CN104484251B - A kind of processing method and processing device of hard disk failure - Google Patents
A kind of processing method and processing device of hard disk failure Download PDFInfo
- Publication number
- CN104484251B CN104484251B CN201410763670.9A CN201410763670A CN104484251B CN 104484251 B CN104484251 B CN 104484251B CN 201410763670 A CN201410763670 A CN 201410763670A CN 104484251 B CN104484251 B CN 104484251B
- Authority
- CN
- China
- Prior art keywords
- sector
- hard disk
- bad track
- data
- failure
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The embodiment of the present invention provides a kind of processing method and processing device of hard disk failure, it is related to technical field of memory, it can solve in the prior art since hard disk reporting fault has delay, hard disc data is lost caused by cannot handling in time hard disk failure so that the problem of storage system poor reliability.Concrete scheme is：Receive the fault message that hard disk reports, fault message includes the address information for the sector for generating bad track, and record the address information of sector, after sector is repaired successfully, sector is scanned according to address property information cycle, it determines whether sector generates bad track again after repairing successfully, if the number that sector generates bad track reaches frequency threshold value, data reconstruction is carried out to the data block where sector.The embodiment of the present invention is for handling hard disk failure.
Description
Technical field
The present invention relates to technical field of memory more particularly to a kind of processing method and processing devices of hard disk failure.
Background technology
Redundant array of independent disks (Redundant Array of Independent Disks, RAID), is a kind of
The disk group or hard disk groups that the independent disk of polylith or hard disk groups collectively form, are referred to as logic hard disk, are one
The high-end storage array of kind.As high-end storage systems nowadays are higher and higher to the reliability requirement of hard disk in storage array, high-end storage
Array by being written and reading data simultaneously on multiple hard disks, to improve the data throughout of storage system, increases average event
Hinder interval time, and fault tolerance is increased by memory image backup or the associated redundant datas such as verification data, such as when depositing
When storing up a certain hard disk failure in array, the data in faulty hard disk can be restored according to the redundant data in other hard disks, avoided
The loss of data in faulty hard disk, to improve the reliability of storage system.
Currently, the most still mechanical hard disk of hard disk in high-end storage array, and due to spread of voltage, improper pass
Machine, high temperature, the reasons such as vibrations when read-write may cause the sector in hard disk to generate bad track.If hard disk has existed bad track, when
When upper layer host business reads the data in bad track, it can just trigger hard disk and bad track is found by itself testing mechanism, and by hard disk
Failure is reported to such as controller of the fault treating apparatus in storage array.That is, hard disk will not be stood after bad track appearance
Hardware fault is reported, until upper layer host business can just report hard disk failure, i.e. hard disk reporting fault to have when reading the bad track
Delay.Since the controller etc. in the storage array after hard disk reporting fault could carry out hard disk failure respective handling, and it is hard
There is delay in disk reporting fault, thus controller cannot in time be handled hard disk failure.If after hard disk generates bad track,
Other hard disks in the bad track where associated redundant data are stored before reporting hard disk failure, in storage array also to break down,
The bad track can not then be repaired according to redundant data, the data in bad track can not be restored, to cause in faulty hard disk
Loss of data, reduce the reliability of storage system.
Invention content
The embodiment of the present invention provides a kind of processing method and processing device of hard disk failure, can solve in the prior art due to hard
Disk reporting fault has delay, and hard disc data is lost caused by cannot handling in time hard disk failure so that storage system
The problem of poor reliability.
In order to achieve the above objectives, the present invention adopts the following technical scheme that：
In a first aspect, a kind of processing method of hard disk failure is provided, including：
The fault message that hard disk reports is received, the fault message includes the address information for the sector for generating bad track, and remembers
Record the address information of the sector；
After the sector is repaired successfully, the sector is scanned according to described address property information cycle, is determined
Whether the sector generates bad track again after repairing successfully；
If the number that the sector generates bad track reaches frequency threshold value, data are carried out to the data block where the sector
Reconstruct.
With reference to first aspect, described to where the sector in the first mode in the cards of first aspect
Data block carries out data reconstruction：
By the clear area of the data block migration where the bad track to this hard disk or other hard in addition to the hard disk
The clear area of disk, and data block described in the hard disk is isolated.
With reference to first aspect or the first possible realization method of first aspect, second in first aspect are possible real
In existing mode, the data block to where the sector carries out data reconstruction and includes：
If the number that the sector generates bad track reaches the frequency threshold value, the sector is continued to repair；
If continuing to repair successfully to the sector, the sector is verified；
If verification failure, will verify the number of failure in the presently described hard disk unit interval and predetermined threshold value compares
Compared with to carry out data reconstruction to the data block where the sector when the number is less than or equal to the predetermined threshold value.
The possible realization method of with reference to first aspect the first, in the third mode in the cards of first aspect
In, the method further includes：
Determine whether the quantity of the data block of presently described hard disk isolation is more than default isolation quantity；
If more than then migrating the whole disk data of the hard disk to the clear area of other hard disks, and by the hard disk
It is set to failure.
Second of possible realization method with reference to first aspect, in the 4th kind of mode in the cards of first aspect
In, the method further includes：
If the number of verification failure is more than the predetermined threshold value in the hard disk unit interval, the whole of the hard disk is migrated
Disk data are set to failure to the clear area of other hard disks, and by the hard disk.
Second aspect provides a kind of processing unit of hard disk failure, including：
Receiving unit, the fault message reported for receiving hard disk, the fault message include the sector for generating bad track
Address information；
Recording unit, the address information for recording the sector；
First determination unit is used for after the sector is repaired successfully, according to the address information of the recording unit records
Periodically the sector is scanned, determines whether the sector generates bad track again after repairing successfully；
Reconfiguration unit, if first determination unit determines that the number of the sector generation bad track reaches frequency threshold value,
Data reconstruction is carried out to the data block where the sector.
In conjunction with second aspect, in the first mode in the cards of second aspect, the reconfiguration unit is specifically used for：
By the clear area of the data block migration where the bad track to this hard disk or other hard in addition to the hard disk
The clear area of disk, and data block described in the hard disk is isolated.
In conjunction with the possible realization method of the first of second aspect or second aspect, second in second aspect may be real
In existing mode, further include：
Unit is repaired, if determining that the number of the sector generation bad track reaches number threshold for first determination unit
Value, then continue to repair to the sector；
Verification unit carries out school if continuing to repair successfully to the sector for the reparation unit to the sector
It tests；
Second determination unit, if for verification unit verification failure, by school in the presently described hard disk unit interval
The number for testing failure is compared with predetermined threshold value, whether is less than or equal to the predetermined threshold value with the determination number；
The reconfiguration unit is specifically used for, if second determination unit determination is, to the data where the sector
Block carries out data reconstruction.
