CN105988891A - Magnetic disk data repairing method and device - Google Patents
Magnetic disk data repairing method and device Download PDFInfo
- Publication number
- CN105988891A CN105988891A CN201510060560.0A CN201510060560A CN105988891A CN 105988891 A CN105988891 A CN 105988891A CN 201510060560 A CN201510060560 A CN 201510060560A CN 105988891 A CN105988891 A CN 105988891A
- Authority
- CN
- China
- Prior art keywords
- disk
- block
- data
- information
- metadata information
- 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.)
- Granted
Links
Abstract
The invention discloses a magnetic disk data repairing method and a magnetic disk data repairing device, and aims at repairing magnetic disc data if magnetic disc faults happen in a CQFS system. The magnetic disk data repairing method comprises the following steps: in the embodiment of the invention, when confirming that the CQFS system has a magnetic disc with faults, implementing quick format operation on the magnetic disc with faults, and confirming corresponding new meta data information on the basis of in-block description information of a data zone retained after quick format in the magnetic disc with faults, wherein the meta data information is used for describing basic information of a data block and corresponds to the in-block description information of the data block one to one; resetting a recycling time stamp in the new meta data information, and storing the new meta data information on the magnetic disc after quick format. Therefore, the invention provides a universal magnetic disc data repairing method for the CQFS system with magnetic disc faults, and after the magnetic disc is repaired, data in all undamaged data zones can be normally read, and can be further normally stored.
Description
Technical field
The present invention relates to computer realm, particularly to a kind of data in magnetic disk restorative procedure and device.
Background technology
File system, can be because some emergency situations (as the situations such as power-off occur) be led during running
Cause file system data is lost, traditional file system meeting configuration inspection instrument, is used for repairing file system number
According to, such as, under Windows operating system, there are new technology file system and CHKDSK checking tool,
Ext3 file system and a fsck.ext3 checking tool is had under linux operating system, and traditional file system
Checking tool can repair partial document system data in some cases, but, once critical data district quilt
Destroying, these instruments will be unable to repair file system data, thus causes whole file system data all to be lost
Lose.
As a example by ext3 file system under linux operating system, in the file system occurring that emergency situations causes
During system loss of data, fsck.ext3 checking tool is used to carry out data in magnetic disk inspection and reparation, in some situation
Under, file system can be repaired；And in some cases, it may appear that the mistakes such as file system superblock damage,
Cause utilizing fsck.ext3 checking tool that whole file system data is recovered.It addition, work as user
When the maloperation of such as disk partition or formatting occurs, also cannot utilize fsck.ext3 checking tool number
According to recovery.
In sum, there is data single point failure point in traditional file systems, once the data on this trouble point are sent out
Raw abnormal, whole file system arises that globality fault, and traditional file systems checking tool also cannot be right
Whole file system data recovers.Further, for round-robin queue's file system (Circular File
System, CQFS) prior art also do not generates a kind of effective data in magnetic disk restorative procedure.
Summary of the invention
The embodiment of the present invention provides a kind of data in magnetic disk restorative procedure and device, deposits in order to solving in prior art
Disk failure time the problem that cannot recover of the data that are not destroyed.
The concrete technical scheme that the embodiment of the present invention provides is as follows:
A kind of data in magnetic disk restorative procedure, including:
Determine time round-robin queue's file system CQFS exists failed disk, described failed disk is performed fast
Speed format manipulation；
Describing information in the block of data field based on the described failed disk retained after quick formatting, it is right to determine
The new metadata information answered, described metadata information is for describing the essential information of data block, with data block
Block in description information one_to_one corresponding；
Reset the recovery time stamp in described new metadata information, described new metadata information is saved in
On disk after quick formatting.
So, it is possible to provide data in magnetic disk restorative procedure general during a kind of CQFS system disk fault, at magnetic
After dish reparation completes, all not destroyed Data Area data all can normally read, and just can proceed with
Often storage.
It is preferred that farther include:
Before described failed disk is performed quick formatting operation, obtain from CQFS running log
The original disk information of described failed disk, described original disk information at least includes: the disk before fault
Path, disk UUID and be ultimately written Block ID.
It is preferred that after performing quick formatting operation, before determining new metadata information, further
Including:
Disk UUID in the CQFS head of the disk after quick formatting is defined as described failed disk
Disk UUID.
In such manner, it is possible to ensure that the disk that user identifies after repair terminates is same.
It is preferred that describe letter in the block of data field based on the described failed disk retained after quick formatting
Breath, determines the new metadata information of correspondence, specifically includes:
By in the block of each data block in the data field of the described failed disk retained after quick formatting
Description information is compared with corresponding new metadata information；
When the field in description information in the block of any one data block with in corresponding new metadata information
Corresponding field inconsistent time, in the block according to described data block, information is described, by the newest metadata
Information is updated.
