JP3601982B2 - Disk array device control method and disk array device - Google Patents

Description

[0001]
[Industrial applications]
The present invention relates to a method for controlling a disk array device and a disk array device, and more particularly to a method for controlling a disk array device that operates a plurality of disk devices in parallel and a disk array device.
[0002]
[Prior art]
In recent years, a general disk array device has a plurality of disk devices, and is configured to perform data processing by operating these in parallel. That is, when a data write command is received from the host device, the data is written to the plurality of disk devices after adding redundant data to the write data. On the other hand, when a data read command is received from the host device, the distributed data is read from a plurality of disk devices and restored, and the restored data is sent to the host device.
[0003]
In such a disk array device, when an error occurs in any one of the disk devices during processing of writing / reading from the host device, the disk device is disconnected from the processing and the operation is degraded. It is configured so that the write / read command can be performed by using the remaining disk devices without using a disk.
[0004]
For example, if a read error occurs due to a minute scratch or the like on a disk (recording medium) at the time of reading data, a read retry that takes about several seconds is performed inside the disk device. If the data is successfully read by this retry, the process is continued.If the error is an unreadable read error that cannot be remedied by the retry, the disk device is disconnected as a failure and the disk array device is degraded. I do. In this case, the disk array device restores the correct data from the remaining data including the redundant data, and sends it to the host device.
[0005]
In the disk array device, the degraded operation itself is not a problem because normal data reading is continued, but maintenance (disk device replacement) is required by transitioning to the degraded operation state, and this maintenance process must be performed. If this were the case, the degenerate operation would be permanent. Since the redundancy is lost until the maintenance processing is completed, there is a high risk that data reading becomes impossible if another failure occurs during that time.
[0006]
In view of such a danger, an example of a disk device capable of continuing stable operation without shifting to permanent degenerate operation due to an error of the disk device is disclosed in Japanese Patent Application No. 44844/1998. Has been described.
[0007]
In the disk array device described in this specification, when an error occurs in any one of a plurality of disk devices at the time of data writing / reading, the disk device is temporarily disconnected from the processing and the disk array is processed. The device is brought into a temporarily degenerate operation state. In this case, the disk array device performs data processing at the time of occurrence of an error inherent in the device (that is, performs data writing / reading using a disk device other than the disk device temporarily disconnected from the processing), and transmits data from the host device. Perform the instructions of
[0008]
If a new write or read command is received from the host device in a state where a disk device disconnected from the process due to the occurrence of an error already exists, the disk is continuously treated as a degraded operation state and the remaining disk devices are used. To execute an instruction from the host device.
[0009]
Further, the disk array device performs the allocation / assignment process to the spare area with the error occurrence location of the temporarily degraded disk device as a defective area at the same time as the execution of the write / read command processing from the host apparatus, and then newly uses it. Performs internal processing such as restoring and writing data from the data written to the remaining disk devices as data in the delivery area, and then releases the temporary degraded state of the disk device in which the error occurred, and temporarily resets the disk array device. It is configured to return from the degraded operation state to the normal operation state.
[0010]
As described above, in the disk array device described in the specification of Japanese Patent Application No. 10-44844, in parallel with the execution of the data write / read command from the host device, the error location of the temporarily degraded disk device is detected. Since the maintenance process and the process of restoring and writing data from another disk device after the maintenance process is completed, the disk array device does not shift to the permanent degraded operation state, It is possible to return to a normal operation state.
[0011]
[Problems to be solved by the invention]
However, the process of alternately assigning the defective area to the spare area and the process of restoring data from the remaining disk devices and writing the data to the spare area are complicated, and these processes are difficult to execute the write / read command from the host device. If performed in parallel, the performance of executing a write / read instruction may be reduced. If the ability to process write / read commands from the host device is reduced, there is a risk that stable operation cannot be continued, especially in a computer system with a short waiting time for data transfer.
[0012]
SUMMARY OF THE INVENTION An object of the present invention is to provide a disk array device capable of performing a recovery process of a disk device which has become in a temporarily degraded state due to occurrence of an error without deteriorating performance with respect to a data write / read request from a host device. That is.
