DATA PROCESSING METHOD FOR DETECTING AND REPAIRING FAILURE OR MULTI-PROCESSOR NETWORK NODE

摘要

PROBLEM TO BE SOLVED: To provide a method for facilitating countermeasures by detecting the failure of a data processing node. SOLUTION: At the time of detecting node failure, at least one monitor data processing node 1202 periodically transmits received acknowledge data packet 1201 to each data processing node. The node 1202 decides that the data processing node has failed to receive a returned acknowledge data packet 1203 due to any failure. The data packet 1203 includes health data related with the present health operating state. The node 1202 transmits a reset data packet 1201 to the decided arbitrary defective data processing node. When the reset has failed to restore the data processing node to the normal operation, routing data in the adjacent data processing node are changed so that the data packet can by-pass the defective node for routing.