in the configuration that prevents this alarm - I would like this suppression built in by default.
“alarm_level_spoolererror=0” in robot.cfg