[0013]
[Means for Solving the Problems]
In order to achieve the above object, a method of controlling a disk array device according to the present invention includes operating a plurality of disk devices in parallel and adding redundant data to the data when receiving a data write command from a host device. Control of a disk array device that writes data to the plurality of disk devices and restores the data allocated to the plurality of disk devices when receiving a data read command from the host device and sends the data to the host device. In the method,
When an error occurs in any of the plurality of disk devices when processing a data write / read command from the host device, the disk device is temporarily disconnected from the processing and a disk device other than the disk device is processed. (Hereinafter referred to as a remaining disk device) to continue the processing,
Executing a command using the remaining disk devices when there is the temporarily disconnected disk device when receiving a data write / read command from the host device;
Performing a maintenance process of the temporarily disconnected disk device only when receiving a command to perform a maintenance process from the host device;
Releasing the temporarily detached state of the temporarily detached disk device when the maintenance processing is completed.
[0014]
Further, the disk array device to which the above control method is applied operates a plurality of disk devices in parallel, adds redundant data to the data when receiving a data write command from a host device, and applies the redundant data to the plurality of disk devices. A disk array device configured to perform distribution and write, and when receiving a data read command from the host device, restore the data distributed to the plurality of disk devices and send the restored data to the host device;
Control means for controlling writing / reading of the data, when an error occurs in any of the plurality of disk devices when processing a data write / read command from the host device, The disk device is temporarily disconnected from the process, and the process is continued using a disk device other than the disk device (hereinafter, referred to as a remaining disk device), or a data write / read command is received from the host device. Control means for executing the command by using the remaining disk devices when there is the temporarily disconnected disk device when
When receiving an instruction to perform a maintenance process from the host device, a maintenance processing unit that performs the maintenance process if the temporarily disconnected disk device exists,
When the maintenance processing is completed, the disk device is provided with a release unit that releases the temporary disconnection state of the disk device.
[0015]
As described above, according to the disk array device control method and the disk array device of the present invention, the maintenance process of the disk device which is in the temporarily degraded operation state due to the occurrence of an error only when a command for performing the maintenance process is received from the host device is performed. To do. With this configuration, the disk array device performs the maintenance process, that is, the process of alternately assigning the defective area of the temporarily degraded disk device to the spare area and the process of restoring data from the remaining disk devices and writing the data to the alternate area. It is not performed simultaneously with the processing of the write / read / read command from the host device. Therefore, at the time of processing a write / read command from the host device, there is no possibility that the performance related to the execution of the command is reduced.
BEST MODE FOR CARRYING OUT THE INVENTION
[0016]
The configuration and operation of a disk array device as one embodiment of the present invention will be described below with reference to the accompanying drawings.
[0017]
FIG. 1 is a block diagram showing a configuration of an embodiment of a disk array device according to the present invention. As shown in FIG. 1, the disk array device according to the present embodiment includes a disk array controller 2, four disk controllers 3-1 to 3-4, and four disk devices respectively controlled by the respective disk controllers. 4-1 to 4-4. The disk array control unit 2 identifies and processes an instruction from the host device 1 and includes an error occurrence position storage unit 21 and a write position storage unit 22 for data written in place of a temporarily degenerated disk. .
[0018]
When the disk array controller 2 receives a data write command from the host device 1, the disk array controller 2 adds redundant data (parity data) to the transmitted data and distributes the data to a plurality of disk controllers 3-1 to 3-4. I do. Each disk control unit writes data to the subordinate disk devices 4-1 to 4-4.
[0019]
On the other hand, when a data read command is received from the host device 1, the data written in the disk devices 4-1 to 4-4 are sent to the corresponding disk control units 3-1 to 3-4, respectively. The data is restored to one data by the array control unit 2 and transmitted to the host device 1.
[0020]
During processing of a write command or a read command from the host device 1, the disk array control unit 2 takes a certain amount of time to perform a processing operation related to read error rescue or the like in any of the disk devices 4-1 to 4-4. If the normal writing / reading is not completed within the time, the relevant disk device is temporarily disconnected from the instruction processing, and the disk array device is set to the temporary degenerate operation state, and the position (block address) where the error has occurred is the error occurrence position. It is stored in the storage unit 21. In this case, the disk array control unit 2 continues and completes the process according to the command from the host device 1 using a disk device other than the disk device.
