Intelligent split-brain prevention system and method for storage gateways

Abstract

The invention discloses an intelligent split-brain prevention system and method for storage gateways. The system comprises an application service layer, a network link layer, a storage management layer and a storage equipment layer; the application service layer is a client application system server and provided with various application systems and database software; the network link layer comprises an IP network and an FC optical fiber network; the storage management layer comprises two storage gateways which serve as nodes, when one node fails, the other node can work normally, an arbitrator is configured between the two storage gateways, and when one node fails, the arbitrator arbitrates and judges the states of the nodes to judge out the failed node; and the storage equipment layer comprises at least two storage arrays. Compared with the prior art, the intelligent split-brain prevention system and method for the storage gateways have the advantages that a three-stage split-brain judgment mechanism is proposed, the Active state and the Inactive state of the nodes can be effectively judged in a high-availability data protection scheme, and then the Primary node is determined to prevent the split-brain phenomenon.

Description

technical field [0001] The invention relates to the technical field of computers, in particular to an intelligent split-brain prevention system and method for a storage gateway. Background technique [0002] In the era of big data, data has become a vital core asset in various industries. Traditional disaster recovery solutions have problems such as low resource utilization, poor availability, long downtime when a failure occurs, slow data recovery, and high risks. Whether the data is safe and whether the business can run continuously without interruption becomes the key for users to evaluate a disaster recovery plan. Traditional data center storage disaster recovery generally adopts the active-standby mode. Only when the production data center storage fails, the disaster recovery center storage will take over the data access business, and this process needs to be performed manually. In addition, in the mode of active and standby data centers, only the main center plays a ...

AI Technical Summary

Problems solved by technology

This kind of disaster recovery system faces the following challenges: 1. When the production center encounters disasters such as power failure, fire, flood, earthquake, etc., the business needs to be manually switched to the disaster recovery center. The business interruption time is long, and the continuous operation of the business cannot be guaranteed.

2. The disaster recovery center cannot provide external services and is idle all the year round, with low resource utilization

Since they lost contact with each other, they all thought that the other party had failed. The HA software on the two nodes was like a "split-brain person", "instinctively" competing for "shared resources" and scrambling to start "application services". Serious consequences will occur: either the shared resources are divided, and the "services" on both sides cannot be started; online log error)

Images