In conjunction with the first possible realization method of second aspect, in the third mode in the cards of second aspect
In, further include：
Third determination unit, for determining whether the quantity of the data block of presently described hard disk isolation is more than default isolation number
Amount；
First processing units, if it is to migrate the whole disk data of the hard disk extremely to be determined for the third determination unit
The clear area of other hard disks, and the hard disk is set to failure.
In conjunction with second of possible realization method of second aspect, in the 4th kind of mode in the cards of second aspect
In, further include：
4th determination unit, for determining it is described default whether the number for verifying failure in the hard disk unit interval is more than
Threshold value；
Second processing unit, if it is to migrate the whole disk data of the hard disk extremely to be determined for the 4th determination unit
The clear area of other hard disks, and the hard disk is set to failure.
The embodiment of the present invention provides a kind of processing method and processing device of hard disk failure, by periodically to generating bad track simultaneously
It repairs successful sector to be scanned, determines whether the sector generates bad track again, can find bad track in time, avoid existing
Hard disk needs that itself detection could be carried out when upper layer host business being waited for read the bad track in technology, is deposited so as to cause reporting fault
The problem of postponing；Data weight is then carried out by the data block where reaching the sector of frequency threshold value to the number for generating bad track
Structure can restore to avoid in the prior art there are the data in the sector of potential bad track in current sector and data block in time
Since present hard discs reporting fault has delay, if storing phase in the bad track of current bad track or current data block before reporting fault
The problem of other hard disks for closing redundant data also break down, then the data in this bad track of hard disk can not be restored, you can to solve
In the prior art since hard disk reporting fault has delay, hard disc data is lost caused by cannot handling in time hard disk failure
It loses so that the problem of storage system poor reliability.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, other are can also be obtained according to these attached drawings
Attached drawing.
Fig. 1 is a kind of structural schematic diagram of storage array RAID-5 in the prior art；
Fig. 2 is a kind of process flow figure of hard disk failure provided in an embodiment of the present invention；
Fig. 3 is a kind of process flow figure for hard disk failure that another embodiment of the present invention provides；
Fig. 4 is a kind of process flow figure for hard disk failure that another embodiment of the present invention provides；
Fig. 5 is a kind of processing unit structure chart for hard disk failure that another embodiment of the present invention provides；
Fig. 6 is the processing unit structure chart of another hard disk failure provided in an embodiment of the present invention；
Fig. 7 is the processing unit structure chart of another hard disk failure provided in an embodiment of the present invention；
Fig. 8 is the processing unit structure chart of another hard disk failure provided in an embodiment of the present invention；
Fig. 9 is a kind of processing unit structure chart for hard disk failure that another embodiment of the present invention provides.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Term " hard disk " and " disk " involved in the embodiment of the present invention have essentially identical meaning.Disk is to pass through magnetism
It is written and read the equipment disk of function, can be non-volatile memory medium, the file preserved after power-off will not be lost.Hard disk is
The tablet storage of disk is attached in hard metal box preferably to be protected.
One group hard disc is usually connected by controller and constitutes a storage array, high-end storage by high-end storage systems nowadays
Array is applied to the higher scene of security request data, thus its reliability requirement is higher.Hard disk in storage array is most
For mechanical hard disk, that is, disk, it is made of the circular discs of painting magnetic material, the storage organization of disc is organized in tracks,
Magnetic track refers to covering the concentric circles in all read-write regions on disc.Magnetic track is divided into smaller read-write according to 512 bytes
Unit, i.e. sector.With the rotation of disk, when read/write head is moved on specific magnetic track, the data of respective sectors in magnetic track
It can be handled by read/write head.
The processing method and processing device for the hard disk failure that various embodiments of the present invention provide can be applied to superfluous using independent disk
The high-end storage array of remaining array RAID technique, including first generation RAID technique RAID1.0 and second generation RAID technique
RAID2.0, wherein RAID can be based on software can also be hardware based.The embodiment of the present invention can be applied to respectively
Kind RAID combinations, are identified, such as RAID-0, RAID-1, RAID-1E, RAID-5, RAID-6, RAID-7 with RAID level,
RAID-10, RAID-50.Different grades of recovery and Preservation tactics may be implemented in different RAID level, meet performance and safety
A variety of needs.The number of hard disk and storage mode needed for various RAID level are known to the public, repeat no more.
Redundant array of independent disks abbreviation disk array RAID, by the way that the independent hard disk of polylith (physical hard disk) is pressed difference
Mode combines to form a hard disk groups (logic hard disk), to provide storage performance more higher than single hard disk and provide number
According to the technology of redundancy.RAID carries out subregion in entire array to data, and extensive by hard disk parity data offer data
Multiple function.It, can be according to other hard in parity data and array when a hard disk is broken down due to generation bad track
Data on disk calculate data lost in faulty hard disk bad track, to carry out bad track reparation.Wherein, can have in RAID
The storage region of multiple hard disks, memory capacity, such as 64k, 128k that several contiguous sectors are constituted etc. is properly termed as one point
Item, the memory capacity that several continuous magnetic tracks are constituted, such as the storage region of 64M are properly termed as a data block (Chunk), across
The relevant data block of logical place of all hard disks is properly termed as a band more in RAID groups, and wherein logical place correlation can be with
Think that the number of the data block being located in same band in a hard disk is identical, but physical location in a hard disk is not necessarily identical.
Illustratively, Fig. 1 provides a kind of structural schematic diagram of RAID-5, and related data is stored in the difference of same band by RAID-5
Upper, including user data and the verification data of hard disk, such as user data can be stored in hard disk 1 to hard disk 4 and be located at item
It, can be by user data according to checking algorithm (such as exclusive or) as shown in the D1 in Fig. 1 in the data block for being 1 with the number in 1
In the data block that the number that the verification data that calculating obtains is stored in hard disk 5 is 1, as shown in the P1 in figure.In same band
In, a data data in the block can be obtained by the data data in the block of other hard disks by checking algorithm, thus its
Its data data in the block are properly termed as the associated redundant data of data in current data block.Certainly, in the embodiment of the present invention
Storage array is not limited in RAID-5.
The embodiment of the present invention can also be applied to the storage system of the miscellaneous equipments such as low, middle-end storage array or server
In, the embodiment of the present invention does not limit.
Hard disk failure generally follows following rule：If a certain sector of hard disk generates bad track, adjacent contiguous sector
The possibility for generating bad track is big, such as may lead to scratch continuous multiple sectors productions in region magnetic sheet due to being scratched etc.
Raw bad track；If the bad track of unrepairable occur in multiple sectors in hard disk, the possibility of hard disk failure is big.
In the prior art, when any sector generates bad track in hard disk, bad track of hard disk will not be reported immediately, but works as upper layer
When host service accesses current sector, hard disk can just detect the bad track by itself testing mechanism, to report the bad track
To such as controller of the fault treating apparatus in storage array.The judgment mechanism whether hard disk fails can there are many, it is exemplary
Can consider that hard disk fails when the sector for generating bad track reaches certain quantity, or when the components such as magnetic head in hard disk damage
Bad when, can consider that hard disk fails, and when upper layer host business is detected hard disk (such as booting detection), hard disk just meeting
It reports hard disk to fail, thus hard disk failure is reported to equally exist delay.That is, hard disk reporting fault has delay, here
Failure may include reporting bad track of hard disk and hard disk being reported to fail.
