Cache service recovery method, system and device based on cloud platform

Abstract

The invention discloses a cache service recovery method, system and device based on a cloud platform, wherein the cache service comprises a plurality of cache instances; the method comprises the following steps: on a cluster of the cache service, sequentially deleting data volumes and instance resources of the cache instances on a downtime server node; reconstructing the cache instance to obtain a first cache instance; scheduling the first cache instance into a redundant server node; starting the first cache instance in the redundant server node to obtain a reconstruction instance; associating the reconstruction instance with a corresponding copy, and synchronizing copy data; and after the copy data is synchronized, obtaining a recovery instance corresponding to the cache instance on the downtime server node. According to the invention, by deleting part of related data of a cache instance on a downtime server node on a cluster, after reconstruction, starting through a redundant server node, and synchronizing copy data, the recovery instance is obtained corresponding to the cache instance on the downtime server node, and after recovery, the original high availability is still maintained.

Description

technical field [0001] The present invention relates to the technical field of cloud computing, in particular to a method, system and device for restoring a cache service based on a cloud platform. Background technique [0002] Driven by cloud native, deploying applications to cloud platforms is an irreversible trend. In the case of few changes to existing business codes, allowing distributed systems to seamlessly enter the cloud has become a key task for cloud transitions, and cloud-native middleware is a key issue in supporting cloud transitions. Middleware usually includes services, functions Computing, microservice system, messaging, etc. Elasticity and high availability are important indicators of the cloud-native environment. Through fast and elastic reconstruction and keeping the system available, the cloud online system is not affected by failures in the environment. [0003] The middleware-cache service deployed on the cloud-native platform usually runs as a clust...

AI Technical Summary

Problems solved by technology

However, before the downtime server node recovers, the cache service will lose its high availability capability, and the cache service is in a relatively unstable state. Therefore, it is necessary to restore the downtime server node and start the cache instance as soon as possible. If the downtime server node and its cache instance Failure to recover in a short period of time or multiple downtimes will affect the high availability of the cache service

Images