Good to know that it worked for you ass well.. I requested a supervision and alarm for this in the console so you get a heads up before you loose access to the cluster so you could schedule a reboot when it is most suitable.