JP7513294B2 - STORAGE APPARATUS, STORAGE SYSTEM, FAILURE RECOVERY METHOD, AND PROGRAM - Google Patents

Description

The present invention relates to a storage device, a storage system, a failure recovery method, and a program.

Conventionally, in a storage system with a RAID (Redundant Array of Independent Disks) configuration, if a disk failure such as a hard disk caused the system to enter a degraded state, the only way to recover was to physically replace the disk. This is because a failure in a hard disk, even if it was only a partial failure, could affect the entire disk. In this regard, non-volatile semiconductor storage devices such as flash memory in SSDs (Solid State Drives) have the advantage that they can continue operation while using other areas without using bad blocks.

Patent Document 1 discloses the following non-volatile semiconductor memory device. A RAID system is constructed within a single non-volatile semiconductor memory device, making it possible to replace faulty areas on a block-by-block basis. When a failure occurs, it is possible to automatically replace the storage area. This makes it unnecessary to replace the storage device, as opposed to constructing a RAID using multiple storage devices.

JP 2013-125513 A

The disclosures of the above-mentioned prior art documents are incorporated herein by reference. The following analysis was conducted by the present inventors.

Incidentally, in addition to failures related to reading and writing data, disk device failures can also be determined to be due to discrepancies in the disk-specific information set on the disk device. For example, a failure can cause a capacity setting of 32 GB to change to 8 GB. In such cases, disk-specific setting information cannot be reconfigured in a storage system where RAID has already been configured, so the disk device must be removed and edited on a separate device for reconfiguration (not in a RAID environment, but in an environment where the system can recognize the disk individually). Therefore, correcting the disk-specific information requires the removal of the device, which ultimately takes the same amount of work as physically replacing the disk.

The present invention aims to provide a storage device, a storage system, a failure recovery method, and a program that can recover from a failure caused by storage (disk) specific information of a storage device (disk device) without removing the device.

According to a first aspect of the present invention or disclosure, a storage device is provided that has a storage unit that holds storage setting information, which is storage setting information, a comparison unit that compares the storage setting information with storage-specific information set in the storage in response to a storage fault notification that is notified when a fault occurs in the storage, and a setting unit that sets the storage setting information as storage-specific information based on the comparison result of the comparison unit.

According to a second aspect of the present invention and disclosure, there is provided a storage system including a control device having: a plurality of storage devices; a failure notification unit that, when a failure is detected in any of the plurality of storage devices, sends a storage failure notification to a storage device involved in the occurrence of the failure among the plurality of storage devices; and a reconstruction unit that, when a setting unit of the storage device involved in the failure sets storage setting information as storage specific information, performs control to reconstruct the storage device involved in the failure.

According to a third aspect of the present invention or disclosure, there is provided a failure recovery method including the steps of: when a failure is detected in any of a plurality of storage devices, sending a storage failure notification to a storage device related to the occurrence of the failure among the plurality of storage devices; in response to the storage failure notification, comparing storage setting information, which is setting information of the storage, with storage specific information set in the storage; if the comparison shows that the storage setting information and the storage specific information do not match, setting the storage setting information as the storage specific information; and performing control to reconstruct the storage device related to the failure.

According to a fourth aspect of the present invention or disclosure, a program is provided for causing a computer to execute the following steps when a fault is detected in one of a plurality of storage devices: sending a storage fault notification to a storage device involved in the fault among the plurality of storage devices; comparing storage setting information, which is setting information for the storage, with storage-specific information set in the storage in response to the storage fault notification; setting the storage setting information as storage-specific information when the storage setting information and the storage-specific information do not match as a result of the comparison; and controlling the reconstruction of the storage device involved in the fault.

According to each of the aspects of the present invention and disclosure, the present invention provides a storage device, a storage system, a failure recovery method, and a program that can recover from a failure caused by storage (disk) specific information of a storage device (disk device) without removing the device.

2 is a block diagram showing an example of a configuration of a storage device according to an embodiment. FIG. FIG. 13 is a diagram showing an example of items of "storage setting information" and "storage specific information." 1 is a conceptual diagram showing an outline of the configuration of a storage system according to a first embodiment. 1 is a block diagram illustrating an example of a configuration of a storage system according to a first embodiment. 4 is a flowchart showing the operation of the storage system of the first embodiment. FIG. 1 is a diagram illustrating an example of a hardware configuration of a storage system according to a first embodiment.

First, an overview of one embodiment will be described. Note that the reference numerals in the drawings attached to this overview are added to each element for convenience as an example to aid in understanding, and the description of this overview is not intended to be limiting in any way. Furthermore, the connection lines between blocks in each figure include both bidirectional and unidirectional ones. One-way arrows are used to indicate the flow of the main signal (data) and do not exclude bidirectionality. Furthermore, although not explicitly shown in the circuit diagrams, block diagrams, internal configuration diagrams, connection diagrams, etc. shown in this disclosure, input ports and output ports exist at the input and output ends of each connection line. The same applies to input/output interfaces.

