JP2010134696A - Raid controller device, processing method, raid controller circuit and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform restoration without replacement of a hard disk in the event of false detection of a hard disk failure even if there is no failure in the hard disk. <P>SOLUTION: A failure detection unit 11 detects a failure which is caused in hard disk drives 20-1 to 20-N. A restart unit 14 restarts the RAID controller device 10 when the failure detection unit 11 detects a plurality of failures of the hard disk drives 20-1 to 20-N within a predetermined time, and a reconfiguration unit 16 reconfigures, when restarting the RAID controller device 10, a logic drive based on logic drive configuration information stored by the hard disk drives 20-1 to 20-N. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a RAID controller device, a processing method, a RAID controller circuit, and a program that operate an array of a plurality of redundant hard disks as one virtual storage means.

  Conventionally, a technique called RAID (Redundant Array of Inexpensive Disks) is used to prevent data loss due to a failure of a storage medium such as a hard disk and to improve input / output processing performance (see, for example, Patent Document 1). ). RAID is a technology for operating a plurality of redundant hard disks virtually as a single hard disk. By making data redundant, even if a hard disk fails, there is no loss of data as a whole. Technology.

Normally, when RAID is applied to a computer, a device called a RAID controller performs data configuration and management of the hard disk. When a hard disk fails, the RAID controller detects a hard disk failure, logically separates the failed hard disk, and operates the computer using another hard disk. When the user replaces the disconnected hard disk with a hard disk that has not failed, the RAID controller reproduces the information on the original hard disk from the information on the other hard disks, and writes the reproduced information on the replaced hard disk. Perform reconfiguration.
JP 2002-373059 A

However, in the case of a conventional RAID controller, if the hard disk cannot be accessed due to a temporary abnormality in the communication path between the RAID controller and the hard disk even though the hard disk has not failed, the hard disk has failed. In some cases, incorrect judgments were made. In this case, the user needs to replace all the hard disks in which the failure is detected and perform data reconstruction even though the hard disk has not failed, and to perform a long recovery operation. It was.
The present invention has been made in view of the above points, and an object of the present invention is to recover without replacing a hard disk when a hard disk failure is detected erroneously even though the hard disk has not failed. A RAID controller device, a processing method, a RAID controller circuit, and a program are provided.

  The present invention has been made to solve the above problems, and is a RAID controller device that combines a plurality of hard disks that store configuration information of virtual storage means in advance and operates as one logical virtual storage means, A failure detecting means for detecting a failure of each of the plurality of hard disks; a restarting means for restarting the apparatus when the failure detecting means detects a failure of a plurality of hard disks within a predetermined time; and Reconstructing means for reconstructing the virtual storage means based on the configuration information.

  Further, the present invention is a processing method using a RAID controller device that combines a plurality of hard disks that store configuration information of virtual storage means in advance and operates as one logical virtual storage means, wherein the failure detection means A failure detecting unit detects a failure of each of the plurality of hard disks, and the restarting unit restarts the apparatus when the failure detecting unit detects a failure of the plurality of hard disks within a predetermined time. The virtual storage means is reconstructed based on the configuration information.

  The present invention also provides a RAID controller circuit that combines a plurality of hard disks that store virtual storage unit configuration information in advance and operates as a single logical virtual storage unit, and detects a failure of each of the plurality of hard disks. A detection circuit, a restart circuit for restarting the apparatus when the failure detection means detects a failure of a plurality of hard disks within a predetermined time, and the virtual storage means based on the configuration information at the time of restart. And a reconstructing circuit for restructuring.

  Also, the present invention provides a RAID controller device that combines a plurality of hard disks that store virtual storage unit configuration information in advance and operates as a single logical virtual storage unit, and detects failure of each of the plurality of hard disks. A restart unit that restarts the device when the failure detection unit detects a failure of a plurality of hard disks within a predetermined time, and a reconfiguration that reconstructs the virtual storage unit based on the configuration information at the time of restart. It is a program for operating as construction means.