The embodiment of the present invention provides a kind of processing method of hard disk failure, and referring to Fig. 2, step may include mainly：
201, device receives the fault message that hard disk reports, and fault message includes the address information for the sector for generating bad track,
And record the address information of sector.
Wherein, device here is the processing unit of hard disk failure, such as can be the controller in storage array.
Due to spread of voltage, abnormal shutdown, high temperature, the reasons such as vibrations, may lead to the sector in hard disk when read-write
It cannot normally be accessed or cannot normally be read and write, i.e., hard disk generates bad track.The device can be by storing in other hard disks
Associated redundant data carries out bad track reparation, to reconstruct the slitting where generating the sector or sector of bad track to when front disk
In hot standby region, loss of data is prevented.Wherein, since sector is to be divided into the memory capacity of hard disk according to storage address
The storage region of 512 bytes, thus, illustratively, the sector institute for generating bad track can be recorded by records storage address
Position.For example, when upper layer host business reads the data in bad track, hard disk is reported for the first time by itself testing mechanism
When generating the sector of bad track, the address information recording of the sector can be got off, in order to pass through the sector progress to the position
Scanning Detction determines if to generate bad track again.
202, after device is repaired successfully in sector, sector is scanned according to address property information cycle, determines sector
Whether after repairing successfully bad track is generated again.
For the sector after being repaired successfully in step 201, periodically the sector can be scanned, such as can be into
Row reads (read) scanning, if can normally be read in read scanning processes, shows that the sector is normal, does not generate again bad
Road；If cannot normally read, show that the sector generates bad track again.Periodicity therein refers to, can to it is any generate it is bad
Road is simultaneously repaired one time cycle of successful sector settings, and respectively to any fan for generating bad track after being spaced the specified time
Area carries out periodic scan, or to all generation bad tracks and can also repair successful sector settings period at the same time,
And after being spaced the specified time, periodic scan is carried out successively to all sectors for generating bad track.
It is of course also possible to including to the slitting where the sector of generation bad track or the sector including currently generating bad track
The region that multiple contiguous sectors are constituted is scanned, and the embodiment of the present invention does not limit the specific range of scanning.
It should be noted that if determining that current sector does not generate bad track again after repairing successfully, then can wait for down
The Scanning Detction in one period, to determine whether current sector generates bad track again.
The embodiment of the present invention to generating bad track and repairing successful sector and be scanned by periodically actively determining
Whether the sector generates bad track again, can find bad track in time, avoids hard disk in the prior art and needs to wait for upper layer host
Business could carry out itself detection when reading the bad track, there are problems that delay so as to cause reporting fault.
If 203, the number of sector generation bad track reaches frequency threshold value, device carries out data to the data block where sector
Reconstruct.
Wherein, the frequency threshold value in this step can be 2,3 or other values, can specifically be set according to actual conditions,
The embodiment of the present invention does not limit.Due to hard disk it is a certain generate bad track sector near sector may also often will produce it is bad
, that is, there is potential bad track in road, such as may cause to scratch due to being scratched etc. magnetic sheet including region includes current bad track
Continuous multiple sectors generate bad tracks.If the number that sector generates bad track reaches frequency threshold value, illustrate that current sector generates again
The possibility of bad track is big, and the poor reliability of current sector, the sector adjacent with current sector is iteratively produced the possibility of bad track
Greatly, thus can be by the data block where current sector, i.e., nearby continuous multiple sectors carry out data reconstructions for current sector, keep away
Exempt from the data potential bad track in the block and causes loss of data.Illustratively, when the frequency threshold value is 2, that is, current sector is determined
Bad track is generated again after first time repairs successfully, that is to say, that, can be to sector institute when second of the generation bad track in current sector
Data block carry out data reconstruction.
Optionally, carrying out data reconstruction to the data block where sector may include：By the data block migration where bad track
The clear area of other hard disks to the clear area of this hard disk or in addition to this hard disk, and data block in isolation hard disk.
Wherein, the data block in hard disk is isolated, can refers to the mapping pass for cancelling upper layer host business with the data block
System, so, upper layer host business will not be written and read operation to the data block again, i.e., the data block will be no longer used.
In addition, carrying out data reconstruction to the data block where sector can also include, according to the migration position of data block where bad track, weight
New structure redundancy check data, to constitute redundancy relationship with the data in other hard disks.
The embodiment of the present invention carries out data reconstruction by the data block where the sector to generating bad track again, can be timely
Restore the data in bad track, avoids in the prior art since present hard discs reporting fault has delay, if before reporting fault
The other hard disks for storing associated redundant data in current bad track or the potential bad track of data block also break down, then in this bad track of hard disk
Data the problem of can not restoring, you can, cannot be in time to hard to solve in the prior art since hard disk reporting fault has delay
Hard disc data is lost caused by disk failure is handled so that the problem of storage system poor reliability.
The embodiment of the present invention provides a kind of processing method of hard disk failure, by periodically to generate bad track and repair at
The sector of work(is scanned, and determines whether the sector generates bad track again, can be found bad track in time, be avoided in the prior art
Hard disk needs that itself detection could be carried out when upper layer host business being waited for read the bad track, exists so as to cause reporting fault and postpones
The problem of；Data reconstruction is then carried out by the data block where reaching the sector of frequency threshold value to the number for generating bad track, it can
To restore to avoid in the prior art due to working as there are the data in the sector of potential bad track in current sector and data block in time
Preceding hard disk reporting fault has delay, if storing relevant redundancy in the bad track of current bad track or current data block before reporting fault
The problem of other hard disks of data also break down, then the data in this bad track of hard disk can not be restored, you can to solve existing skill
Since hard disk reporting fault has delay in art, hard disc data is lost caused by cannot handling in time hard disk failure, makes
The problem of obtaining storage system poor reliability.
The embodiment of the present invention will be described the processing method of hard disk failure by taking RAID-5 shown in Fig. 1 as an example, ginseng
See Fig. 3, may include following key step：
301, device receives the fault message that hard disk reports, and fault message includes the address information for the sector for generating bad track,
And record the address information of sector.
Wherein, device here is the processing unit of hard disk failure, such as can be the controller in storage array.
Due to spread of voltage, abnormal shutdown, high temperature, many reasons such as vibrations, may cause in hard disk when read-write
Sector generates bad track.Bad track reparation can be carried out by the associated redundant data stored in other hard disks by generating the hard disk of bad track,
To restore the data in bad track, loss of data is prevented.Wherein, when upper layer host business reads the data in bad track, hard disk
By itself testing mechanism, when reporting the sector for generating bad track for the first time, the sector position for generating bad track can be recorded below
Storage address information determines if to generate bad track again in order to be detected by the sector to the position.