Figure 1 is a block diagram showing an example of the configuration of a storage device according to one embodiment. As shown in this figure, a storage device 10 according to one embodiment has a holding unit 11, a matching unit 12, and a setting unit 13.

The storage unit 11 stores storage setting information, which is storage setting information. "Storage" refers to SSDs (Solid State Drives), hard disks, and other rewritable storage media. "Storage setting information" refers to information that includes at least one of the following: storage capacity, issued command queue settings, and cache enable/disable settings, and is information that includes the same items as the storage specific information described below. In SSDs, hard disks, etc., the storage setting information is stored as a list in an EPROM (Erasable Programmable Read Only Memory) in the built-in board.

The collation unit 12 compares the storage setting information with the storage specific information set in the storage in response to a storage fault notification that is notified when a fault occurs in the storage. A "storage fault notification" is a command notified from a host or controller in which a fault is detected. "Storage specific information" is information that includes the same items as the above-mentioned "storage setting information." "Storage specific information" is unique information that includes various setting information related to the storage device. Since it is set at the factory, it does not need to be rewritten during normal operation, so it is necessary to update it using dedicated software. In particular, for storage devices that have a RAID configuration, it is necessary to remove them from the system and connect them to a PC or the like in a standalone state to make the changes.

Figure 2 shows an example of an item in "storage specific information." This item is also common to "storage setting information." A setting value is assigned to each item in this figure and is stored in the storage device.

The setting unit 13 sets the storage setting information as the storage specific information based on the result of the comparison by the comparison unit. Depending on the result of the comparison by the comparison unit, which is a match or mismatch, the setting information held is rewritten as storage specific information.

In one embodiment of the storage device, if storage-specific information is lost or corrupted due to a failure, it would normally be necessary to replace the storage device or remove the storage device to correct the storage-specific information. However, in response to a failure notification command, the stored storage setting information is written as storage-specific information, making it possible to reconstruct the disk as a new disk. In other words, it is possible to reconstruct the disk without removing the storage device.

Specific embodiments are described in more detail below with reference to the drawings. Note that the same components in each embodiment are given the same reference numerals and their description is omitted.

[First embodiment]
3 is a conceptual diagram showing an outline of the configuration of a storage system according to the first embodiment. As shown in this figure, the storage system 100 of this embodiment is composed of a host device 104 connected to a controller 101, the controller 101, SSD #1 (102), and SSD #2 (103). As shown in the figure, the storage system 100 may include a host device 104 such as a server. SSD #1 (102) and SSD #2 (103) have a RAID configuration. Note that, regarding SSDs, the number of SSDs is not limited to two, since two or more SSDs are connected to the controller depending on the type of RAID configuration.

Figure 4 is a block diagram showing an example of the configuration of a storage system according to the first embodiment. As shown in this figure, the storage system 100 of this embodiment includes a control device 14 and storage devices #1 (17) and #2 (18). The control device 14 has a fault notification unit 15 and a reconstruction unit 16. The storage system 100 may also have a host device. Although not shown, as described above, the storage devices #1 and #2 (17 and 18) have a retention unit 11, a collation unit 12, and a setting unit 13. Note that these configurations have been described in the above embodiment, so explanations will be omitted.

The control device 14 is generally a RAID controller. However, although the functions of the control device 14 are realized by a software module or a hardware module, these functions may also be realized by placing a software module on the host device side and executing this.

When a fault is detected in one of the multiple storage devices, the fault notification unit 15 notifies a storage fault notification to the storage device involved in the fault. The fault may be detected in the control device 14 of the storage system 100, or in the host device 104 to which the storage system 100 is connected. In this embodiment, a "fault" includes a situation in which disk-specific information in a storage device is lost or destroyed.

When the setting unit 13 of the storage device 10 involved in the failure sets the storage setting information as storage specific information, the reconstruction unit 16 performs control to reconstruct the storage device involved in the failure. Reconstruction is performed so that the storage device in which lost or broken disk specific information has been rewritten with the storage setting information can be used again. The reconstruction process corresponds to the process of a rebuild operation in a RAID configuration.

As described above, the comparison unit 12 of the storage device compares the storage setting information with the storage-specific information set in the storage in response to a storage fault notification notified when a fault occurs in the storage device, and may also have an exchange notification unit (not shown) that notifies the user that the storage needs to be replaced if the storage setting information and the storage-specific information match.