[0021]
Further, when a data write command is received from the host device 1 while the disk array device is in the temporary degenerate operation state due to the occurrence of an error as described above, the disk device in the degenerate operation is continuously treated as the degenerate operation state, and In addition to executing the processing for the write command using the disk device, the position (block address) where the data is written is stored in the write position storage unit 22.
[0022]
Further, when a new data read command is received from the host device 1 while the disk array device is in the temporary degenerate operation state, the disk device in the temporary degenerate operation is continuously treated as a degenerate operation, and the disk array device is The data is restored using the data restoration function provided (that is, the data is restored using the remaining disk devices) and sent to the host device.
[0023]
As a control method at the time of writing data in the disk array device and a data restoring method at the time of reading data, a conventional general data processing method of the disk array device or the like can be used. Since it is described in detail in the application specification of Hei 10-44844, the description thereof will be omitted in this specification.
[0024]
Further, when the disk array device receives an error recovery command from the host device 1, it is first determined whether or not there is a disk device in the temporary degenerate operation. Here, if there is a disk device that is in the temporary degenerate operation, the error occurrence position is read from the error occurrence position storage unit 21 and this position is set as a defective area, and internal processing such as alternate allocation to a spare area is performed. Data to be written to this disk device is recovered from a disk device other than the disk device and written.
[0025]
Before this recovery process, if a data write command is received from the host device 1 during the temporary degenerate operation of the disk array device and processing for the command is performed, the write position storage unit 22 is stored in the write position storage unit 22 during the temporary degenerate operation. The data to be written to the disk device is recovered from the remaining disk devices to the registered write position, and the data is written to the temporarily degenerated disk device. When these restoration processes are completed, the temporary degraded operation state of the disk array device is released to return to the normal operation state.
[0026]
The operation of the disk array device of this embodiment will be described below with reference to the flowcharts shown in FIGS.
[0027]
As shown in FIG. 2, when the disk array device starts operating, the disk array control unit 2 waits for an instruction from the host device 1 (step S1 in FIG. 2). When a data read command is received from the host device 1 (step S4 in FIG. 2), it is first determined whether there is a disk device in a temporarily degenerated state (step S5 in FIG. 2). Here, if there is no disk device in the temporarily degenerate state, the disk controller 3-1 to 3-4 provided in the disk array device is instructed to read out to the subordinate disk devices 4-1 to 4-4 (FIG. Two steps S6). As a result, the disk devices 4-1 to 4-4 sequentially read the data of the respective devices and supply the data to the disk array control unit 2 via the disk control units 3-1 to 3-4. The disk array control unit 2 confirms that no error has occurred in the constituent disks, and upon receiving all these data normally (step S7 in FIG. 2), discards parity data from the read data and replaces the restored data. The command is sent to the host device (step S8 in FIG. 2), and the operation for the command is completed.
[0028]
On the other hand, when an error has occurred in any of the disk devices constituting the disk array device upon receiving the data read command from the host device 1 (step S7 in FIG. 2), the disk device in which the error has occurred is temporarily stored. The separation is performed to cause the degenerate operation (temporary degenerate state), and the error occurrence position is stored in the error occurrence position storage unit 21 (step S9 in FIG. 2). In this case, the disk array control unit 2 reads data from the remaining disk devices, recovers the data (step S10 in FIG. 2), and sends the data to the host device 1 (step S11 in FIG. 2).
[0029]
If there is a disk device in the degraded operation state when the data read command is received from the host device 1 (step S5 in FIG. 2), the disk device in the degraded state is continuously treated as the degraded operation state as shown in FIG. The data is recovered from the remaining disk devices and sent to the host device 1 (step S21 in FIG. 4).
[0030]
When the disk array controller 2 receives a data write command from the host device 1 (step S3 in FIG. 2), the processing is performed according to the flowchart shown in FIG. First, the disk array control unit 2 determines whether there is a disk device in a temporarily degenerated state (step S13 in FIG. 3). If there is no disk device in the temporarily degenerate state, the disk controller 3-1 to 3-4 is instructed to write to the subordinate disk devices 4-1 to 4-4 (step S14 in FIG. 3). After receiving the data to be written from the host device 1 and adding the parity data, the data is distributed to the disk controllers 3-1 to 3-4 and sent out. These data are written to the subordinate disk devices 4-1 to 4-4 by the disk control units 3-1 to 3-4.
