Computer system fault diagnosis

• Main Authors: MILLER, JOSEPH P, HULL, DANIEL S, CULLEY, PAUL R, TAVALLAEI, SIAMAK
• Format: Patent
• Language: English; French; German
• Published: 10.12.2003
• Edition: 7
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

Faults in a computer system having circuits are managed by fault detectors connected to detect fault states of respective circuits. A fault manager associates the fault states with the respective circuits. The fault manager includes a system manager connected to identify which of the circuits is causing faulty operation in the computer system. The fault detectors associated with the respective circuits are configured to detect faulty operation of and to generate fault state information for the respective circuits. A central manager is connected to accumulate fault state information from the fault detectors. One of the circuits includes a bus, and the fault state includes a bus error condition. The bus is connected to multiple devices, and the fault manager identifies which of the multiple devices causes the bus error condition. One of the circuits includes multiple modules, and the fault manager identifies fault states of the multiple modules. The modules include state machines. One of the circuits includes an internal clock, and the fault state of the circuit includes the internal clock not functioning properly. One of the circuits includes a temperature sensor, and the fault state of the circuit includes a high temperature condition detected by the temperature sensor.