It is preferred that the recovery time stamp reset in described new metadata information, specifically include:
In original disk information according to described failed disk, record is ultimately written Block ID, at quick lattice
Disk after formula positions corresponding data block；
Recovery time in new metadata information corresponding for corresponding data block is stabbed and is set to zero；This
Sample, can be updated to again write by the current location of corresponding data block original position during data；
Be set to the data block of zero based on recovery time stamp, successively by corresponding for other all data blocks described newly
Metadata information recovery time stamp reset in order.
So, recovery spatial order can be made to meet original order.
A kind of data in magnetic disk prosthetic device, including:
Quick formatting unit, in time there is failed disk in determining CQFS, to described failed disk
Execution quick formatting operates；
Metadata recovery unit, for data field based on the described failed disk retained after quick formatting
Description information in block, determines the new metadata information of correspondence, and described metadata information is used for describing data block
Essential information, describe information one_to_one corresponding in the block of data block, and reset described new metadata letter
Recovery time stamp in breath；
Storage unit, the disk after described new metadata information is saved in quick formatting.
So, it is possible to provide data in magnetic disk restorative procedure general during a kind of CQFS system disk fault, at magnetic
After dish reparation completes, all not destroyed Data Area data all can normally read, and just can proceed with
Often storage.
It is preferred that farther include:
Acquiring unit, for, before described failed disk is performed quick formatting operation, transporting from CQFS
The original disk information of the described failed disk obtained in row daily record, described original disk information at least includes:
Disk path, disk UUID before fault and be ultimately written Block ID.
It is preferred that perform quick formatting operation after, before determining new metadata information, described soon
Speed formatting unit is further used for:
Disk UUID in the CQFS head of the disk after quick formatting is defined as described failed disk
Disk UUID.
In such manner, it is possible to ensure that the disk that user identifies after repair terminates is same.
It is preferred that describe letter in the block of the data field of the described failed disk retained after based on quick formatting
Breath, when determining the new metadata information of correspondence, described metadata recovery unit specifically for:
By in the block of each data block in the data field of the described failed disk retained after quick formatting
Description information is compared with corresponding new metadata information；
When the field in description information in the block of any one data block with in corresponding new metadata information
Corresponding field inconsistent time, in the block according to described data block, information is described, by the newest metadata
Information is updated.
During it is preferred that the recovery time in resetting described new metadata information stabs, described metadata is recovered
Unit specifically for:
In original disk information according to described failed disk, record is ultimately written Block ID, at quick lattice
Disk after formula positions corresponding data block；
Recovery time in new metadata information corresponding for corresponding data block is stabbed and is set to zero；This
Sample, can be updated to again write by the current location of corresponding data block original position during data；
Be set to the data block of zero based on recovery time stamp, successively by corresponding for other all data blocks described newly
Metadata information recovery time stamp reset in order.
So, recovery spatial order can be made to meet original order.
Accompanying drawing explanation
Fig. 1 is that in the embodiment of the present invention, data in magnetic disk repairs flow chart；
Fig. 2 is system construction drawing in the embodiment of the present invention.
Detailed description of the invention
In the embodiment of the present invention, based on CQFS file system, when occurring that Data Area data is destroyed, defeated
Enter the disk path that failed disk is original, the globally unique identification strings (Universally that disk is original
Unique Identifier, UUID) and it is ultimately written block ID, by failed disk again quick formatting,
I.e. metadata is all eliminated, and will not there is the metadata that failed disk is original and depend in follow-up repair process
Relying, therefore the data in magnetic disk restorative procedure of the embodiment of the present invention has versatility.After quick formatting,
Perform complete disk consistency check, recover metadata from data field, and reset the recovery time of metadata
Stamp, after metadata being deposited, completes disk repair process.This provides one and can repair CQFS system
The generic diskette data recovery method of system disk failure, can protect not destroyed Data Area data, and
Can proceed with normal storage.
In order to be better understood from this programme, below the structure of CQFS file system is described in detail.
One physical disk CQFS is divided into following region:
Table 1
Below the region in table 1 is illustrated:
CQFS Header in table 1 is used for describing CQFS version, disk UUID, data block quantity and
The configuration informations such as data block size information；
The Block i.e. database information of Info Table in table 1 describes list, is used for storing metadata information,
Such as table 2:
Table 2
| Block Info | Block Info | … | Block Info |
Data Area in table 1 is data field, is split the data block into formed objects, such as table 3:
Table 3
| Block Data | Block Data | … | Block Data |
The structure of each Block Data such as table 4 in table 3:
Table 4
| Frame Data Area | Index | Description |
Frame Data Area in table 4 is frame data region, and Index is the index information of data block,
Description is the description information of data block.