[0031]
At this time, in each of the disk devices 4-1 to 4-4, reading of the data management area (ID section) is performed prior to writing of data. It is confirmed whether or not a read error has occurred in the read operation of the ID portion. If no read error has occurred (step S15 in FIG. 3), it means that the disk device has no abnormality, and the allocated data is written. (Step S16 in FIG. 3).
[0032]
On the other hand, if the above-described read error of the ID portion occurs in any of the disk devices upon receiving the data write command from the host device 1 (step S15 in FIG. 3), the disk device in which the error has occurred is disconnected from the processing. In this case, the error occurrence address is stored in the error occurrence position storage unit 21 (step S17 in FIG. 3). The write data from the host device 1 is written to the remaining disk devices by skipping the disk device in which the error has occurred (step S18 in FIG. 3).
[0033]
If there is a disk device in the degraded operation state when the data write command is received from the host device 1 (step S13 in FIG. 3), the disk device in the degraded state is continuously treated as the degraded operation state, and the data is transferred to the remaining disk devices. Is written (step S19 in FIG. 3). At the same time, the position (block address) where the data was written is sequentially registered in the write position storage unit 22 (step S20 in FIG. 3).
[0034]
Further, when the disk array control unit 2 receives an error recovery instruction from the host device 1 (Step S2 in FIG. 2), the processing is performed according to the flowchart shown in FIG. It is determined whether there is a temporarily degenerated disk device (Step S22 in FIG. 5). If there is a temporarily degenerated disk device, the error occurrence position of this disk device is read from the error occurrence position storage unit 21; This position is designated as a defective area, and an alternate allocation to a spare area (not shown) provided in a normal disk device is instructed. At the same time, the subsequent use of an error occurrence location is prohibited to prevent the error recurrence at the same location. (Step S23 in FIG. 5). When this process is completed, the data to be written is recovered from the remaining disk devices in the area to which the replacement has been performed, and the data is written to the area after the replacement (step S24 in FIG. 5).
[0035]
Next, it is determined whether or not data writing has been performed during the temporary degeneration operation of the disk array device (step S25 in FIG. 5). Here, if data writing has been performed during the temporary degenerate operation, the write position (block address) stored in the write position storage unit 22 is read, and the data in that area is replaced with the data of the remaining disk device. Are restored by reading them out, and sequentially written to the disk drive in the degenerate operation state (FIG. 5, step 26). When all the writing processing is completed, the disk device is released from the temporarily disconnected state (temporary degenerate state), and the disk array device is returned from the degraded operation state to the normal operation state (step S27 in FIG. 5). On the other hand, if data writing has not been performed during the temporary degeneration operation, the disk device is released from the temporary degeneration state and returned to the normal operation state (steps S25 and S27 in FIG. 5).
[0036]
This recovery processing is performed when the host device does not request data write / read processing in accordance with the operation status of the system. For example, a recovery command may be issued when the host device does not transmit a write command or a read command for a certain period of time or longer. Alternatively, it may be set so that an error recovery command is executed when the operation is stopped. For example, if an error recovery command is issued when the operation is stopped every day, the normal operation can be performed with the error recovered at the next operation start.
[0037]
If the command received from the host device 1 is not a read command, a write command, or an error recovery command, a process corresponding to the command is performed (step S12 in FIG. 2), and the operation for the command is completed. .
[0038]
Although the configuration of the example of the present invention has been described in detail above, the present invention is not limited to the above-described embodiment, and another configuration may be adopted according to a use situation or the like. For example, in the above-described example, a disk array device having four disk devices is configured, but three or less disk devices or five or more disk devices may be provided.