  According to the present invention, the restarting means restarts the RAID controller device when a plurality of hard disk failures are detected within a predetermined time. Since it is unlikely that multiple hard disks will fail at close timing, in this case, there may be a cause other than hard disk failure, such as a temporary failure in the communication path between the failure detection means and the hard disk due to noise contamination. high. Therefore, by restarting the RAID controller device, the rebuilding unit can attempt to restore the virtual storage unit by rebuilding the virtual storage unit based on the configuration information.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a schematic block diagram showing the configuration of a RAID controller device according to an embodiment of the present invention.
The RAID controller device 10 is connected to a plurality of hard disk devices 20-1 to 20-N via a connection path 3 (bus). Further, the RAID controller device 10 constructs a logical drive (virtual storage means) by combining the hard disk devices 20-1 to 20-N, and causes the computer 1 to recognize and operate the logical drive.
The RAID controller device 1 also includes a failure detection unit 11, a failure detection pattern monitoring unit 12, a hard disk information storage unit 13, a restart unit 14, a restart count storage unit 15, and a reconstruction unit 16. Prepare.
The failure detection unit 11 detects a failure that has occurred in the hard disk devices 20-1 to 20 -N via the connection path 30.
The failure detection pattern monitoring unit 12 determines whether a plurality of failures of the hard disk devices 20-1 to 20-N are detected within a predetermined time.
The hard disk information storage unit 13 stores the identification numbers of the hard disk devices 20-1 to 20-N that have detected the failure in association with the time at which the failure was detected.
The restart unit 14 restarts the RAID controller device 1.
The restart number storage unit 15 stores a restart number indicating the number of times the restart unit 14 restarts the RAID controller device 1.
The reconstruction unit 16 reconstructs the logical drive based on the logical drive configuration information stored in the hard disk devices 20-1 to 20-N.
In addition, the hard disk devices 20-1 to 20 -N include a hard disk 21 and a control unit 2222.
The hard disk 21 stores configuration information of the logical drive such as RAID specifications and data stored in the logical drive in advance.
The control unit 2222 detects a failure of the hard disk 21.

In the RAID controller device 1 of the present embodiment, the failure detection unit 11 detects a failure that has occurred in the hard disk devices 20-1 to 20-N, and the failure detection unit 11 includes the hard disk devices 20-1 to 20-N. When a plurality of failures are detected within a predetermined time, the restart unit 14 restarts the own device, and the rebuilding unit 16 stores the logical drives stored in the hard disks 21 of the hard disk devices 20-1 to 20-N at the time of restart. Rebuild the logical drive based on the configuration information.
As a result, when the hard disk failure is erroneously detected even though the hard disk devices 20-1 to 20-N have not failed, the RAID controller device 1 restores the logical drive without replacing the hard disk. I do.

Next, the operation of the RAID controller device 1 will be described.
FIG. 2 is a flowchart showing the operation of the RAID controller device.
First, the failure detection unit 11 communicates with the hard disk devices 20-1 to 20-N, and determines whether there is a failure in the hard disk devices 20-1 to 20-N (step S1).
The determination of the presence or absence of a failure is performed as follows, for example.
The failure detection unit 11 transmits a failure detection signal to the control unit 22 of each of the hard disk devices 20-1 to 20-N via the connection path 30. When the control unit 22 of each of the hard disk devices 20-1 to 20-N receives the failure detection signal, it determines the failure of the corresponding hard disk 21.
As a result of the failure determination, the control unit 22 transmits a response signal to the failure detection signal to the failure detection unit 11 when it is determined that no failure has occurred in the hard disk 21. On the other hand, when it is determined that a failure has occurred in the hard disk 21 as a result of the failure determination, the control unit 22 does not transmit a response signal for the failure detection signal to the failure detection unit 11.
When the failure detection unit 11 receives a response signal from the control unit 22 of the hard disk devices 20-1 to 20-N, the failure detection unit 11 determines that no failure has occurred, and the control unit 22 of the hard disk devices 20-1 to 20-N. If a response signal cannot be received from the mobile phone, it is determined that a failure has occurred.
At this time, even when a temporary abnormality occurs in the connection path 30 due to noise or the like, the failure detection unit 11 cannot receive the response signal, and thus the hard disk devices 20-1 to 20-20 that have not received the response signal. -N determines that a failure has occurred.

