CN101281452A - Automatic hard disk reconstruction method - Google Patents

Abstract

The invention provides an automatic hard disk reconstruction method which is suitable for a host adopting a disk array, wherein the disk array comprises a plurality of first hard disks. Each first hard disk is connected to a hard disk slot of the host through a small computer system interface. The method comprises the following steps: the first hard disks are initialized first, and data configuration information of the first hard disks is recorded. Then, the pins configured on each hard disk slot are used for detecting whether a hard disk is inserted into or pulled out of one of the hard disk slots. And when the second hard disk is detected to be inserted, searching data configuration information according to the plugging data, wherein the data configuration information corresponds to the hard disk slot into which the second hard disk is inserted. And then, according to the data configuration information, the backup data stored in other first hard disks are used for reconstructing a second hard disk.

Description

Hard Disk Automatic Rebuild Method

technical field

The present invention relates to a method for rebuilding a hard disk, and in particular relates to a method for automatically rebuilding a hard disk.

Background technique

Nowadays, network technology is becoming more and more mature. Whether it is in enterprises or schools, the proportion of information transmission and exchange relying on the network is getting higher and higher, and the reliability of the server as an information exchange platform is also becoming more and more important. Therefore, when there is a problem with the hardware of the server, it is often necessary to replace part of the hardware without stopping the server, so as to ensure that the communication of information is not interrupted. In addition, the server also needs to have the function of data fault tolerance, so that the data can be restored in time when the data is damaged.

In order to replace part of the hardware without stopping the server, the server needs to provide the function of hot pluggable hardware, so that the system can run without being affected after the hardware is plugged and unplugged. In addition, Redundant Array of Independent Disks (RAID) can provide the fault tolerant function of the storage device in the server, so that the data stored in the server can be restored when it is partially damaged.

Redundant disk array is a virtual group composed of multiple disks through hardware or software controllers. Replacement of hard disks that need to be shut down and expansion of hard disk capacity, etc. Redundant disk arrays can adopt different policies to manage disk groups according to different disk array levels. Taking level 0 as an example, each data will be divided and stored in different disks. When obtaining a piece of data, the host can read the divided data from different disks in parallel, so the speed of reading data will be much faster than reading data from a single disk. In addition, taking Level 1 as an example, data will be stored on different disks at the same time, that is, different disks will keep copies of the same data. Therefore, when one of the disks is damaged, data can still be effectively read by other disks. Therefore, after the damaged disk is replaced, the redundant disk array can use the data in other disks to reconstruct the data of the replaced hard disk.

In today's server environment, there are not a few hard disks connected to the host computer via Small Computer System Interface (SCSI). However, the small computer system interface does not provide the functions of hot plugging and automatic data reconstruction. Therefore, at present, it is necessary to use an additional control chip to issue commands to the small computer system interface, so as to achieve the functions of disk hot swap and automatic data reconstruction, such as using a GEN318 chip. However, by adding hardware to provide the above-mentioned functions, the production cost must increase accordingly. Correspondingly, the selling price will also rise accordingly, making the competitiveness of the product vulnerable to being reduced by the price factor.

Contents of the invention

The invention provides a method for automatically rebuilding a hard disk, which provides functions of hot plugging and automatic rebuilding of the hard disk by recording the insertion and removal data of the hard disk, and saves the cost of purchasing hardware equipment.

The invention proposes a method for automatically rebuilding a hard disk, which is suitable for a host using a disk array, wherein the disk array includes a plurality of first hard disks. Each first hard disk is connected to the hard disk slot of the host through a Small Computer System Interface (SCSI). The method includes the following steps: first, initialize the first hard disks, and record the data configuration information of the first hard disks. Then, through the pins (pins) arranged on each hard disk slot, it is detected whether a hard disk is inserted into or pulled out from these hard disk slots. When it is detected that a second hard disk is inserted, the data configuration information corresponding to the hard disk slot into which the second hard disk is inserted is searched according to the insertion and removal data. Then, according to the data configuration information, the second hard disk is rebuilt using the backup data stored in other first hard disks.