The storage device of the storage system of this embodiment has a RAID configuration, but the RAID configuration may be any configuration other than RAID 0. For example, it may be a RAID 1 or RAID 10 configuration that configures mirroring, or a RAID 5 or RAID 6 configuration that uses parity.

[System Operation]
5 is a flow chart showing the operation of the storage system of this embodiment. As shown in this figure, if a failure is detected during system operation (step S01, Y), the storage setting information is checked to see if it matches the storage specific information (step S02). If the information matches (step S02, Y), there is a possibility of a disk drive or other failure, so the storage device must be replaced (step S07). After replacement, a reconstruction process is executed (step S08).

Next, if the storage setting information does not match the storage specific information (step S02, N), it is determined that the failure is due to the storage specific information, and the storage setting information is set as the storage specific information (step S03). The control device 14 (RAID controller) then recognizes it as a new storage device (step S04). In this state, recovery is possible by performing the rebuild process that is normally performed in a RAID configuration (step S05).

If the system recovers after the reconstruction process (step S06, Y), it can return to normal operation. If the system does not recover (step S06, N), this means that there is another fault, and the storage device will be replaced (step S07).

[Hardware configuration]
The storage system of this embodiment has a configuration exemplified in Fig. 6. For example, the storage system has a host device 104, a RAID controller 101, an SSD #1 (102) and an SSD #2 (103), and each SSD has an I/O interface (105, 110), an EPROM (non-volatile memory) (106, 111), a cache memory (107, 112), a processor (108, 113), and a flash memory (109, 114).

Although not shown, the host device 104 and the RAID controller 101 have at least a CPU, memory, and an I/O interface, which are components that make up a computer.

The configuration shown in FIG. 6 is not intended to limit the hardware configuration of the storage system. The storage system may include hardware not shown. Furthermore, the number of SSDs and other devices included in the storage system is not intended to be limited to the example shown in FIG. 6. For example, three or more SSDs may be included in the storage device 10.

The functions of the storage system are realized by a set of programs (processing modules) such as a setting program stored in the EPROM (106, 111) of the SSD (102, 103), a matching program, a fault notification program, a reconstruction program, and a replacement notification program stored in the storage area of the RAID controller 101 or the host device 104, and a set of data such as storage setting information and storage specific information used by the setting programs. The processing modules are realized, for example, by the processor (108, 113) executing each program stored in the EPROM (106, 111). The programs can be downloaded via a network or updated using a storage medium that stores the programs. Furthermore, the processing modules may be realized by semiconductor chips. That is, it is sufficient to have some means for executing the functions performed by the processing modules using some hardware and/or software.

[Hardware Operation]
When the storage system detects a fault during operation, a fault notification program is executed in the CPU of the host device 104 or the RAID controller 101. The following describes an example in which a fault occurs in SSD #1. The program sends a notification message via the I/O interface 105 to the storage device in which the fault was detected. Upon receiving the message, the program is read from the EPROM 106 and the verification program is put into execution on the processor 108.

The comparison program compares the storage setting information stored as a list in EPROM 106 or flash memory 109 with the storage-specific information. If the results match, the replacement notification program is executed in processor 108 and sends a replacement notification message via I/O interface 105. Upon receiving the replacement notification message, host device 104 outputs information that the storage device needs to be replaced via a display device or the like.

If the comparison results in a mismatch, the configuration program is read from EPROM 106 and executed on processor 108. The program reads the storage configuration information stored in EPROM 106 or flash memory, and writes it to a storage area such as EPROM 106 where storage-specific information is stored.

After writing, the reconstruction program is called from memory on the host device 104 or the RAID controller 101 and executed by the CPU. If the RAID controller 101 detects a failure and the storage setting information is written, it recognizes the storage device as a new one and executes a reconstruction (rebuild) process, such as transferring data from the other storage device where no failure has occurred.

[Effects]
With the storage system according to the first embodiment described above, particularly in an SSD, which is a storage device having a RAID configuration, it is possible to have the storage device recognized as a new storage device without removing the device by repairing the storage-specific information, and it is possible to recover from a system failure due to a breakdown by performing a reconstruction (rebuild) process.

