Method and device for handling nodes in distributed system

Abstract

The embodiments of the invention provide a method and a device for handling nodes in a distributed system. The nodes include service nodes and a center node. The method includes the following steps: acquiring survival state information of the service nodes; acquiring current system information of the center node; determining whether each service node is abnormal according to the survival state information and the current system information; if there is an abnormal service node, acquiring center state information of the center node; and handling the abnormal service node according to the center state information. According to the embodiments of the invention, an abnormal service node can be handled adaptively according to the state of the center node. Therefore, misjudgment on the state of the service nodes because of the problems of the center node is reduced, and the error probability of the center node is reduced.

Description

technical field [0001] The present application relates to the technical field of data processing, in particular to a node processing method in a distributed system and a node processing device in a distributed system. Background technique [0002] A distributed system is a system composed of one or more independent nodes that are geographically and physically dispersed, and the nodes include service nodes and central nodes. The central node can coordinate and deploy service nodes. These nodes are connected together to realize the sharing of resources, and the distributed system is equivalent to a unified whole. [0003] During the operation of the distributed system, the monitoring of the survival status of the service nodes is a very important link. The usual practice is that each service node in the distributed system sends survival status information to the central node at a predetermined period. After the central node receives the survival state information, it is used...

AI Technical Summary

Problems solved by technology

However, the central node may not be able to receive the survival status information reported by the service node in time due to network delay, or may not be able to process the survival status information in time due to the overload of system resources. These situations may lead to the failure of the service node. The survival status information is lost, or the next update time is invalid, etc., then the central node may misjudge the survival status of the service node

Images