Hard disk hang-up fault detection and processing methods and apparatuses for distributed storage system

Abstract

The invention discloses hard disk hang-up fault detection and processing methods and apparatuses for a distributed storage system. Whether a target hard disk has a hang-up fault or not is judged by detecting executive time of an access request corresponding to the target hard disk, so that the hang-up fault of the target hard disk can be timely discovered; after it is discovered that the target hard disk has the hang-up fault, a faulted hard disk is prevented from being accessed again through a state tag on one hand and system resources occupied by the faulted hard disk are cleaned up on the other hand, so that other processes can reallocate and apply the system resources, negative influence possibly caused by the hang-up fault of the hard disk is reduced, and the purpose of stopping loss is achieved. According to the hang-up fault detection and processing schemes provided by the methods and the apparatuses, a detection tool can be provided without depending on a hard disk manufacturer, new hardware does not need to be added to the hard disk, and no manual intervention is needed, so that the simplicity and easiness are achieved, and the production and use costs of the hard disk are not influenced.

Description

Technical field [0001] The invention relates to the field of computer technology, in particular to a method and device for detecting and processing a hard disk hanging fault of a distributed storage system. Background technique [0002] A distributed storage system is a storage system built on a local file system, which distributes data to multiple hard drives. For distributed storage systems, the entire link from the local file system to the internal hard disks may fail. Among them, the hard disk hang up failure, which means that the hard disk cannot respond to normal operations. The input and output operations cannot be stopped because the entire link does not respond. Improper handling of the hung hard disk may cause the entire access process to become unresponsive, resulting in inaccessibility of the data managed by the process, increased front-end request latency, increased system load, and decreased data availability. Therefore, it is a key issue to ensure the performance...

AI Technical Summary

Problems solved by technology

[0004] However, the above-mentioned processing methods have certain defects, including the need to rely on additional auxiliary tools, affecting the availability of system resources, etc.

Specifically, the above method (1) needs to rely on the tools provided by the hard disk manufacturer, and is not applicable to the situation that the hard disk cannot accept offline instructions, and the actual application success rate is low; method (2) needs to add new hardware to the hard disk (i.e. hardware switch), resulting in an increase in the cost of hard disk development and maintenance, and narrow scope of application; method (3) introduces human intervention, during machine restart, the availability of the machine itself and the storage system is reduced, and there is a possibility of restart failure, even if the restart Success also requires the storage system to be able to avoid the use of hung hard disks, which has higher requirements on the storage system; in method (4), the original process cannot release memory resources because of thread hangs, which makes the system memory usage high, even if it is restarted The available resources of the system will also be reduced

Images