Service data synchronization method, data recovery method, data recovery device and network device

Abstract

The invention discloses a service data synchronization method, a data recovery method, a data recovery device and a network device. The synchronization method comprises the following steps that: a first server used as the master server executes a synchronization processing affair; the synchronization processing affair comprises the following steps of: executing a first service affair to generate a first service data, recording an identifier of the first service affair in a local first service affair record list, generating a service affair log and writing the identifier of the first service affair and the service affair log in a service affair log sheet of a second server, and setting the submission state information of the first service affair as not submitted; if the synchronization processing affair is executed successfully, the first server sets the submission state of the first service affair as submitted; and the first server executes the synchronization affair so as to synchronize the first service data with the second server. With the adoption of the technical scheme of the invention, the databases of the master server and a standby server can be completely consistent before and after switching.

Description

technical field [0001] The invention relates to network communication technology, in particular to a service data synchronization method, recovery method and device, and network equipment. Background technique [0002] In order to improve the availability of the server system, the active-standby redundancy design is often adopted to shorten the service unavailable time due to the failure of the main server. High-availability cluster is one of the active and standby redundant designs of server systems, and it is a server cluster technology aimed at reducing service interruption time. In a high-availability cluster, if a server fails, its standby server will take over the responsibility of the server in a short time, so that the overall service of the high-availability cluster is as available as possible, thereby reducing the risk caused by hardware and software abnormalities in the cluster. coming loss. [0003] In order for the standby server to take over the responsibilit...

AI Technical Summary

Problems solved by technology

Since data replication in units of transactions is asynchronous, real-time performance is affected by factors such as the network and the load of the primary server, so there is a delay in the data synchronization process of the database replication technology, which may cause it to be too late in the process of switching between the primary and secondary servers. The data is synchronized to the standby server, resulting in inconsistencies between the databases of the active and standby servers before and after the switchover

Images