If the failure detection unit 11 determines in step S1 that no failure has occurred in all the hard disk devices 20-1 to 20-N (step S1: NO), the RAID controller device 10 returns to step S1, Subsequently, the failure detection of the hard disk devices 20-1 to 20-N is continued. The failure detection is performed at predetermined intervals.
On the other hand, when it is determined that a failure has occurred in at least one of the hard disk devices 20-1 to 20-N (step S1: YES), the failure detection unit 11 determines that a failure has occurred. -1 to 20-N are logically separated from the RAID configuration (step S2).

When the failure detection unit 11 disconnects the hard disk devices 20-1 to 20-N, the failure detection pattern monitoring unit 12 and the identification numbers of the hard disk devices 20-1 to 20-N that detected the failure from the failure detection unit 11 The time when the failure is detected is acquired (step S3).
When the failure detection pattern monitoring unit 12 acquires the identification number and the time, the failure detection pattern monitoring unit 12 compares the time when the failure of the other hard disk devices 20-1 to 20-N stored in the hard disk information storage unit 13 is detected with the acquired time. Then, it is determined whether or not there is a hard disk whose failure has already been detected within a predetermined time (step S4). Here, the predetermined time is, for example, a short time such as 10 milliseconds or 1 second, in which a failure occurs in a plurality of hard disk devices 20-1 to 20-N within the time, and failure detection is performed. That is longer than the time difference until a failure of each of the hard disk devices 20-1 to 20-N is erroneously detected due to an abnormality in the connection path 30 between the unit 11 and the hard disk devices 20-1 to 20-N And

If the failure detection pattern monitoring unit 12 determines that there is no hard disk device 20-1 to 20-N in which a failure is detected within a predetermined time (step S4: NO), the hard disk device 20 that has detected the failure in step S3. The identification numbers of −1 to 20-N and the time when the failure is detected are associated and registered in the hard disk information storage unit 13 (step S5). Here, the reason why the identification numbers and times of the hard disk devices 20-1 to 20-N are registered is that when the failure detection unit 11 detects the failure of the hard disks 20-1 to 20-N next time in step S1, the failure is detected. In step S4, the detection pattern monitoring unit 12 detects the failure of the hard disk devices 20-1 to 20-N that detected the failure and the failure of the hard disk devices 20-1 to 20-N that detected the failure this time. This is to compare the time.
When the failure detection pattern monitoring unit 12 registers the identification numbers of the hard disk devices 20-1 to 20-N and the time when the failure is detected, the RAID controller device 10 returns to step S1 and continues to the hard disk devices 20-1 to 20-. Continue detecting N failures.
When the failure detection pattern monitoring unit 12 determines that there is a hard disk device 20-1 to 20-N in which a failure is detected within a predetermined time (step S4: YES), the RAID controller device is processed by a process described later. 10. When the failure detection pattern monitoring unit 12 determines that there is no hard disk device 20-1 to 20-N in which a failure is detected within a predetermined time (step S4: NO), the restart is performed. Do not perform startup. The reason for this will be described below. If no failure of the other hard disk devices 20-1 to 20-N is detected within a predetermined time, the hard disk devices 20-1 to 20-N that have detected the failure in step S1 have been detected independently. . For this reason, there is a low possibility that a failure is erroneously detected due to a temporary abnormality in the connection path 30 due to noise or the like, and there is a possibility that the hard disk devices 20-1 to 20-N where the failure is detected actually have a failure. high. Since the restart is performed to reconnect to the hard disk devices 20-1 to 20-N that have been disconnected due to a temporary abnormality in the connection path 30, the hard disk devices 20-1 to 20-N are connected. If there is an actual failure, there is no need to perform a restart. This is the reason why the restart is not executed.