In an embodiment of the present invention, the above-mentioned step of detecting whether a hard disk is inserted into or pulled out of the hard disk slot includes: when a hard disk is inserted into or pulled out of these hard disk slots, an interrupt signal is generated through a pin configured on the hard disk slot (interrupt), when an interrupt signal is received, it scans all hard disk slots to find out which hard disk slot has a hard disk inserted or pulled out.

In an embodiment of the present invention, after the above step of detecting whether a hard disk is inserted into or removed from the hard disk slot, it may further include recording insertion and removal data when it is detected that the first hard disk is pulled out from the hard disk slot.

In an embodiment of the present invention, before the step of rebuilding the second hard disk, it also includes setting the state of the second hard disk as hot spare (hot spare), clearing all data in the second hard disk, and checking whether the second hard disk is There are steps such as damaged magnetic sectors.

In one embodiment of the present invention, the above data configuration information includes the disk array level and the disk array group to which each first hard disk belongs, and the plug-in data includes hard disk insertion and removal records and corresponding hard disk slot numbers .

In an embodiment of the present invention, the above disk array may be a redundant array of independent disks (Redundant Array of Independent Disks, RAID), and the level of the redundant array of independent disks may include one of 1, 5 and 10. In addition, the above pins may be general purpose input/output (GPIO) pins.

The present invention configures a pin on each hard disk slot, and can automatically detect hard disk plugging and unplugging through the interrupt signal generated by the pin when plugging and unplugging the hard disk, and record it as plugging data, and the plugging data is used to judge whether the hard disk needs automatic Rebuild, when it is determined that automatic rebuilding is required, automatic rebuilding will be performed. Furthermore, when the plug-in data records that the hard disk is pulled out from one of the hard disk slots, and then if there is a record that a hard disk is inserted into the hard disk slot, the present invention judges that the inserted hard disk is used to replace the original hard disk. Hard disk, that is, to automatically rebuild data from the backup data of other hard disks. In addition, the present invention can achieve the functions of hot plugging and automatic data reconstruction by using low-cost pins, so the overall production cost will not be burdened.

In order to make the above-mentioned features and advantages of the present invention more comprehensible, a number of embodiments are exemplified below and described in detail with accompanying drawings.

Description of drawings

FIG. 1 is a flowchart of a method for automatically rebuilding a hard disk according to a first embodiment of the present invention.

FIG. 2 is a flowchart of a method for automatically rebuilding a hard disk according to a second embodiment of the present invention.

Detailed ways

The methods for automatically rebuilding hard disks in the following embodiments are applicable to hosts using disk arrays. Wherein, the disk array includes a plurality of first hard disks, and each first hard disk is connected to the hard disk slot of the host through the small computer system interface. Wherein, the disk array is, for example, a redundant disk array. In addition, in order to achieve hot-swapping and automatic rebuilding functions at a lower cost, the following embodiments can detect the status of hard disk insertion and removal in real time through the pins configured on each hard disk slot, and automatically reset the hard disk when necessary. Complete the HDD reconstruction work. For the detailed process, please refer to the descriptions of the following embodiments.

[first embodiment]

FIG. 1 is a flowchart of a method for automatically rebuilding a hard disk according to a first embodiment of the present invention. Please refer to FIG. 1 , first, step S110 is performed at the initial stage of system establishment to initialize the first hard disks, and record and save the data configuration information of the first hard disks. Because these first hard disks can belong to different disk array groups according to requirements during system planning. Wherein, these different disk array groups can simultaneously provide services of different disk array levels on the host.

For example, some groups in the host are responsible for providing level 0 services to speed up the reading speed, while another part of the groups provide level 1 or level 5 fault tolerance and backup services. Therefore, after initializing the first hard disks, it is necessary to save the data configuration information corresponding to the disk array level and the disk array group to which each first hard disk belongs. When the hard disk is replaced in the future, the data configuration information can be used for comparison to determine the service that the replaced hard disk should provide, and the original service status can be quickly restored. Since this embodiment can provide the function of automatically rebuilding the hard disk, and the premise of rebuilding is that there needs to be backup data in the disk array, so the level of the disk array can include one of 1, 5 and 10 to provide the backup data to achieve Fault tolerance and reconstruction capabilities.

