Quick fault-tolerant processing method for data-writing Redis failure

Abstract

The invention discloses a quick fault-tolerant processing method for data-writing Redis failure. The method includes S1, adding a Redis service failed in play as a fault-tolerant task to a fault-tolerant to-be-processed queue; S2, acquiring a fault-tolerant to-be-processed task from the fault-tolerant to-be-processed queue, replaying the Redis service if the fault-tolerant to-be-processed task exists, adding one time in replaying the Redis service, continuously performing the S3, and performing the S4 if the fault-tolerant to-be-processed task does not exist; S3, judging whether replaying of the Redis service in the S2 is successful or not, if yes, storing the Redis service to a Redis server cluster, skipping to the S2 for continuous performance, and if not, adding the Redis service to the fault-tolerant to-be-processed queue and skipping to the S2 for continuous performance; S4, closing the Redis service. The task failed in processing is retried continuously till processed successfully, the problem about data loss caused by interruption of network connection or Redis service is effectively solved, and smoothness in service performance is guaranteed.

Description

technical field [0001] The invention relates to a fault-tolerant processing method, in particular to a fast fault-tolerant processing method for failure of data writing into Redis. Background technique [0002] Redis is currently the most commonly used cache data storage system. Compared with cache storage systems such as Memcached, Redis not only supports publish / subscribe, master-slave and other deployment methods; it also supports more data types, and each data type provides Rich operating instructions. Redis middleware provides unified interface access services for other business systems. Master-slave and sentinel integrated service management can increase server fault tolerance under disasters, but because the service is provided through the network, communication may not be established correctly during use due to network signals or Redis itself. At this time, In-progress business data will be lost. Contents of the invention [0003] The purpose of the present inve...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a fast fault-tolerant processing method for the failure of data writing into Redis, to solve the problem that during the use of Redis service, communication may not be established correctly due to network signals or Redis itself, and the business data being executed will be lost question

Images