When the failure detection pattern monitoring unit 12 determines that there is a hard disk device 20-1 to 20-N in which a failure is detected within a predetermined time (step S4: YES), the restarting unit 14 is a restart count storage unit. It is determined whether or not the number of restarts stored in 15 is less than a predetermined maximum number of repetitions (predetermined number) (step S6). As described above, the number of restarts indicates the number of times that the restart unit 14 restarts the RAID controller device 1.
If the restarting unit 14 determines that the restart count is equal to or greater than the maximum number of repeats (step S6: NO), the RAID controller device 10 returns to step S1 and continues to fail the hard disk devices 20-1 to 20-N. Continue detection. The reason why the restart is not executed when it is determined that the number of restarts is equal to or greater than the maximum number of repetitions will be described later.
If the restarting unit 14 determines that the restart count is less than the maximum repeat count (step S6: YES), the restart unit 14 restarts the RAID controller device 10 (step S7). Here, when the failure detection unit 11 erroneously detects the occurrence of a failure due to a temporary abnormality in the connection path 30 due to noise mixing or the like, the temporary abnormality is resolved when the RAID controller device 10 is restarted. If this occurs, communication with the hard disk devices 20-1 to 20-N that have been determined to have failed can be performed normally, and the failure is not detected. Therefore, the restart is performed in step S7.
When the restart unit 14 executes the restart, the restart unit 14 rewrites the restart number stored in the restart number storage unit 15 to a value obtained by adding 1 to the restart number (step S8). However, if the restart performed by the restart unit 14 is the first time, and the restart count storage unit 15 does not store the restart count, the restart unit 14 stores the restart count storage unit 15. It is assumed that “1” is registered as the number of restarts.

When the restart unit 14 rewrites the number of restarts, the failure detection unit 11 communicates with the hard disk devices 20-1 to 20-N determined to have detected the failure within a predetermined time in step S4, and whether or not there is a failure. Is determined (step S9). At this time, there is no failure in the hard disk devices 20-1 to 20-N determined to have detected the failure within the predetermined time in step S4, and the determination in step S1 is a temporary abnormality in the connection path 30 due to noise mixing or the like. In the case of an erroneous determination due to the above, there is normally a high possibility that communication is normally performed by restarting the RAID controller device 10. Therefore, in step S9, the failure detection unit 11 communicates with the hard disk devices 20-1 to 20-N that are determined to have detected the failure within the predetermined time in step S4, and determines whether or not there is a failure. It is determined whether or not communication has been started.
For example, when the failure detection pattern monitoring unit 12 determines in step S4 that a failure has occurred in the hard disk device 20-1 and the hard disk device 20-2 within a predetermined time, the failure detection unit 11 in step S9 It is determined whether there is a failure in the hard disk device 20-1 and the hard disk device 20-2. The determination of the presence or absence of the failure is performed by the same process as the determination executed in step S1.

In step S9, when the failure detection unit 11 determines that all the hard disk devices 20-1 to 20-N determined to have detected the failure within the predetermined time in step S4 have a failure (step S9: YES) Returning to step S6, restart is executed. The reason for returning to step S6 is as follows.
When there is an actual failure in the hard disk devices 20-1 to 20-N determined to have detected the failure within the predetermined time in step S4, or when a failure such as disconnection of the connection path 30 has occurred. Since steps S7 and S8 are repeated, the restart will continue to be executed. Therefore, when the number of restarts exceeds the maximum number of repetitions in step S6, the determination of the failure occurrence by the failure detection unit 11 is not an erroneous determination due to a temporary abnormality in the connection path 30, but the hard disk device 20-1 to -1. Since there is a high possibility that a failure has actually occurred in 20-N or a failure such as disconnection of the connection path 30, the hard disk device returns to step S1 without further restarting. The failure detection of 20-1 to 20-N is continued.

  When it is determined in step S9 that the failure detection unit 11 has no failure in at least one of the hard disk devices 20-1 to 20-N determined to have detected the failure within the predetermined time in step S4 (step S9). : NO), the rebuilding unit 15 acquires the configuration information of the logical drive from the hard disk devices 20-1 to 20-N (step S10). When acquiring the logical drive configuration information, the rebuilding unit 15 reconstructs the logical drive by combining the hard disk devices 20-1 to 20-N based on the acquired logical drive configuration information (step S11).