Illustratively, in step 301, carrying out bad track reparation to the sector A for generating bad track in the data block of hard disk 1 can be with
The hot standby sector for specifying a certain sector B to be used as sector A by rebuilding mapping relations in the hot standby region of present hard discs, and
According to the associated redundant data and checking algorithm (such as exclusive or algorithm) in other hard disks in the B of sector, restores current and generate
Data in the sector of bad track simultaneously rebuild data redundancy relationship.After rebuilding mapping relations, the sector B in hot standby region
The sector A being remapped in data block 1.Meanwhile cancelling reflecting for upper layer host business and the sector A in former data block 1
Relationship is penetrated, upper layer host business will not be written and read operation to the sector A in former data block 1 again, i.e., sector A in former data block 1
It is isolated and is no longer used.
Wherein, the sector near the sector of bad track is generated since hard disk is a certain to will also tend to generate bad track, it is possible to exist
Potential bad track, such as it includes continuous more including current bad track that may lead to scratch region magnetic sheet due to being scratched etc.
A sector generates bad track.Thus, when the sector A in hard disk generates bad track, the contiguous sector near the A of sector also generates bad track
Possibility is big, when carrying out bad track reparation by the hot standby region of hard disk, can carry out data reconstruction to the sector A for generating bad track,
Can also be to multiple contiguous sectors near the A of sector, such as slitting where the A of sector carries out data reconstruction, the embodiment of the present invention
It does not limit.
In addition, due to other hard disk failures or associated redundant data loss etc., for generating bad track and repairing failure
Sector, it is believed that the sector unrepairable, and bad block label (Bad Sector Tag, BST) is carried out to the sector, to
It abandons using the sector.
302, after sector is repaired successfully, device is scanned sector according to address property information cycle, determines sector
Whether after repairing successfully bad track is generated again.
For the sector after being repaired successfully in step 301, illustratively, can be determined according to the storage address information of record
It generates bad track and repairs the position of successful sector, (be actually periodically then to be mapped as fanning in hot standby region to sector A
The position of the sector B of area A) carry out read scannings show that the sector is normal if can normally be read in read scanning processes,
Bad track is not generated again；If cannot normally read, show that the sector generates bad track again.Periodicity therein refers to, can
It is right respectively to any generation bad track and to repair one time cycle of successful sector settings, and after being spaced the specified time
Any sector for generating bad track carries out periodic scan, or can also to all sector settings period at the same time, and
After being spaced the specified time, periodic scan is carried out successively to all sectors for generating bad track.
It should be noted that being only to determine sector after repairing successfully whether again by read scannings in the embodiment of the present invention
The secondary one way in which for generating bad track can also use other manner, the embodiment of the present invention not to limit.
It is of course also possible to including to the slitting where the sector of generation bad track or the sector including currently generating bad track
The region that multiple contiguous sectors are constituted is scanned, and the embodiment of the present invention does not limit the specific range of scanning.
The embodiment of the present invention by periodically actively to generate bad track and repair successful sector carry out read scannings come
It determines whether the sector generates bad track again, can find bad track in time, avoid hard disk in the prior art and need to wait for upper layer
Host service could carry out itself detection when reading the bad track, there are problems that delay so as to cause reporting fault.
If 303, the number of sector generation bad track reaches frequency threshold value, device carries out data to the data block where sector
Reconstruct.
When generating bad track due to a certain sector of hard disk, near there may be potential bad tracks for contiguous sector.If is produced from sector
The number of raw bad track reaches frequency threshold value, illustrate current sector generate again bad track possibility it is big, the reliability of current sector
Difference, the possibility that the sector adjacent with current sector is iteratively produced bad track is big, thus can by the data block where current sector,
I.e. nearby continuous multiple sectors carry out data reconstruction for current sector, avoid the data potential bad track in the block that data is caused to lose
It loses.Illustratively, when the frequency threshold value is 2, that is, it is bad to determine that current sector generates again after first time bad track reparation success
When road, then data reconstruction can be carried out to the data block where sector.Certainly, the frequency threshold value in this step can be 3 or its
It is worth, and can specifically be set according to actual conditions, the embodiment of the present invention does not limit.
Wherein, carrying out data reconstruction to the data block where sector may include：Extremely by the data block migration where bad track
The clear area of this hard disk or the clear area of other hard disks in addition to this hard disk, and data block in isolation hard disk.
It here can be by the clear area in this hard disk or the clear area of other hard disks in addition to this hard disk to production
Data block where the sector of raw bad track carries out data reconstruction, you can with by the sky of the data block migration where bad track to this hard disk
Not busy region or the clear area of other hard disks in addition to this hard disk.
Can refer to cancelling upper layer host business and the data block specifically, the data block in hard disk is isolated
Mapping relations, upper layer host business will not be written and read operation to the data block again, i.e., the data block is no longer used.In addition,
Carrying out data reconstruction to the data block where sector can also include, according to the migration position of data block where bad track, structure again
Redundancy check data is built, to constitute redundancy relationship with the data in other hard disks.
The embodiment of the present invention carries out data reconstruction by the data block where the sector to generating bad track again, can be timely
Restore the data in bad track, avoids in the prior art since present hard discs reporting fault has delay, if before reporting fault
The other hard disks for storing associated redundant data in current bad track or the potential bad track of data block also break down, then in this bad track of hard disk
Data the problem of can not restoring, you can, cannot be in time to hard to solve in the prior art since hard disk reporting fault has delay
Hard disc data is lost caused by disk failure is handled so that the problem of storage system poor reliability.
It should be noted that if the number that sector generates bad track is not up to frequency threshold value, then continue to repair, if repairing
Whether Scanning Detction that is successful then waiting for next period is generated with to determine current sector in Scanning Detction next time bad again
Road, and determine whether the number for generating bad track reaches frequency threshold value；BST labels are carried out if repairing failure.
304, device determines whether the quantity of the data block of present hard discs isolation is more than default isolation quantity, if more than then
The whole disk data of migration hard disk are set to failure to the clear area of other hard disks, and by hard disk.
In this step, however, it is determined that the quantity of the data block of present hard discs isolation is more than default isolation quantity (such as can be with
3), then show in present hard discs have multiple and different positions data sector in the block repairs successfully after another generation it is bad
Road, thus the hard disk generate again bad track possibility it is big, hard disk failure possibility it is big, the reliability of hard disk is low, so as to
To predict that present hard discs will fail.The whole disk data of hard disk before hard disk failure, can be thus migrated in advance to other hard
Hard disk is then actively set to failure, to notify user's removable disk, so as to will currently will in advance by the clear area of disk in advance
In Data Migration to the hard disk areas of safety in the hard disk of failure, avoid in the prior art due to present hard discs reporting fault
There is delay so that after hard disk failure, report other hard disks of associated redundant data in storage hard disk before hard disk failure
It breaks down, then the problem of data in this hard disk can not be restored.
