JPS63160071A - Automatic data transfer system from defect track data to substitute track - Google Patents

Abstract

PURPOSE:To reduce mis-log information supervision caused by a person by detecting the propriety of alternative track during I/O request to a magnetic disk, assigning the alternative track while it is readable and transferring information so as to prevent the destruction of information in advance. CONSTITUTION:The erroneous log content of the device is loaded and a CPU uses the value (b) of the supervisory rule list 21 of a memory device 20 as to whether the error is due to the track face. When the error exists and a frequency function gm obtaining an mis-state function (f) is initialized to the frequency accumulated value M in the case of new track address, where (g) is the operator for initial setting. When the function exists already, the function gm is obtained from various information relating to the latest error on the management list 22 to calculate the total sum M. In the case of M>Mmax, the alternative track request is given. The erroneous track address is scanned, the track is closed and the information is transferred to the alternative track, the defect is set to release the blocking. The defect track is eliminated from the management list 22 finally.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to the operability of a computational wall system, and in particular to predicting defective addresses in magnetic disk drives, automatically allocating alternative tracks, and automatically migrating data. Regarding the method.

Conventional dynamic defective track processing uses the track surface inspection function in the volume initialization program to determine the degree of defect, and then allocates a replacement track to the track determined to be defective. , automatic data migration was not realized.

Problems to be Solved by the Invention Magnetic disk drives are central to data and program storage in computer systems, and with their increasing density and capacity, proactive measures such as monitoring error log information are becoming increasingly important. Due to the increase in the number of devices and the number of devices, there was a drawback in terms of man-hours for human analysis of error log information.

The present invention has been developed in order to solve the above-mentioned problems inherent in the conventional technology. Therefore, an object of the present invention is to analyze, predict, and analyze the track status of a magnetic disk drive of a customer system as a system function. By automatically executing these measures, we aim to improve the reliability and operability of the customer's system, as well as reduce the number of man-hours required for maintenance.

Means for Solving the Problems In order to achieve the above object, the data automatic migration method of defective track data to an alternative track according to the present invention inputs the intermittent error state of the magnetic disk drive of the customer system from error log information. Then, the contents of the error log are analyzed, and when the error is caused by the track surface, the past error status of the corresponding error address is checked, and it is determined whether or not it should be treated as a defective address. If it is a defective address, an alternative address is selected. This enables the data on the corresponding error address to be migrated to an alternative track while the track is being prepared and a retry is possible.

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to FIG. 1 and a preferred embodiment thereof.
This will be explained in detail using the first page, FIG. 3, and the reference figures.

FIG. 1 is a system block diagram showing one embodiment of the present invention.

In FIG. 1, reference number 10 indicates a central processing unit;
The central processing unit IQ is equipped with a storage device, and the storage device is equipped with a magnetic disk device monitoring rule table and a magnetic disk management table. The target device is a magnetic disk device, 7/, JJ, , 7.7.

Table 1 shows a magnetic disk device monitoring rule table 2/, and Table 1 is a magnetic disk management table.

Table 1 Magnetic disk device monitoring rule table Chapter - Table Magnetic disk management table.

On the magnetic disk device monitoring rule table l in Table 1, information on the error status bit position due to the track surface is shown in bl,
b2. . . . are prepared as selection conditions for errors caused by the track surface, which is the subject of the present invention. The error state function is a function whose parameters are the occurrence time interval and the IO request interval, and the error state frequency function gm determined by this function is determined. g is an operator such as addition or initialization, and m is a fixed value for each error state function fK.

The error state frequency cumulative limit value M max is the allowable limit value of the cumulative error state frequency value M explained in the magnetic disk management table of Table 2, and is a value for determining alternative track processing.

The magnetic disk management table in Table 1 is managed for each target error track address, and the error state frequency is determined by the latest elvish occurrence time, the number of work 0 requests at that time, and the function f calculated from them. The function gmt cumulatively calculates 9M (=Σgm) and has an alternative track processing request status indicating whether M>Mmax.

FIG. 1 is a process flow diagram of a program that detects the necessity of alternative track processing. When there is an error in the magnetic disk device and its contents are logged, this process also starts.
This is achieved by the following processing. Load the error log contents of the magnetic disk device (Step 101), and check whether the error state status therein is an error state caused by the track surface targeted for this process k)1-b2+...
If the error does not apply, the process ends, but if it is the target of this process, check whether the corresponding track address is the first error or an existing error (step 10.1). If it is a new track address, find its error state function f, and calculate the corresponding error state frequency function gm as the cumulative error state frequency value M.
Initialize K (steps 10u, 10!r).

In other words, g in the new case is an initial setting operator.

If it already exists, the latest error occurrence time hms on the magnetic disk management table in Table 1, the number of IO requests at the time of the latest error occurrence, and the latest error occurrence time hmsi on the error log.
, function f from IO bad request number 1 when the latest error occurs
and find the error state frequency function gm.

A cumulative error state frequency value M of the corresponding track address is calculated (step 10). If the calculation result M is larger than the limit value M max for accumulating error status B on the magnetic disk device monitoring rule table 1 in Table 1, the alternative track processing request is set to ON (step 101).
, requests activation of alternative track processing (FIG. 3) (step tOW).

FIG. 3 shows a process flow diagram of the alternative track processing program.

Next, referring to FIG. 3, in response to a request to start the alternative track processing program a program shown in %8g (step 10), this processing is started, searches for an error track address with a 1 alternative track processing request, and searches for the corresponding error track address. Prepare a replacement track for the track (step=01).

When preparations for data migration are completed, the track is blocked and access from other processing programs is prohibited (step 0I2).

The data of the track is read and written to the alternative track (step oJ), and after data migration is completed, the track is set as a defective track and the blockage is released (step oII). Finally, the track address is deleted from the second magnetic disk management table (step - ○).

Effects of the Invention As is clear from the above description, according to the present invention, it is possible to detect whether a track needs to be replaced during 10 requests to a normal magnetic disk, and to replace the track before it becomes a fixed error, that is, before it becomes readable. Because it is possible to allocate an alternative truck to the system and move the data, data destruction can be prevented and business downtime can be significantly improved.

Furthermore, the effect of reducing the need for monitoring error log information by humans can also be obtained.

[Brief explanation of the drawing]

FIG. 1 is a system block configuration diagram showing an embodiment of the present invention. FIG. 1 is a flowchart showing an outline of a program that detects whether or not to perform alternative track processing. FIG. 2 is a flowchart showing an outline of a program to be implemented. 10...Central processing unit, J...Storage device, -/...
・Magnetic disk device monitoring rule table %-...Magnetic disk management table, 7/,,? Yu, 33... Magnetic disk device patent applicant Representative of NEC Field Service Co., Ltd. Patent attorney Yutabe Kumagai Figure 1, Figure 2F5! J Figure 3

Claims (1)

[Claims] In a computer system that has error log information for magnetic disk units, a specific address on the magnetic disk unit is determined to be a bad track before an input/output error that cannot be retried occurs, and an alternative track is allocated and data is migrated through human intervention. An automatic data migration method for defective track data to an alternative track, which is characterized in that it is realized without the need for data transfer.