JPS6160154A - Control system for access to memory - Google Patents

Abstract

PURPOSE:To detect early the fault of a defective record and also to prevent previously the generation of the head crush, by using plural memories successively and in an on-line system. CONSTITUTION:Both VOL0 and VOL1 are used with switching every fixed period of time. If an error is detected with a reading action out of the VOL1, a fault flag is turned on with an on-volume. Then an on-line switch flag 3 is changed over with an on-line switch change over program. Then the volume turned on the fault flag is formed again with exchange, etc. Thus the volume repeats the normal actions and starts a patrol for detection of faults.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an access control method to a storage device that can detect failures of duplexed or multiplexed magnetic disk devices at an early stage.

In order to increase the reliability of magnetic disk drives used in information processing devices, duplication is implemented, recognizing that redundancy increases. If this redundancy fully achieves its intended purpose, no problems will occur, but with conventional redundancy systems, there are factors that can cause unexpected failures at the beginning. However, situations may arise where the purpose of duplication is not fully achieved.

[Conventional technology]

In conventional duplication systems, one of the systems is used online and the other is operated as a backup system, and when a failure occurs in the online system, the online system and the backup system are switched to reduce the effects of the failure. In such an operation, if the online system is operating normally, the backup system will not be accessed at all, or a head seek will be performed periodically, or a specific record will be read periodically. It is. This latter procedure is intended to ensure the normal operation of the backup system.

[Problem that the invention seeks to solve]

The normal operation guarantee means of the backup system described above does not access the records used in regular im work, so even if a malfunction such as a defect occurs in the record, its discovery will be delayed. Therefore, there is a possibility that it will develop into a serious problem when used as an online system.

Means for Solving Problem c] The present invention provides an access control method to a storage device that can solve the above problem, and the means is connected to a central processing unit and receives access from the central processing unit so that the same storage content can be stored. In an information processing device having a plurality of storage devices storing , a switching means for instructing one of the plurality of storage devices that can be accessed by the central processing unit, and one of the plurality of devices 1a. The storage device is configured to include means for controlling the switching means so that the storage devices are accessed by the central processing unit in a predetermined order for a predetermined period of time, so as to perform a failure patrol of the plurality of storage devices. be.

[Effect]

According to the method of the present invention, a plurality of storage devices are connected to the central processing unit one at a time at regular intervals in a predetermined order, that is, the plurality of storage devices are sequentially used online. , its operation can be confirmed, and malfunctions such as bad records can be discovered at an early stage, and the possibility of occurrence of a failure can be prevented.

〔Example〕

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows an embodiment of the invention. In this figure, ■
is a central processing unit (hereinafter abbreviated as CPU), to which the main memory 2 is connected as well as disk controllers DKCO and DKCI, and to this disk controller is a magnetic disk device (hereinafter abbreviated as volume). ) VOLO and VOLI are connected. In addition,
A disk controller may be shared by each volume.

The main memory 2 stores an access program for accessing volumes vot, o or VOLI, and an online switch switching program.
Online switch flag 3 and volume V act as an online switch that allows the access program to access volume OI, 0 or volume VOL1.
Fault flag for OLO4. and volume VOL
There is an area for a fault flag 42 for 1.

Next, a detailed explanation of the present invention under the above-mentioned configuration system will be given.

'' In the dual configuration system, if neither volume VOLO nor volume VOLI has a failure and is operating normally, the processing flow of the online switch changeover program shown in FIG. S2.
Through S5, the online volume is switched between volume VOLO and volume VOL1 at predetermined time intervals. The operation sequence of the online switch switching program, online switch flag, and online volume in this case is shown in FIG. During such operations, as is clear from FIG. 3, the CPU refers to the online switch flag, and the access program accesses the online volume via the disk controller DKCO or DKCI in accordance with the instruction of the flag.

If such an access occurs, for example, when reading from volume VOLO, and an error occurs during reading (Y in step S2 in Figure 5) (same for writing, see Figure 6), online volume failure occurs. A flag (in this case, flag 41) is turned on (step S3 in FIG. 5).

Then, in the next online switch switching program processing flow, the determination in step S2 in FIG. 2 becomes affirmative (Y in step S2), and it is determined in step S3 in FIG. 2 whether or not the spare volume failure flag is on. It will be judged. In this case, if the spare volume failure flag is off, the online switch flag 3 is switched to switch the spare volume online (step S4 in FIG. 5). Then, in step S1 in which the online switch switching program is executed, the online switch flag 3 is switched again, but the second
At step S4 in the figure, flag 3 is switched to a state indicating that the spare volume should be used online within a short time without causing any operational problems. FIG. 4 shows such an operation sequence.

When volume duplication is re-established by replacing the volume with the failure flag turned on, the normal operation described above will be repeated, and a patrol will be started for failures that may occur in any of the volumes. Become.

In the second embodiment, a case of duplication has been described, but the invention is not limited to this.

〔Effect of the invention〕

As explained above, according to the present invention, it is possible to (1) detect failures in standby storage devices at an early stage and minimize the development of serious failures caused by such failures; (2) also cause head crashes to occur. Effects such as being able to prevent such problems from occurring can be obtained.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an embodiment of the present invention, FIG. 2 is a flowchart of an online switch switching program, and FIG. 3 is a diagram showing the operation sequence of the online switch switching program, online switch flag, and online volume in normal operation. , FIG. 4 is an online switch changeover program when a read error occurs. FIG. 5 is a flowchart of the volume access program for reading, and FIG. 6 is a flowchart of the volume access program for writing. In the figure, 1 is the CPU, 2 is the main memory, 3 is the online switch flag, 41.42 is the failure flag, DKCO,
DKCI is a disk controller, VOLO, VOLl
is a magnetic disk device. Figure 5 Figure 6 n0

Claims (1)

[Claims] In an information processing device having a plurality of storage devices that are connected to a central processing unit and receive access from the central processing unit and store the same storage contents, switching that instructs one of the plurality of storage devices that the central processing unit can access. and means for controlling the switching means to make one of the plurality of storage devices available for access by the central processing unit in a predetermined order and for a predetermined period of time. 1. A method for controlling access to a storage device, characterized in that the method is configured to perform the following.