JP2008234385A - Redundant system, server, information processor, control method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology capable of coping with a failure even if a HDD failure cannot be detected in a RAID device main body while a temporary HDD I/O access error can be detected by a counter server side in occurrence of a failure of a HDD inside a RAID device. <P>SOLUTION: A counter server interface part is arranged in the RAID device, and disconnecting operation of a designated HDD is carried out not only by failure detection and HDD disconnection processing determination inside the RAID device but also by determination by the counter server side on the outside. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a control technique when a failure occurs in an HDD, and more particularly, to a control technique performed by a device provided outside the device having the HDD when a failure occurs in the HDD.

  In general, a RAID (Redundant Array of Inexpensive Disk) device performs HDD failure monitoring on the RAID device main body, and when a failure is detected, it automatically detaches the failed HDD and automatically installs a spare HDD (for example, Patent Documents). 1).

However, depending on the content of the failure and the timing of various operations in the RAID device body, for example, while the opposing server side detects a temporary HDD I / O access error, the RAID device body has not detected an HDD failure. In some cases, a failure could not be detected completely on the RAID device side.
JP-A-9-319526

  The problem to be solved by the present invention is to improve the above-mentioned problem. A failure occurred in the HDD in the RAID device, and a temporary HDD I / O access error could be detected on the opposite server side. Therefore, it is to provide a technique capable of handling a failure even when an HDD failure cannot be detected in the RAID apparatus main body.

  The present invention for solving the above problems is a redundant system, comprising an information processing device having a storage device, and a server provided outside the information processing device, wherein the server is in the storage device. When a failure is detected, the information processing apparatus includes an instruction unit that instructs to disconnect the storage device, and the information processing apparatus includes a control unit that disconnects the storage device based on the instruction. And

  The present invention for solving the above problem is a server, and when a failure is detected in a storage device in an information processing device connected to the own device, the storage device is disconnected from the information processing device. It is characterized by instructing.

  The present invention for solving the above problem is a RAID device, and when an external device detects a failure of a storage device provided in the own device, the device is based on an instruction from the external device. The storage device is disconnected.

  The present invention for solving the above problem is a control method of a redundant system having an information processing apparatus having a storage device and a server provided outside the information processing apparatus, and detects a failure in the storage device A detection step; an instruction step for instructing the information processing apparatus to disconnect the storage device when the server detects the failure; and a disconnection step of disconnecting the storage device based on the instruction. Features.

  The present invention for solving the above-described problem is a server program, and when the program detects a failure in a storage device in an information processing apparatus connected to the server, the information processing apparatus And a process of instructing to disconnect the storage device.

  The present invention for solving the above problems is a RAID device program, wherein the program detects a failure of a storage device provided in the RAID device by an external device. The storage device is disconnected based on an instruction from the external device.

  In the present invention, an interface is provided on the RAID device side so that a specific HDD can be disconnected from the opposite server of the RAID device. When an abnormality such as an I / O access error is detected on the opposite server side, the HDD is specified by an instruction from the server. It is configured to separate and remove the suspected part. By implementing such an interface on the RAID side, a failure according to the reliability required for the system, such as a process of disconnecting a suspected HDD with a single access error or a process of disconnecting a suspected HDD with multiple errors. The detection function can be flexibly realized on the server side.

  According to the present invention, in the present RAID device, when an abnormality occurs on the opposite server side but an HDD failure cannot be detected on the main body of the RAID device, the suspected HDD can be disconnected by an instruction on the server side. Effective fault detection becomes possible.

  Further, according to the present invention, the failure detection function can be easily enhanced only by adding processing to the server connected to the RAID device.

  The RAID device of the present invention has an interface capable of executing a failure detection process for an HDD mounted on a RAID device and a disconnection process for the failed HDD from an opposing server connected to the RAID device.

  By realizing such a RAID device, it is possible to realize more fault detection functions than those possessed by the RAID device by adding functions to the server side, and flexibly fail according to the required system reliability. The detection function can be expanded.

  In order to explain the features of the present invention, it will be specifically described below with reference to the drawings.

  A first embodiment for carrying out the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a configuration diagram of a redundant system according to the present invention.

  The redundant system of the present invention includes a RAID device 1 and a counter server 7.

  The RAID device 1 includes an HDD (Hard Disk Drive) 2 and a controller 3.

  The HDD 2 holds various data.

  The controller 3 includes an I / O access processing unit 4, a failure detection / separation processing unit 5, and a counter server interface unit 6.

  The I / O access processing unit 4 performs data read / write (read / write) access processing between the opposing server 7 and the HDD 2.

  The failure detection / separation processing unit 5 detects a failure related to the HDD 2. Further, when some trouble occurs in the HDD 2, the HDD 2 is disconnected.