A part or all of the above-described embodiments can be described as the following supplementary notes. However, the following supplementary notes are merely examples of the present invention, and the present invention is not limited to such cases.
[Appendix 1]
This is the same as the storage device according to the first aspect described above.
[Appendix 2]
The storage device is preferably the storage device of appendix 1, which is configured with an SSD (solid state drive).
[Appendix 3]
Preferably, the storage device of appendix 2, wherein the storage setting information and the storage specific information include at least one of information on storage capacity, issued command queue setting, and cache enable/disable setting.
[Appendix 4]
Preferably, in the storage device according to appendix 3, when the collation by the collation unit shows that the storage setting information and the storage specific information do not match, the setting unit sets the storage setting information as the storage specific information.
[Appendix 5]
Preferably, the storage device of appendix 4 has an exchange notification unit that notifies the user that the storage device needs to be replaced when the storage setting information and the storage specific information match as a result of the comparison by the comparison unit.
[Appendix 6]
This is the same as the storage system according to the second aspect described above.
[Appendix 7]
Preferably, the storage system according to claim 6, wherein the plurality of storage devices are configured in a mirroring configuration.
[Appendix 8]
A storage system preferably as described in appendix 6, wherein at least one of the multiple storage devices has a parity area.
[Appendix 9]
This is the same as the failure recovery method according to the third aspect described above.
[Appendix 10]
This is as per the program related to the fourth perspective mentioned above.
In addition, Supplementary Notes 9 and 10 can be expanded into Supplementary Notes 2 to 5, similar to Supplementary Note 1.

The disclosures of the above cited patent documents are incorporated herein by reference. Within the framework of the entire disclosure of the present invention (including the scope of the claims), modifications and adjustments of the embodiments and the embodiments are possible based on the basic technical ideas. Furthermore, within the framework of the entire disclosure of the present invention, various combinations and selections (including partial deletions) of the various disclosed elements (including each element of each claim, each embodiment and each element of each embodiment, each element of each drawing, etc.) are possible. In other words, the present invention naturally includes various modifications and corrections that a person skilled in the art would be able to make in accordance with the entire disclosure, including the scope of the claims, and the technical ideas. In particular, with regard to the numerical ranges described in this document, any numerical value or subrange included in the range should be interpreted as being specifically described even if not otherwise specified.

10: storage device 11: holding unit 12: collation unit 13: setting unit 14: control device 15: fault notification unit 16: reconstruction unit 17: storage device #1
18: Storage device #2
100: Storage system 101: RAID controller, controller 102: SSD #1
103: SSD #2
104: host device 105, 110: I/O interface 106, 111: EPROM
107, 112: cache memory 108, 113: processor 109, 114: flash memory

Claims (9)

A storage unit that stores storage setting information, which is setting information of the storage;
a collation unit that collates the storage setting information with storage specific information set in the storage in response to a storage failure notification that is notified when a failure occurs in the storage;
a setting unit that sets the storage setting information as the storage unique information when the storage setting information and the storage unique information do not match as a result of the comparison by the comparison unit;
an exchange notification unit that notifies the user that a storage device needs to be replaced when the storage setting information and the storage specific information match as a result of the comparison by the comparison unit;
A storage device having the above configuration. The storage device of claim 1, wherein the storage device is configured with an SSD (Solid State Drive). The storage device of claim 2, wherein the storage setting information and the storage specific information include at least one of the following information: storage capacity, issued command queue setting, and cache enable/disable setting. when the comparison result by the comparison unit indicates that the storage setting information and the storage specific information do not match, the setting unit sets the storage setting information as the storage specific information.
The storage device of claim 3. A plurality of storage devices according to any one of claims 1 to 4 ;
a fault notification unit that notifies a storage fault notification to a storage device involved in the occurrence of the fault among the plurality of storage devices when a fault is detected in any of the plurality of storage devices;
a reconstruction unit that performs control for reconstructing the storage device related to the failure when the setting unit of the storage device related to the failure sets the storage setting information as the storage specific information;
A storage system including a control device having a 6. The storage system according to claim 5 , wherein the plurality of storage devices are configured in a mirroring configuration. 6. The storage system according to claim 5 , wherein at least one of said plurality of storage devices has a parity area. A disaster recovery method in which a computer executes the following steps:
when a fault is detected in any one of a plurality of storage devices, sending a storage fault notification to a storage device involved in the fault occurrence among the plurality of storage devices;
In response to the notification, comparing storage setting information, which is setting information of the storage, with storage specific information set in the storage;
a step of setting the storage setting information as the storage specific information when the storage setting information and the storage specific information do not match as a result of the comparison;
a step of notifying the user that the storage device needs to be replaced when the storage setting information and the storage specific information match as a result of the comparison;
performing control for reconstructing the storage device related to the failure;
Disaster recovery methods, including: When a fault is detected in any one of a plurality of storage devices, a storage fault notification is sent to a storage device involved in the fault occurrence among the plurality of storage devices;
In response to the notification, a process of comparing storage setting information, which is setting information of the storage, with storage specific information set in the storage;
a process of setting the storage setting information as the storage specific information when the storage setting information and the storage specific information do not match as a result of the comparison;
If the storage setting information and the storage specific information match as a result of the comparison, a process of notifying that the storage device needs to be replaced;
A process of performing control for reconstructing the storage device related to the failure;
A program for causing a computer to execute the following.

Images