Illustratively, hard disk is set to failure in advance in the embodiment of the present invention, it can be by by having in the disk state
Effect state is set to failure state, and is reported to user.Wherein, the state of hard disk may include：It can be normal to characterize hard disk
The effective status of work, and the failure state for showing hard disk cisco unity malfunction.
It should be noted that here migration hard disk whole disk data to the clear area of other hard disks in other hard disks
When, it can be migrated to the clear area of some hard disk, can also be migrated to the clear area of multiple hard disks, the present invention is real
Example is applied not limit.
The embodiment of the present invention provides a kind of processing method of hard disk failure, by periodically to generate bad track and repair at
The sector of work(is scanned, and determines whether the sector generates bad track again, can be found bad track in time, be avoided in the prior art
Hard disk needs that itself detection could be carried out when upper layer host business being waited for read the bad track, exists so as to cause reporting fault and postpones
The problem of；Data reconstruction is then carried out by the data block where reaching the sector of frequency threshold value to the number for generating bad track, it can
To restore in time in current sector and data block, there are the data in the sector of potential bad track；In turn, in present hard discs isolation
When the quantity of data block is more than default isolation quantity, by the whole disk data of migration hard disk to the clear area of other hard disks, and
Hard disk is set to failure, it, can be in advance by the Data Migration in the hard disk that will currently fail to safety to notify user's removable disk
Hard disk areas in, avoid in the prior art since hard disk reporting fault has delay, if being stored before reporting fault current
The problem of other hard disks of associated redundant data also break down in hard disk, then the data in this hard disk can not be restored, you can with
It solves in the prior art since hard disk reporting fault has delay, hard disk number caused by cannot handling in time hard disk failure
According to loss so that the problem of storage system poor reliability.
The embodiment of the present invention will be described the processing method of hard disk failure by taking RAID-5 shown in Fig. 1 as an example, ginseng
See Fig. 4, may include following key step：
401, device receives the fault message that hard disk reports, and fault message includes the address information for the sector for generating bad track,
And record the address information of sector.
402, after sector is repaired successfully, device is scanned sector according to address property information cycle, determines sector
Whether after repairing successfully bad track is generated again.
Wherein, the detailed process of step 401~402 may refer to step 301~302.
If 403, the number of sector generation bad track reaches frequency threshold value, device continues to repair to sector.
In this step, if the number that sector generates bad track reaches frequency threshold value, in the hot standby region of present hard discs,
Continue to repair the slitting where current sector or current sector, specific repair process may refer to step 301.It is exemplary
, when the frequency threshold value is 2, that is, determine that current sector generates bad track again after first time bad track reparation success, that is,
It says in current sector after second generates bad track, continues to repair the sector.Certain frequency threshold value can also be it
Its numerical value, the embodiment of the present invention do not limit.
It should be noted that if the number that sector generates bad track is not up to frequency threshold value, then continue to repair, if repairing
Whether Scanning Detction that is successful then waiting for next period is generated with to determine current sector in Scanning Detction next time bad again
Road, and determine whether the number for generating bad track reaches frequency threshold value；BST labels are carried out if repairing failure.
If 404, device continues to repair successfully to sector, sector is verified.
If continuing to repair successfully to current sector in step 403, can school be carried out to the sector immediately in step 404
It tests.For example, can carrying out Verify verifications according to certain checking algorithm such as exclusive or algorithm to the sector, being continued with detecting
Whether the sector after repairing successfully is readable, i.e. detection continues whether the sector after repairing successfully can generate bad track again.It is exemplary
, the associated redundant data stored in other hard disks in same band in data and array in the sector can be read, and
XOR operation is carried out, if exclusive or result is 0, illustrates that the sector after continuing to repair successfully is readable, does not generate bad track again, then school
Success is tested, the Scanning Detction in next period is continued waiting for；Otherwise, illustrate that the sector after continuing to repair successfully is unreadable, it is current to fan
Area generates bad track again, i.e. current sector is failed immediately after continuing to repair successfully, verification failure.
It should be noted that if repairing failure is continued to sector, it may be considered that the sector unrepairable, thus can be right
It carries out BST labels.
If 405, device verification failure, the number that failure is verified in the present hard discs unit interval is carried out with predetermined threshold value
Compare, then enters step 406 or step 407.
406, when number is less than or equal to predetermined threshold value, device carries out data reconstruction to the data block where sector.
If verifying failure in step 404, show that current sector is failed or current sector immediately after continuing to repair successfully
There is only writeable unreadable situation, at this time it is considered that current sector unrepairable.Again since current sector may nearby deposit
In potential bad track, current sector unrepairable, then nearby sector generates bad track and the possibility of unrepairable is big for current sector, this
When the number that failure is verified in the present hard discs unit interval is compared with predetermined threshold value, if it is less than or equal to default threshold
Value, although illustrate present hard discs occur the sector unrepairable of bad track possibility it is big, which still passes through data weight
Data block where sector of the structure to generating bad track migrates, and the data block is isolated, thus can be to where current sector
Data block carry out data reconstruction, so as to restore in time in current sector and data block, there are in the sector of potential bad track
Data avoid in the prior art since present hard discs reporting fault has delay, if storing current bad track before reporting fault
Or other hard disks of associated redundant data also break down in the potential bad track of data block, then the data in this bad track of hard disk can not be extensive
Multiple problem, you can with solve in the prior art since hard disk reporting fault has delay, cannot in time to hard disk failure at
Hard disc data is lost caused by reason so that the problem of storage system poor reliability.
If 407, the number of verification failure is more than predetermined threshold value in the hard disk unit interval, device migrates the whole disk number of hard disk
According to the clear area of extremely other hard disks, and hard disk is set to failure.
If it is determined that the number of verification failure is more than predetermined threshold value (such as 3 times per hour) in the hard disk unit interval, then it can be with
Show in present hard discs there are multiple sectors for occurring bad track repeatedly, and sector unrepairable, thus the sector in present hard discs
Bad track is generated again and the possibility of unrepairable is big, and the possibility of hard disk failure is big, and the reliability of hard disk is low.It thus can be
Before hard disk failure, the whole disk data of hard disk are migrated in advance to the clear area of other hard disks, are then actively set to hard disk in advance
Failure, to notify user's removable disk, so as in advance by the hard disk of the Data Migration in the hard disk that will currently fail to safety
In region, avoid in the prior art after the failure of this hard disk, before hard disk reporting fault, relevant redundancy in storage hard disk
When other hard disks of data also break down, the problem of data in this hard disk can not be restored.
It is similar with step 304, hard disk is set to failure in advance in the embodiment of the present invention, it can also be by by the disk state
In effective status be set to failure state, and be reported to user.Wherein, the state of hard disk may include：To characterize hard disk energy
The effective status enough worked normally, and the failure state for showing hard disk cisco unity malfunction.And migrate the whole disk of hard disk
When other hard disks in data to the clear area of other hard disks, it can be migrated to the clear area of some hard disk,
It can migrate to the clear area of multiple hard disks, the embodiment of the present invention does not limit.
