Middleware self-management hot backup method and middleware self-management hot backup system in cluster

Abstract

The invention discloses a middleware self-management hot backup method in a cluster. When failure of a main service node in a cluster system occurs, backup nodes can automatically sense the failure, and nodes in the backup state are automatically transferred to an election state; and based on internal multicast, one or multiple novel main service nodes can be automatically elected from all the backup nodes according to pre-specified rules, and client service node switching is notified. The invention further provides a system for achieving the middleware self-management hot backup method in a cluster. The system and a service middleware system are deployed in the same machine, the system is used to call interface of an operation system and acquire corresponding operation information of the process of service middleware according to the set information of the service middleware, and then according to returned information, the system is used to determine self working status namely main status or backup status or election status. In different working status, different logic is performed. According to the middleware self-management hot backup method and the middleware self-management hot backup system in cluster, the middleware self-management hot backup system can determine self-working status through multicast election rules; and on the basis of multicast interworking, backup and switching of the entire system can be scheduled and managed, and self-management of monitoring backup is achieved.

Description

technical field [0001] The invention belongs to the technical field of redundant hot backup in computers, and in particular relates to a self-management hot backup method and system for cloud services and key service middleware in clusters. Background technique [0002] With the rapid development of cloud computing and clusters, reliability has been widely valued as an important indicator to measure service quality. How to conveniently and quickly build high-reliability distributed services has become an important issue. [0003] The currently known dynamic hot backup technology is implemented in two ways: [0004] (1) Inside the service software, write a hot backup redundant module to form a dedicated hot backup system. [0005] (2) Use special hardware and software equipment for mirror backup and hot switching. [0006] As a software implementation, because it has no versatility, it increases the difficulty of building a backup system. At the same time, adding a hot back...

AI Technical Summary

Problems solved by technology

[0006] As a software implementation, due to its lack of versatility, it increases the difficulty of building a backup system. At the same time, adding a hot backup mechanism increases the complexity of the service software, which reduces the stability of the service middleware itself.

However, using the software and hardware mirroring backup method will undoubtedly increase the cost of system construction due to the additional investment in mirroring equipment. Since most services are middleware systems and will not generate new data, the mirroring method cannot achieve its due effect. role, which also resulted in a waste of investment

At the same time, for a large number of existing middleware systems without hot standby function, it is impossible to use software to build a middleware system group with highly reliable hot standby function

Images