Configuration file synchronization system and method based on memory and lightweight database

Abstract

The invention relates to a configuration file synchronization system and method based on a memory and a lightweight database, and relates to the technical field of data storage. The invention provides a lightweight and rapid configuration file synchronization system and method for a multi-node cluster or a distributed cluster, and through reasonable design of a memory database, a back-end database and a management module, a management service single node fault can be avoided, and the system management performance and reliability are improved; the node adding step is simplified, and the node adding difficulty is reduced.

Description

technical field [0001] The invention relates to the technical field of data storage, in particular to a configuration file synchronization system and method based on memory and a lightweight database. Background technique [0002] In cluster management based on local storage, the management and control of all nodes in the cluster is generally centralized and unified management through a single node, and the configuration data of the cluster is stored centrally on the management node. Cluster management fails. In cluster management based on distributed storage, configuration files can be stored in distributed systems, but for uninitialized distributed storage systems, configuration files cannot be stored or shared in the cluster. Furthermore, based on distributed systems For configuration file storage, the storage backend is too heavy and the configuration is complex. For new nodes that are about to join the cluster, a complete distributed storage system must be installed in...

AI Technical Summary

Problems solved by technology

[0002]In the cluster management based on local storage, the management and control of all nodes in the cluster are generally centralized and unified management through a single node, and the configuration data of the cluster is stored centrally in the management On the node, the unexpected failure of the management node in this way will cause the failure of the entire cluster management

In cluster management based on distributed storage, configuration files can be stored in distributed systems, but for uninitialized distributed storage systems, configuration files cannot be stored or shared in the cluster. Furthermore, based on distributed systems For configuration file storage, the storage backend is too heavy and the configuration is complex. For new nodes that are about to join the cluster, a complete distributed storage system must be installed in advance, and the cluster can only be joined after being authorized by the distributed storage system. Significantly prolong the joining time of nodes and increase the difficulty of joining nodes

Images