The embodiment of the present invention provides a kind of processing method of hard disk failure, by periodically to generate bad track and repair at
The sector of work(is scanned, and determines whether the sector generates bad track again, can be found bad track in time, be avoided in the prior art
Hard disk needs that itself detection could be carried out when upper layer host business being waited for read the bad track, exists so as to cause reporting fault and postpones
The problem of；It is then verified by reaching frequency threshold value to the number for generating bad track and continuing to repair successful sector, and right
Data block where verifying the sector of failure carries out data reconstruction, can restore that there are potential in current sector and data block in time
Data in the sector of bad track；In turn, when verifying the number of failure within the present hard discs unit interval more than predetermined threshold value, pass through
The whole disk data of migration hard disk are set to failure to the clear area of other hard disks, and by hard disk, to notify user's removable disk, Ke Yiti
In the preceding Data Migration by the hard disk that will currently fail to the hard disk areas of safety, avoid in the prior art due to hard disk
Reporting fault has delay, if event also occurs for the other hard disks for storing associated redundant data in present hard discs before reporting fault
The problem of barrier, the then data in this hard disk can not be restored, you can to solve in the prior art since hard disk reporting fault has delay,
Hard disc data is lost caused by cannot handling in time hard disk failure so that the problem of storage system poor reliability.
The embodiment of the present invention provides a kind of processing unit 500 of hard disk failure, which is used to handle hard disk failure,
Can be the controller in storage array, as shown in figure 5, the device 500 may include：
Receiving unit 501, can be used for receiving the fault message that hard disk reports, and fault message includes the sector for generating bad track
Address information.
Recording unit 502 can be used for recording the sector that bad track is generated in the fault message that receiving unit 501 receives
Address information.
First determination unit 503, can be used for after sector is repaired successfully, be believed according to the address that recording unit 502 records
Breath is periodically scanned sector, determines whether sector generates bad track again after repairing successfully.
Reconfiguration unit 504, if can be used for the first determination unit 503 determines that the number of sector generation bad track reaches number threshold
Value then carries out data reconstruction to the data block where sector.
Optionally, reconfiguration unit 504 can be specifically used for：
By the clear area of the data block migration where bad track to this hard disk or the sky of other hard disks in addition to this hard disk
Not busy region, and data block in isolation hard disk.
Optionally, referring to Fig. 6, device 500 can also include：
Unit 505 is repaired, if can be used for the first determination unit 503 determines that the number of sector generation bad track reaches number threshold
Value, then continue to repair to sector；
Verification unit 506 continues to repair successfully, be verified to sector if can be used for repairing unit 505 to sector；
Second determination unit 507 will be in the present hard discs unit interval if can be used for the verification failure of verification unit 506
The number of verification failure is compared with predetermined threshold value, whether is less than or equal to predetermined threshold value with determined number；
Reconfiguration unit 504 can be specifically used for, if the determination of the second determination unit 507 is, to the data block where sector
Carry out data reconstruction.
Optionally, referring to Fig. 7, device 500 can also include：
Third determination unit 508, be determined for the data block of present hard discs isolation quantity whether be more than it is default every
From quantity；
First processing units 509, if can be used for the determination of third determination unit 508 is the whole disk data for migrating hard disk
To the clear area of other hard disks, and hard disk is set to failure.
Optionally, referring to Fig. 8, device 500 can also include：
Whether 4th determination unit 510 is determined for the number of verification failure in the hard disk unit interval more than default
Threshold value；
Second processing unit 511 can be used for, if the determination of the 4th determination unit 510 is the whole disk data for migrating hard disk
To the clear area of other hard disks, and hard disk is set to failure.
The embodiment of the present invention provides a kind of processing unit 500 of hard disk failure, can be used for handling hard disk failure, passes through week
It is scanned to generating bad track and repairing successful sector to phase property, determines whether the sector generates bad track again, it can be timely
It was found that bad track, avoids hard disk in the prior art and needs that itself inspection could be carried out when upper layer host business being waited for read the bad track
It surveys, there are problems that delay so as to cause reporting fault；Then reach the sector of frequency threshold value by the number to generation bad track
The data block at place carries out data reconstruction, can restore the data in bad track in time, avoids in the prior art due to current hard
Disk reporting fault has delay, if storing associated redundant data in the bad track of current bad track or current data block before reporting fault
Other hard disks also break down, then the problem of data in this bad track of hard disk can not be restored, you can to solve in the prior art
Since hard disk reporting fault has delay, hard disc data is lost caused by cannot handling in time hard disk failure so that deposits
The problem of storage system poor reliability.
The embodiment of the present invention provides a kind of processing unit 900 of hard disk failure, which is used to handle hard disk failure,
Can be the controller in storage array, as shown in figure 9, the device 900 may include：Bus 904；And it is connected to bus
904 processor 901, memory 902, receiver 903；Wherein the memory 902 is for storing instruction and data；The receiver
903 execution instructions include the address information for the sector for generating bad track for receiving the fault message that hard disk reports, fault message；
The processor 901 executes the instruction and is used for, and the address of the sector of bad track is generated in the fault message that record identifier receiver 903 receives
Information；The processor 901 executes the instruction and is additionally operable to, after sector is repaired successfully, according to the address information of recording unit records
Periodically sector is scanned, determines whether sector generates bad track again after repairing successfully；The processor 901 executes should
Instruction is additionally operable to, however, it is determined that the number that sector generates bad track reaches frequency threshold value, then carries out data to the data block where sector
Reconstruct；The memory 902 executes the instruction and is additionally operable to, the address information for the sector that storage processor 901 records.
In embodiments of the present invention, optionally, the processor 901 execute the instruction for the data block where sector into
Row data reconstruction includes：
By the clear area of the data block migration where bad track to this hard disk or the sky of other hard disks in addition to this hard disk
Not busy region, and data block in isolation hard disk.
In embodiments of the present invention, optionally, the processor 901 execute the instruction for the data block where sector into
Row data reconstruction includes：
If the number that sector generates bad track reaches frequency threshold value, sector is continued to repair；
If continuing to repair successfully to sector, sector is verified；
If verification failure, will verify the number of failure in the present hard discs unit interval and predetermined threshold value is compared, with
Data reconstruction is carried out to the data block where sector when number is less than or equal to predetermined threshold value.
In embodiments of the present invention, optionally, which executes the instruction and is additionally operable to：
Determine whether the quantity of the data block of present hard discs isolation is more than default isolation quantity；
If more than the whole disk data for then migrating hard disk are set to failure to the clear area of other hard disks, and by hard disk.
In embodiments of the present invention, optionally, which executes the instruction and is additionally operable to：
If the number of verification failure is more than predetermined threshold value in the hard disk unit interval, the whole disk data of hard disk are migrated to other
The clear area of hard disk, and hard disk is set to failure.