Next, proceed to step S120, and detect whether a hard disk is inserted into the hard disk slot or whether a hard disk is pulled out through the pins arranged on each hard disk slot. Most of the hard disk slots on the server are installed on the backplane to facilitate the replacement of the hard disk, so the pins used to detect the insertion and removal of the hard disk can be conveniently configured on the backplane, and there is no need for complicated wiring work like configuring chips , so the impact on production cost can be avoided. Wherein, the pin is, for example, a general input and output pin.

Next, proceed to step S130, when it is detected that a second hard disk is inserted into the hard disk slot, search for the data configuration information according to the insertion and removal data, so as to search for the data configuration information corresponding to the hard disk slot. The above insertion and removal data may include hard disk insertion and removal records and corresponding hard disk slot numbers. Specifically, if there is a record of the first hard disk being pulled out of the hard disk slot before the second hard disk is inserted into the hard disk slot in the data of plugging and unplugging, it can be determined that the second hard disk is used to replace the pulled out first hard disk. a hard drive. Therefore, the data configuration information can be queried through the information corresponding to the hard disk slot, and the disk array group and the corresponding disk array level originally configured by the first hard disk pulled out from the hard disk slot can be found out.

Finally, proceed to step S140, after obtaining information such as the corresponding disk array group and disk array level according to the aforementioned data configuration information, cooperate with the backup status of the corresponding disk array level, and then use the backup data stored in other first hard disks to rebuild the first Two hard drives. For example, the hard disks corresponding to disk array level 1 will save the same data copy, so when rebuilding, the data of the hard disks in the group will be completely written to the second hard disk, while other levels have different backup status.

It is worth noting that during the process of hard disk insertion and removal, the user does not need to manually issue an instruction to inform the operating system to remove or add a new hard disk, but the method automatically detects it to provide the hot-swap function. In addition, when the first hard disk is damaged and replaced with the second hard disk, the method can also automatically rebuild the second hard disk through the backup data of other hard disks, and the user does not need to manually rebuild the data. That is to say, the hard disk automatic rebuilding method of this embodiment only needs to add low-cost pins to achieve the functions of hot swapping and automatic data rebuilding.

[Second embodiment]

FIG. 2 is a flowchart of a method for automatically rebuilding a hard disk according to a second embodiment of the present invention. This embodiment is similar to the first embodiment, the differences will be described below, and similar steps will not be described in detail. Referring to FIG. 2 , these first hard disks are initialized at the initial stage of system establishment, and the data configuration information of these first hard disks is recorded and saved (step S210 ).

Next, proceed to step S220, and detect whether a hard disk is inserted into or pulled out of the hard disk slot through the pins arranged on each hard disk slot. Wherein, step S220 includes two sub-steps of S222 and S224: step S222, when a hard disk is inserted into or pulled out of the hard disk slot, an interrupt signal is generated through a pin; step S224: when the interrupt signal is received, scan all hard disk slots, to find out which hard disk slot has a hard disk inserted or removed. In more detail, the action of plugging and unplugging the hard disk will trigger the pins of the hard disk slot to generate an interrupt signal to the operating system, and when the interrupt signal is received, the driver will drive the control chip of the small computer system interface to Scan the bus of all hard disk slots.

After the operating system receives the interrupt signal, it proceeds to step S230. When it detects that the first hard disk is pulled out from the hard disk slot, it will record a piece of plugging and unplugging data. The insertion/removal data may include the record of the hard disk being pulled out and the corresponding hard disk slot number, so as to be used for comparison in the future. Furthermore, if there is a second hard disk inserted into this hard disk slot later, because the hard disk slot has a record of pulling out the first hard disk, it can be judged that the second hard disk is used to replace this hard disk. first hard drive. Therefore, when it is detected that a second hard disk is inserted, the data configuration information corresponding to the hard disk slot into which the second hard disk is inserted is searched according to the above-mentioned insertion and removal data (step S240), to determine whether the second hard disk is used for replacement Damaged first hard drive.

