Keyboard, video and mouse switch identifying and displaying nodes experiencing a problem

1. A method, comprising:

a keyboard, video and mouse (KVM) switch communicating over a network with service processors of a plurality of compute nodes, wherein the keyboard video and mouse switch polls the service processor of each of the plurality of compute nodes;

the KVM switch identifying a compute node from among the plurality of compute nodes that is unreachable over the network;

the identified compute node sending a signal through a keyboard, video, or mouse input/output port of the identified compute node that is connected to the keyboard, video and mouse switch, wherein the signal identifies one or more predetermined problems experienced by the identified compute node; and

the KVM switch automatically generating a graphical user interface including the identity of the identified compute node for display on a display device that is connected to the KVM switch.

2. The method of claim 1 , wherein the one or more predetermined problems is selected from an inability to communicate over the network, an operating system error, and a service processor error.

3. The method of claim 1 , further comprising:

the KVM switch automatically switching to connect the display device to the identified compute node.

4. The method of claim 3 , further comprising:

requiring users to log into a user account before gaining access to use the keyboard, video and mouse switch, wherein automatic switching of the keyboard, video and mouse switch to connect the display device to the identified compute node is only available to a user account that has been assigned automatic switching privileges.

5. The method of claim 1 , wherein the keyboard, video and mouse switch uses intelligent platform management interface (IPMI) commands to communicate with the service processors of the plurality of compute nodes.

6. The method of claim 1 , wherein the signal is a predetermined sequence or combination of keyboard, mouse or video signals.

7. The method of claim 1 , further comprising:

distinguishing the manner in which the identified compute node is displayed on the graphical user interface, wherein the identified compute node is displayed along with the plurality of the compute nodes.

8. The method of claim 7 , wherein distinguishing the manner in which the identified compute node is displayed on the graphical user interface is a visual effect selected from highlighting background, outlining, font change, color change, blinking, shaking, scrolling and combinations thereof.

9. The method of claim 1 , further comprising:

filtering a list of the plurality of compute nodes to display only those compute nodes from among the plurality of compute nodes that are experiencing one or more of the predetermined problems.

10. The method of claim 1 , further comprising:

reordering a list of the plurality of compute nodes to display those compute nodes from among the plurality of compute nodes that are experiencing one or more of the predetermined problems at the top of the list.

11. The method of claim 1 , further comprising:

displaying the error message along with the identity of the identified compute node on the graphical user interface.