Thus, according to the present embodiment, the failure detection unit 11 detects a failure that has occurred in the hard disk devices 20-1 to 20-N, and the restart unit 14 detects a failure in the hard disk devices 20-1 to 20-N. When multiple devices are detected within a predetermined time, the device itself is restarted, and the rebuilding unit 16 rebuilds the logical drive based on the logical drive configuration information stored in the hard disk devices 20-1 to 20-N at the time of restarting. To do.
As a result, when the failure detection unit 11 determines that the plurality of hard disk devices 20-1 to 20-N have failed at close timings, recovery of the logical drive can be attempted. At this time, when it is determined that the hard disk devices 20-1 to 20-N are out of order due to a temporary abnormality in the communication path 30, the logical drive can be restored.

  As described above, the embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to the above, and various design changes and the like can be made without departing from the scope of the present invention. It is possible to

  The RAID controller device 10 described above has a computer system therein. The operation of each processing unit described above is stored in a computer-readable recording medium in the form of a program, and the above processing is performed by the computer reading and executing this program. Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Alternatively, the computer program may be distributed to the computer via a communication line, and the computer that has received the distribution may execute the program.

  The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

It is a schematic block diagram which shows the structure of the RAID controller apparatus by one Embodiment of this invention. It is a flowchart which shows operation | movement of a RAID controller apparatus.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Computer 10 ... RAID controller apparatus 11 ... Failure detection part 12 ... Failure detection pattern monitoring part 13 ... Hard disk information storage part 14 ... Reboot part 15 ... Reboot number memory | storage part 16 ... Reconstruction part 20-1-20-N ... Hard disk device 21 ... Hard disk 22 ... Control unit 30 ... Connection path

Claims (5)

A RAID controller device that combines a plurality of hard disks that store configuration information of virtual storage means in advance and operates as one logical virtual storage means,
A failure detection means for detecting a failure of each of the plurality of hard disks;
Restarting means for restarting the apparatus when the failure detecting means detects a failure of a plurality of hard disks within a predetermined time;
Rebuilding means for rebuilding the virtual storage means based on the configuration information upon restart;
A RAID controller device comprising: The failure detection means detects a failure of a plurality of hard disks that have detected a failure within the predetermined time after restart by the restart means,
When the restarting unit detects a failure of a plurality of hard disks that have detected a failure within the predetermined time, the restarting unit performs restarting again,
The number of times that the restarting unit executes the restart is less than a predetermined number of times,
The RAID controller device according to claim 1. A processing method using a RAID controller device that combines a plurality of hard disks that store configuration information of virtual storage means in advance and operates as one logical virtual storage means,
The failure detection means detects a failure of each of the plurality of hard disks,
The restarting means restarts the apparatus when the failure detecting means detects a failure of a plurality of hard disks within a predetermined time,
The rebuilding means rebuilds the virtual storage means based on the configuration information at the time of restarting.
A processing method characterized by the above. A RAID controller circuit that combines a plurality of hard disks that store configuration information of virtual storage means in advance and operates as one logical virtual storage means,
A failure detection circuit for detecting a failure of each of the plurality of hard disks;
A restart circuit for restarting the apparatus when the failure detection means detects a failure of a plurality of hard disks within a predetermined time;
A reconfiguration circuit that reconstructs the virtual storage means based on the configuration information upon restart;
A RAID controller circuit comprising: A RAID controller device that combines a plurality of hard disks that store the configuration information of the virtual storage means in advance and operates as one logical virtual storage means,
Failure detection means for detecting a failure of each of the plurality of hard disks;
Restarting means for restarting the apparatus when the failure detecting means detects failures of a plurality of hard disks within a predetermined time;
Rebuilding means for rebuilding the virtual storage means based on the configuration information at the time of restart;
Program to operate as.

Images