  The opposing server interface unit 6 performs a connection process with the failure detection / separation processing unit 5 in order to enable the HDD 2 to be disconnected from the external opposing server 7.

  The opposing server 7 is provided outside the RAID device 1 and is connected to each other via a network. The opposing server 7 includes a control unit that performs a process of disconnecting the HDD 2 when a failure related to the HDD 2 is detected.

  Next, the operation of the present invention will be described with reference to the drawings. In the following description, the case of performing data read processing will be described, but data write processing may be used. Further, the opposite server 7 will be described using a case where the HDD 2 is disconnected by a single access error, but it is also different if the HDD 2 is disconnected by a plurality of errors.

  FIG. 2 is a time chart for explaining the operation of the present invention.

  The opposing server 7 performs data read processing on the HDD 2.

  The I / O access processing unit 4 receives a data read process for the HDD 2 of the opposite server 7.

  The I / O access processing unit 4 accesses the HDD 2. At this time, if some failure occurs and the I / O access processing unit 4 cannot access the HDD 2, the opposite server 7 is notified that the failure has occurred.

  The opposing server 7 transmits an HDD disconnection request to the opposing server interface unit 6.

  The opposing server interface unit 6 receives the HDD disconnection request from the opposing server 7 and notifies the failure detection / separation processing unit 5 to perform processing for disconnecting the HDD 2.

  The failure detection / separation processing unit 5 discards the notification from the opposing server interface unit 6 when the failure has already been detected and the HDD disconnection processing has ended, and when the failure has not been detected, Based on the notification from the opposite server interface unit 6, the HDD 2 is disconnected.

  On the other hand, when the I / O access processing unit 4 receives data read processing for the HDD 2 of the opposing server 7 and accesses the HDD 2, the I / O access processing unit 4 can read the data normally without causing a failure. Receives the data read by the I / O access processing unit 4 and performs normal operation.

  As described above, the feature of the present invention is that the opposite server interface unit 6 is provided in the controller 3 and the specified HDD 2 is disconnected not only by the determination of the failure detection / disconnection processing unit 5 but also by the determination of the external opposite server 7 side. This is a possible point.

  In the above description, the case where the server 7 detects a failure that has occurred in the data read / data access operation has been described. However, if the failure can be detected by the opposite server 7, a failure that has occurred during another operation is detected. It may be the case.

  The RAID device and the opposite server of the present invention described above can be configured by hardware as is apparent from the above description, but can also be realized by a computer program.

  In this case, functions and operations similar to those of the above-described embodiment are realized by a processor that operates according to a program stored in the program memory. Note that only a part of the functions of the above-described embodiment can be realized by a computer program.

FIG. 1 is a block diagram for explaining the configuration of the present invention. FIG. 2 is a time chart for explaining the operation of the present invention.

Explanation of symbols

1 RAID device 2 HDD (Hard Disk Drive)
3 Controller 4 I / O access processing unit 5 Fault detection / separation processing unit 6 Counter server interface unit 7 Counter server 7

Claims (10)

A redundant system,
An information processing apparatus having a storage device;
A server provided outside the information processing apparatus,
The server includes an instruction unit that instructs the information processing device to disconnect the storage device when a failure is detected in the storage device.
The redundant system according to claim 1, wherein the information processing apparatus includes a control unit that disconnects the storage device based on the instruction.   The redundant system according to claim 1, wherein the control unit disconnects the storage device based on the instruction when a failure in the storage device cannot be detected by the own device.   The redundant system according to claim 1, wherein the information processing apparatus is a RAID apparatus.   The redundant system according to claim 1, wherein the server detects the failure when reading or writing data held by the storage device. A server,
A server characterized by instructing the information processing device to disconnect the storage device when a failure is detected in the storage device in the information processing device connected to the own device.   6. The server according to claim 5, wherein the failure is detected when reading or writing data held in the storage device. A RAID device,
A RAID device characterized in that, when a device provided outside detects a failure of a storage device provided in the device itself, the storage device is disconnected based on an instruction from the external device. A redundant system control method comprising an information processing apparatus having a storage device and a server provided outside the information processing apparatus,
A detection step of detecting a failure in the storage device;
An instruction step for instructing the information processing device to disconnect the storage device when the server detects the failure;
And a disconnecting step of disconnecting the storage device based on the instruction. A server program, the program on the server,
A program that, when detecting a failure in a storage device in an information processing device connected to the own device, causes the information processing device to instruct to disconnect the storage device. A RAID device program, the program being stored in the RAID device,
A program that, when a device provided externally detects a failure of a storage device provided in its own device, causes the storage device to be disconnected based on an instruction from the external device.

Images