Next, proceed to step S250, setting the state of the second hard disk as hot backup. In a redundant disk array, the hard disk in the hot backup state is not usually used to store data, but directly replaces the broken hard disk to store data when a hard disk fails, and rebuilds the original data on this hard disk. When the damaged first hard disk is pulled out and the inserted second hard disk is set as a hot spare, the controller of the redundant disk array will replace the original first hard disk with the second hard disk, and the original Data stored on the first hard drive is recreated on the second hard drive.

The step S260 of data reconstruction includes three sub-steps of S262～S266: S262: clear all data in the second hard disk, such as formatting the second hard disk; In the process of rebuilding or reading and writing data in the future, abnormal data reading or writing occurs; S266, according to the record in the data configuration information, initialize the second hard disk to the corresponding disk array level, In combination with the backup status corresponding to the level of the disk array, this embodiment can use the backup data stored in other first hard disks to rebuild the second hard disk, thereby completing the work of automatic data reconstruction.

In summary, the present invention configures pins on each hard disk slot, and can automatically detect hard disk plugging and unplugging through the interrupt signal generated by the pins when plugging and unplugging the hard disk, and provides the function of hot swapping, so that users do not need to bother Manually issue commands to detect hard disk insertion and removal. In addition, the present invention judges whether the hard disk needs to be automatically rebuilt through the recording of the hard disk insertion and removal data, and automatically rebuilds when it is determined that automatic rebuilding is required. Therefore, the user does not need to perform the time-consuming task of data reconstruction by the user after replacing the hard disk. Furthermore, when the plug-in data records that the hard disk is pulled out from one of the hard disk slots, and then if there is a record that a hard disk is inserted into the hard disk slot, the present invention judges that the inserted hard disk is used to replace the original hard disk. Hard disk, that is, to automatically rebuild data from the backup data of other hard disks. In addition, the present invention can achieve the aforementioned functions of hot plugging and automatic data reconstruction by using low-cost pins, so the overall production cost will not be burdened.

Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention, and anyone with ordinary knowledge in the technical field may make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be defined by the claims.

Claims (11)

1. method for automatically rebuilding hard disk, be applicable to a main frame that adopts a disk array, wherein this disk array comprises a plurality of first hard disks, and each those first hard disk is hard disk slots that are connected to this main frame by a small computer system interface, and this method comprises the following steps:

Those first hard disks of initialization, and write down a data configuration information of those first hard disks;

By be disposed at the stitch on each those hard disk slot, detect whether have hard disk to insert or extract those hard disk slots one of them;

When having detected one second hard disk and inserted, this data configuration information of this hard disk slot that is inserted corresponding to this second hard disk according to a plug data search; And

According to this data configuration information, utilize the Backup Data leave other those first hard disk in to rebuild this second hard disk.

2. whether method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, detect to have hard disk to insert or to extract one of them step of those hard disk slots to comprise:

Have hard disk to insert or extract those hard disk slots one of them the time, produce a look-at-me by this stitch; And

When receiving this look-at-me, promptly scan all hard disk slots, there is hard disk to insert or extracts to find out which hard disk slot.

3. method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, detect whether have hard disk to insert or extract one of them step of those hard disk slots after, also comprise:

When detect those first hard disks one of them when the hard disk slot is extracted, write down this plug data.

4. method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, before the step of rebuilding this second hard disk, also comprises:

The state of setting this second hard disk is a Hot Spare.

5. method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, the step of rebuilding this second hard disk comprises:

Remove all data in this second hard disk.

6. method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, the step of rebuilding this second hard disk comprises:

Check whether this second hard disk has the magnetic region of damage.

7. method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, this data configuration information comprises disk array grade under each those first hard disk and affiliated disk array group.

8. method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, these plug data comprise record and the corresponding hard disk slot numbering that hard disk inserts, extracts.

9. method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, this disk array is a redundance type magnetic disc array.

10. method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, the grade of this redundance type magnetic disc array comprise 1,5 and 10 one of them.

11. method for automatically rebuilding hard disk as claimed in claim 1 is characterized in that, this stitch is a general input and output stitch.

Images