[0039]
【The invention's effect】
As described above, according to the present invention, when an error occurs in any one of the disk devices constituting the disk array device, the disk device is temporarily disconnected from the processing to be in the temporary degenerate operation state, and the remaining disk devices are used. Performs data write / read processing. Further, when a command for performing a maintenance process is received from the host device, the maintenance process for the disk device in which the error has occurred is performed periodically, and the maintenance process is performed in parallel with the execution of the write / read command from the host device. Therefore, even if an error occurs in any one of the disk devices, the permanent degraded operation state does not continue, and the performance of executing the write / read command from the host device does not decrease. . That is, it is possible to keep the processing capability for the write / read command from the host device in a good state, and to continue the stable operation of the system.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a disk array device according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating an operation of the disk array device illustrated in FIG. 1;
FIG. 3 is a flowchart illustrating an operation of the disk array device when processing a data write instruction.
FIG. 4 is a flowchart for explaining the operation of the disk array device when a disk device in a temporarily degenerate operation state already exists when performing a data read process;
FIG. 5 is a flowchart illustrating an operation of the disk array device when an error recovery command is received.
[Explanation of symbols]
1 Host device 2 Disk array control unit 21 Error occurrence position storage unit 22 Data write position storage units 3-1 to 3-4 Disk control units 4-1 to 4-4 Disk devices

Claims (5)

A plurality of disk devices are operated in parallel, and when receiving a data write command from the host device, redundant data is added to the data and the data is distributed to the plurality of disk devices and written, and data is read from the host device. A method of controlling a disk array device, which restores data allocated to the plurality of disk devices when receiving the command and sends the data to the host device,
When an error occurs in any of the plurality of disk devices when processing a data write / read command from the host device, an error occurrence position is recorded and the disk device is temporarily disconnected from the process. The disk array device is managed as a degraded operation (hereinafter, referred to as a temporary degraded operation), and the above processing is continued using a disk device other than the disk device (hereinafter, referred to as a remaining disk device) using redundancy. The process of
When receiving a data write / read command from the host device during the temporary degenerate operation, executing the command using the remaining disk devices using redundancy;
Only when a command for performing a maintenance process is received from the host device, the error occurrence position is set as a defective area, and internal processing for alternate allocation to a spare area is performed, and the disk device temporarily disconnected from the remaining disk device is executed. Performing a maintenance process of restoring data to be written to the
A step of releasing the temporary retraction operation state and returning to a normal operation state after the maintenance processing is completed. 2. The method for controlling a disk array device according to claim 1, further comprising receiving a data write command from the host device during the temporary retraction operation, and writing data to be written to the temporarily disconnected disk device. A step of storing a position where the writing is performed when writing to the remaining disk devices, and the maintenance step recovering the data from the stored writing position. Control method. 3. The disk array device control method according to claim 1, wherein the host device issues a command for performing the maintenance process when the disk array device starts or stops operating. Method. A plurality of disk devices are operated in parallel, and when receiving a data write command from the host device, redundant data is added to the data and the data is distributed to the plurality of disk devices and written, and data is read from the host device. A disk array device configured to restore the data allocated to the plurality of disk devices when receiving the command and to send the restored data to the host device;
Control means for controlling writing / reading of the data, when an error occurs in any of the plurality of disk devices when processing a data write / read command from the host device, A disk device is temporarily disconnected from processing, managed as a degraded operation as a disk array device (hereinafter, referred to as a temporary degraded operation), and a disk device other than the disk device is used by utilizing redundancy (hereinafter, the remaining disk devices). the process continues with the previous SL above using the remaining disk apparatus using redundancy when subjected to write / read command data from the host device during temporary degeneration operation instruction using the called) and Controlling means to execute;
First storage means for recording an error occurrence position when an error occurs in any one of the disk devices;
When a command for performing a maintenance process is received from the host device, the error occurrence position is set as a defective region, and internal processing for replacement is performed in the spare region, and the disk device temporarily disconnected from the remaining disk devices is Maintenance processing means for performing maintenance processing for restoring data to be written and writing the data;
When the maintenance process is completed, the disk array device comprises a release unit that releases the temporary retraction operation state and returns to the normal operation. 5. The disk array device according to claim 4, wherein a data write command is received from the host device during the temporary degenerate operation, and the control unit writes the data to the remaining disk devices. A disk array device comprising a second storage unit for storing the data, wherein the maintenance processing unit calls a write position stored in the second storage unit and recovers the data from the write position. .

Images