The embodiment of the present invention provides a kind of processing unit 900 of hard disk failure, can be used for handling hard disk failure, passes through week
It is scanned to generating bad track and repairing successful sector to phase property, determines whether the sector generates bad track again, it can be timely
It was found that bad track, avoids hard disk in the prior art and needs that itself inspection could be carried out when upper layer host business being waited for read the bad track
It surveys, there are problems that delay so as to cause reporting fault；Then reach the sector of frequency threshold value by the number to generation bad track
The data block at place carries out data reconstruction, can restore the data in bad track in time, avoids in the prior art due to current hard
Disk reporting fault has delay, if storing associated redundant data in the bad track of current bad track or current data block before reporting fault
Other hard disks also break down, then the problem of data in this bad track of hard disk can not be restored, you can to solve in the prior art
Since hard disk reporting fault has delay, hard disc data is lost caused by cannot handling in time hard disk failure so that deposits
The problem of storage system poor reliability.
In several embodiments provided herein, it should be understood that disclosed method and apparatus can pass through it
Its mode is realized.For example, apparatus embodiments described above are merely indicative, for example, the division of the unit, only
Only a kind of division of logic function, formula that in actual implementation, there may be another division manner, such as multiple units or component can be tied
Another system is closed or is desirably integrated into, or some features can be ignored or not executed.Another point, it is shown or discussed
Mutual coupling, direct-coupling or communication connection can be the INDIRECT COUPLING or logical by some interfaces, device or unit
Letter connection can be electrical, machinery or other forms.
In addition, in equipment in each embodiment of the present invention, each functional unit can be integrated in a processing unit,
Can also be that the independent physics of each unit includes, it can also be during two or more units be integrated in one unit.And it is above-mentioned
Each unit both may be used hardware form realize, can also be realized in the form of hardware adds SFU software functional unit.
Realize that all or part of step of above method embodiment can be completed by the relevant hardware of program instruction, it is preceding
The program stated can be stored in a computer read/write memory medium, and when being executed, execution includes above method reality to the program
The step of applying；And storage medium above-mentioned includes：USB flash disk, mobile hard disk, read-only memory (English：Read Only
Memory, abbreviation：ROM), random access memory (English：Random Access Memory, abbreviation：RAM), magnetic disc or
The various media that can store program code such as CD.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (10)
1. a kind of processing method of hard disk failure, which is characterized in that including：
The fault message that hard disk reports is received, the fault message includes the address information for the sector for generating bad track, and records institute
State the address information of sector；
After the sector is repaired successfully, the sector is scanned according to described address property information cycle, described in determination
Whether sector generates bad track again after repairing successfully；
If the number that the sector generates bad track reaches frequency threshold value, data weight is carried out to the data block where the sector
Structure.
2. according to the method described in claim 1, it is characterized in that, the data block to where the sector carries out data weight
Structure includes：
By the clear area of the data block migration where the bad track to this hard disk or other hard disks in addition to the hard disk
Clear area, and data block described in the hard disk is isolated.
3. method according to claim 1 or 2, which is characterized in that the data block to where the sector is into line number
Include according to reconstruct：
If the number that the sector generates bad track reaches the frequency threshold value, the sector is continued to repair；
If continuing to repair successfully to the sector, the sector is verified；
If verification failure, will verify the number of failure in the presently described hard disk unit interval and predetermined threshold value is compared, with
Data reconstruction is carried out to the data block where the sector when the number is less than or equal to the predetermined threshold value.
4. according to the method described in claim 2, it is characterized in that, the method further includes：
Determine whether the quantity of the data block of presently described hard disk isolation is more than default isolation quantity；
If more than the whole disk data for then migrating the hard disk are set to the clear area of other hard disks, and by the hard disk
Failure.
5. according to the method described in claim 3, it is characterized in that, the method further includes：
If the number of verification failure is more than the predetermined threshold value in the hard disk unit interval, the whole disk number of the hard disk is migrated
According to the clear area of other hard disks extremely in addition to the hard disk, and the hard disk is set to failure.
6. a kind of processing unit of hard disk failure, which is characterized in that including：
Receiving unit, the fault message reported for receiving hard disk, the fault message include the address for the sector for generating bad track
Information；
Recording unit, the ground for recording the sector for generating bad track described in the fault message that the receiving unit receives
Location information；
First determination unit was used for after the sector is repaired successfully, according to the address information period of the recording unit records
Property the sector is scanned, determines whether the sector generates bad track again after repairing successfully；
Reconfiguration unit, if first determination unit determines that the number of the sector generation bad track reaches frequency threshold value, to institute
Data block where stating sector carries out data reconstruction.
7. device according to claim 6, which is characterized in that the reconfiguration unit is specifically used for：
By the clear area of the data block migration where the bad track to this hard disk or other hard disks in addition to the hard disk
Clear area, and data block described in the hard disk is isolated.
8. the device described according to claim 6 or 7, which is characterized in that further include：
Unit is repaired, if determining that the number of the sector generation bad track reaches frequency threshold value for first unit, to the fan
Area continues to repair；
Verification unit verifies the sector if continuing to repair successfully to the sector for the reparation unit；
Second determination unit is lost if for verification unit verification failure by being verified in the presently described hard disk unit interval
The number lost is compared with predetermined threshold value, whether is less than or equal to the predetermined threshold value with the determination number；
The reconfiguration unit is specifically used for, if second determination unit determines that the number is less than or equal to the default threshold
Value then carries out data reconstruction to the data block where the sector.
9. device according to claim 7, which is characterized in that further include：
Third determination unit, for determining whether the quantity of the data block of presently described hard disk isolation is more than default isolation quantity；
First processing units, if it is default to determine that the quantity of the data block of presently described hard disk isolation is more than for third determination unit
Quantity is isolated, then the whole disk data for migrating the hard disk are set to mistake to the clear area of other hard disks, and by the hard disk
Effect.
