Method and device for detecting discs

Abstract

The embodiment of the invention discloses a method for detecting discs, comprising the following steps: calculating the response time the discs need to perform the data read-write operation when the discs finish the operation; making the discs offline when the response time exceeds the preset time threshold value; and detecting the off-line discs to judge whether the discs are out of work. The embodiment of the invention also discloses a disc detection device and a storage controller. The embodiment of the invention can distinguish disks which get out of work for the time being but can still be used from the discs really out of work.

Description

technical field [0001] The invention relates to the storage field, in particular to a disk detection method and device. Background technique [0002] A disk array combines a large number of disks to provide external storage services through a storage controller. The service host is connected to the disk array, and stores data in or reads data from the disks in the disk array. The business host has set a timeout period for each IO request, and if the execution time of the IO request expires, the execution will fail. When there is a disk failure in the disk array, the IO response is slow. It may happen that the business host sends ten IO request commands, but the disk only processes one in time, and the other nine IO request commands fail to execute. Then the nine IO request commands may be lost. In some business scenarios, if the real-time requirements of the business host are high, business interruption may also be caused. [0003] In the prior art, the disk that execute...

AI Technical Summary

Problems solved by technology

In some business scenarios, if the real-time requirements of the business host are high, it may also cause business interruption

[0003] In the prior art, the disk that executes the IO request timeout is usually determined as a failed disk, and a normal disk is used instead of the disk to continue working, and the disk may be due to an abnormal external environment, resulting in a short-term IO response slow, once the external environment If the disk returns to normal, the disk also returns to normal, so the existing technology does not distinguish this kind of disk from the disk that has actually failed, and may cause the disk that has only a short-term failure but can continue to be used to be suspended, which is not conducive to the normal operation of the array operation, it will also cause a waste of resources

Images