Fault processing method of computer cluster system

Abstract

The invention discloses a fault processing method of a computer cluster system. The method comprises the following steps: (A) at least two nodes in the computer cluster system are selected and are set as management nodes which bear the fault processing and the management of the computer cluster system, one node in the management nodes is taken as a main node, and other nodes are taken as standby nodes, (B) a bottom monitoring service module of each node in the computer cluster system monitors the operation state of the node and software and hardware loads and judges whether a fault appears or not, and if so, the bottom monitoring service module notifies a message middleware service module to send a fault massage to a management center service module of the main node; and (C) the management center service module of the main node carries out fault processing according to the fault message. According to the technical scheme of the invention, in the condition that human intervention is not needed, the automatic processing function of the cluster computer system fault can be realized.

Description

technical field [0001] The present application relates to computer technology, in particular to a computer cluster system, and in particular to a method for troubleshooting a computer cluster system. Background technique [0002] With the advancement of information technology, both enterprises and other organizations are increasingly dependent on computer systems. With the rapid expansion of data volume, a single computer can no longer meet its needs, and the use of supercomputers will greatly increase the cost of computers. In this case, computer cluster technology emerged as the times require. [0003] A computer cluster system is connected by a group of loosely integrated computer software or hardware, and highly closely cooperates to complete computing work. Multiple computer devices forming a computer cluster system can be regarded as a computer logically. A single computer in a computer cluster system is usually called a node, and the computer cluster system can be c...

AI Technical Summary

Problems solved by technology

[0005] For dealing with failures in the computer cluster system, the usual method is for maintenance personnel to enter the computer room to find the faulty machine among multiple nodes in the computer cluster system, then determine the cause of the machine's failure, and then perform maintenance work. When the number of nodes increases, it may be necessary Increase the number and workload of maintenance personnel, not only the cost is high, but also the work efficiency is very low

Images