10. device according to claim 8, which is characterized in that further include：
4th determination unit, for determining whether the number for verifying failure in the hard disk unit interval is more than the default threshold
Value；
Second processing unit, if being determined for the 4th determination unit, the number that failure is verified in the hard disk unit interval is big
In the predetermined threshold value, then the clear area of other hard disks of the whole disk data of the hard disk extremely in addition to the hard disk is migrated,
And the hard disk is set to failure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410763670.9A CN104484251B (en) | 2014-12-11 | 2014-12-11 | A kind of processing method and processing device of hard disk failure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410763670.9A CN104484251B (en) | 2014-12-11 | 2014-12-11 | A kind of processing method and processing device of hard disk failure |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104484251A CN104484251A (en) | 2015-04-01 |
| CN104484251B true CN104484251B (en) | 2018-10-02 |
Family
ID=52758795
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410763670.9A Active CN104484251B (en) | 2014-12-11 | 2014-12-11 | A kind of processing method and processing device of hard disk failure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104484251B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105702273B (en) * | 2016-02-29 | 2018-07-03 | 四川效率源信息安全技术股份有限公司 | The method of bad failure after being got well before a kind of reparation Hitachi hard disk |
| CN107577545B (en) * | 2016-07-05 | 2021-02-02 | 北京金山云网络技术有限公司 | Method and device for detecting and repairing fault disk |
| CN107731262A (en) * | 2017-09-29 | 2018-02-23 | 郑州云海信息技术有限公司 | A kind of storage disk bad block management method, device, equipment and readable storage medium storing program for executing |
| CN108182126B (en) * | 2018-01-02 | 2020-11-20 | 联想(北京)有限公司 | Hard disk processing method and electronic equipment |
| CN110018783B (en) * | 2018-01-09 | 2022-12-20 | 阿里巴巴集团控股有限公司 | Data storage method, device and system |
| CN108536548B (en) | 2018-04-10 | 2020-12-29 | 网宿科技股份有限公司 | Method and device for processing bad track of disk and computer storage medium |
| CN109298979B (en) * | 2018-08-17 | 2022-08-12 | 新华三技术有限公司成都分公司 | Hard disk fault processing method and device |
| CN110879761A (en) | 2018-09-05 | 2020-03-13 | 华为技术有限公司 | Hard disk fault processing method, array controller and hard disk |
| CN109582237B (en) * | 2018-11-30 | 2021-12-17 | 上海工程技术大学 | Bad sector mapping method based on mechanical hard disk |
| CN109799953B (en) * | 2018-12-14 | 2022-10-11 | 深圳创新科技术有限公司 | Method and device for rapidly rebuilding RAID (redundant array of independent disks) |
| CN111274070B (en) * | 2019-11-04 | 2021-10-15 | 华为技术有限公司 | Hard disk detection method and device and electronic equipment |
| CN112835745B (en) * | 2021-02-09 | 2022-04-01 | 天津易鼎丰动力科技有限公司 | High-reliability storage method of embedded system |
| CN113190179B (en) * | 2021-05-26 | 2022-02-11 | 北京自由猫科技有限公司 | Method for prolonging service life of mechanical hard disk, storage device and system |
| CN113608673A (en) * | 2021-06-23 | 2021-11-05 | 南京科海智博信息技术有限公司 | Electronic file long-term storage system |
| CN116664116B (en) * | 2023-07-25 | 2023-11-14 | 江苏华存电子科技有限公司 | Recovery detection method and system for returned firmware |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1707683A (en) * | 2004-06-10 | 2005-12-14 | 英业达股份有限公司 | Rigid disc state monitoring method |
| CN102945191A (en) * | 2012-11-15 | 2013-02-27 | 浪潮电子信息产业股份有限公司 | RAID5 (redundant array of independent disk 5) data transfer method |
| CN103400602A (en) * | 2013-07-24 | 2013-11-20 | 大唐移动通信设备有限公司 | Automatic hard disk bad track repairing method and equipment |
| CN103700387A (en) * | 2013-09-18 | 2014-04-02 | 福建鼎恩协创电子科技有限公司 | Device and method for monitoring and repairing damaged sector of hard disc |
| US8880472B2 (en) * | 2012-02-15 | 2014-11-04 | Tivarri Ltd. | Method of backing-up, and making available, electronic data and software initially stored on a client server |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102750213B (en) * | 2012-06-18 | 2016-04-06 | 深圳市锐明技术股份有限公司 | Disk detects, disposal route and detection, disposal system |
| CN103389918A (en) * | 2013-07-24 | 2013-11-13 | 北京鲸鲨软件科技有限公司 | Repair method for false fault in RAID (Redundant Array of Independent Disks) system |
-
2014
- 2014-12-11 CN CN201410763670.9A patent/CN104484251B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1707683A (en) * | 2004-06-10 | 2005-12-14 | 英业达股份有限公司 | Rigid disc state monitoring method |
| US8880472B2 (en) * | 2012-02-15 | 2014-11-04 | Tivarri Ltd. | Method of backing-up, and making available, electronic data and software initially stored on a client server |
| CN102945191A (en) * | 2012-11-15 | 2013-02-27 | 浪潮电子信息产业股份有限公司 | RAID5 (redundant array of independent disk 5) data transfer method |
| CN103400602A (en) * | 2013-07-24 | 2013-11-20 | 大唐移动通信设备有限公司 | Automatic hard disk bad track repairing method and equipment |
| CN103700387A (en) * | 2013-09-18 | 2014-04-02 | 福建鼎恩协创电子科技有限公司 | Device and method for monitoring and repairing damaged sector of hard disc |
Non-Patent Citations (1)
| Title |
|---|
| OceanStor S5000 系列 存储系统 特性描述_(V100R005C02);华为技术有限公司;《http://support.huawei.com/enterprise/docinforeader.action?contentId=DOC0000662048&partNo=10052》;20120730;73、74页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104484251A (en) | 2015-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104484251B (en) | A kind of processing method and processing device of hard disk failure | |
| US10025666B2 (en) | RAID surveyor | |
| CN102708019B (en) | Method, device and system for hard disk data recovery | |
| US8589724B2 (en) | Rapid rebuild of a data set | |
| US6959413B2 (en) | Method of handling unreadable blocks during rebuilding of a RAID device | |
| CN104035830B (en) | A kind of data reconstruction method and device | |
| CN103309775B (en) | A kind of fault-tolerance approach of high-reliability disk array | |
| US10013321B1 (en) | Early raid rebuild to improve reliability | |
| CN100530125C (en) | Safety storage method for data | |
| CN101916173B (en) | RAID (Redundant Array of Independent Disks) based data reading and writing method and system thereof | |
| US8930750B2 (en) | Systems and methods for preventing data loss | |
| TWI258661B (en) | Efficient media scan operations for storage systems | |
| US9104604B2 (en) | Preventing unrecoverable errors during a disk regeneration in a disk array | |
| CN101840360A (en) | Rapid reconstruction method and device of RAID (Redundant Array of Independent Disk) system | |
| CN102508733B (en) | A kind of data processing method based on disk array and disk array manager | |
| CN103699457A (en) | Method and device for restoring disk arrays based on stripping | |
| CN102207895A (en) | Data reconstruction method and device of redundant array of independent disk (RAID) | |
| US20060215456A1 (en) | Disk array data protective system and method | |
| CN111816239B (en) | Disk detection method and device, electronic equipment and machine-readable storage medium | |
| CN106990918A (en) | Trigger the method and device that RAID array is rebuild | |
| CN105183590A (en) | Disk array fault tolerance processing method | |
| CN107122261B (en) | Data reading and writing method and device of storage equipment | |
| JP4605374B2 (en) | Storage device testing method and apparatus | |
| CN111381997A (en) | RAID reconstruction method and device | |
| CN101359299A (en) | Write operation method for mapping cheap magnetic disc redundant array equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |