Method for automatic diagnosing system information

Abstract

A self-diagnostic method of the system information to monitor the state of the physics memory device in the system detects the bad devices according to the change of the state of the memory device. The method collects the all present state information of the physics memory device by the system management module, also it detects if the memory device is in the normal condition according to the comparing of the state information and the system log, the data information in the system management module, the normal information. The method can aim at many physic memory device such as the SATA / SCSI hard disk, the hard disk array and the memory device, especially it can detect the hard disk accurately and sensitively.

Description

technical field [0001] The invention relates to a method for automatically diagnosing system information, in particular to a method for diagnosing whether the condition of a physical storage device in an automatic monitoring system is abnormal. Background technique [0002] At present, in the known technology, the physical storage devices in the system, such as the hard disk, are monitored in an "active" way, that is, the working status of the hard disk is actively monitored through a program within a certain period of time, wherein Including: sending some hard disk commands, judging the health status of the hard disk by observing the status information returned from the hard disk; or directly performing I / O operations (input / output operations or read / write operations) on the hard disk to test the performance of the hard disk. The advantage of this method is that the diagnosis is directly targeted at the target hard disk, and the detection result is straightforward. But the...

AI Technical Summary

Problems solved by technology

[0003] 1. Each detection requires the program to actively monitor the physical storage device, and the program can only monitor the storage device on a regular basis. For hot-swappable / unpluggable physical storage devices, such as hot-swappable / unpluggable hard disks, it is necessary to Monitor its working status and performance at any time, so the monitoring method of the known technology has poor scheduling flexibility

[0004] 2, compared with equipment such as system bus, internal memory, the access speed of hard disk is relatively slow, and often can use multiple hard disks in server system, like this, known hard disk monitoring method can monitor each hard disk in parallel usually, therefore increased system Resource overhead, if you switch to serial monitoring of each hard disk, the monitoring cycle will be relatively extended, reducing monitoring efficiency

[0005] 3. The monitoring method of the known technology, its test itself will increase the load of physical storage devices such as hard disks, and long-term detection is not conducive to the health of the hard disk itself, and will reduce the service life of the hard disk

[0006] 4. The monitoring method of the known technology can only judge whether the storage device (hard disk) is normal or damaged by means of the execution result of the program, and false positives or false negatives often occur, which requires technicians to adopt other methods The status of the storage device (hard disk) is further verified, so the accuracy and sensitivity of its detection are relatively low

Images