The corresponding metadata information (i.e. Block Info) of each data block (i.e. Block Data) in table 3,
Metadata information is database description structure, recites the essential information of respective data blocks, such as table 5:
Table 5
| Block ID | Stream ID | Recycle TS | Begin TS | End TS | Lock Count | … |
Backup Area in table 1 is backup area, is backed up in three regions of other in table 1, as
Table 6:
Table 6
In CQFS file system, CQFS Header size is fixed as 4KB；Block Info size is
32 bytes；Block Info Table size is data block quantity (i.e. block Num) and Block Info
The product of structure size；Block Allocation Table size is fixed as 4000KB；Block Size is several
In formal applied environment, 16MB all it is fixed as according to block size.
Seen from the above description, computing formula such as formula [1] institute of data block data amount (i.e. block Num)
Show:
Block Num=(disk Size-2*header Size-2*block Allocation Table Size)/(2
* Block Info+block Size) 1 formula [1]
Wherein, disk Size refers to disk size, it is seen then that only it is to be understood that disk size, can calculate file
The data block quantity of system, and extrapolate the size of Block Info Table further.
On the premise of calculating the size obtaining Block Info Table, by formula [2] computable number according to district
Side-play amount (i.e. Data Area Offset), formula [2] is as follows:
Data area Offset=block Info Table Offset+block Info Table Size+block
Allocation Table Size formula [2]
From formula [1] and formula [2], only it is to be understood that disk size, the skew of data field can be extrapolated
Amount, further, can calculate this according to the reckoning result of Block ID and the side-play amount of data field
Block deviation post on an entire disc, shown in specific formula for calculation such as formula [3]:
Disk Offset=data Area Offset+block ID*block Size formula [3]
Wherein, disk Offset is the disk skew of corresponding Block, and data Area Offset is data field
Side-play amount (can be calculated by formula [2]), and block Size is that data block size is (in formal applied environment
All it is fixed as 16MB), depending on order that block ID writes according to data block and position.
Can be exited by above formula, the skew of CQFS system regional and disk size and data block
Size is relevant, and data block size is substantially fixing, therefore, only it is to be understood that disk size, so that it may calculate
Go out the side-play amount of regional.It is to say, under CQFS system, at disk size and data block size not
In the case of change, the file system structure that twice formatting of same disk is generated is on all four.By
Format in each time and all can generate different disk UUID, the unit therefore twice formatting inserted
For data, only disk UUID is different, and other data are duplicate.
Above-mentioned characteristic based on CQFS system, the embodiment of the present invention is for CQFS system disk accident design
The universal method that a kind of data in magnetic disk is repaired, refering to shown in Fig. 1, concrete process is as follows:
Step 100: determine when there is failed disk in CQFS system, performs quick format to failed disk
Change operation.
In the embodiment of the present invention, disk failures refers to that data are destroyed, and non-physical is destroyed, as long as
Being that disk size is not changed in, either CQFS Header is destroyed, or CQFS is Block Info Table
Destroyed, all can carry out data reparation by the method for embodiment of the present invention design, not broken after having repaired
Bad data all can be recovered normally to use, and specifically used actual application scenarios at least includes following several:
1) CQFS disk is changed by Windows disk management tool initialization, the most former partition table.
2) CQFS disk is turned to new technology file system by quick format.
3) CQFS disk is turned to CQFS file system by quick format, and now file system is for changing,
But meta-data region resets, user cannot read data.
4) CQFS disk is destroyed more than 16TB, 32-bit operating system.
When CQFS system disk breaks down, for the following operation of this failed disk execution:
First, the original disk information of the described failed disk obtained from CQFS running log, described former
Beginning disc information at least includes: the disk path before fault, disk globally unique identifier character string UUID
And it is ultimately written block ID.
Then, failed disk is performed quick formatting operation.
Specifically, during quick formatting, CQFS metadata information is reinitialized by system,
And the district's data that do not clear data, i.e. can improve the speed of formatting, it is ensured that be formatted in the shortest time
Complete, also protect user data simultaneously and be not eliminated, in order to be continuing with after the complete disk of follow-up reparation.
After to disk formatting, a new UUID can be generated in CQFS Header, therefore exist
After having performed failed disk quick formatting operation, by former in CQFS Header (i.e. CQFS head)
The disk UUID come inserts, and so, can make the CQFS head after quick formatting is that this failed disk is with former
The holding come is consistent, it is possible to ensure that the disk that user identifies after repair terminates is same.
So far, the file system structure that the disk of the disk after quick formatting and initial format is generated
It is on all four, and through inserting original disk UUID so that the data in magnetic disk after twice formatting
Completely the same, and now, the metadata information of failed disk is eliminated.
Step 110: the metadata information of data field based on the failed disk retained after quick formatting, really
Fixed corresponding new metadata information, this metadata information is for describing the essential information of data block, with data
Description information one_to_one corresponding in the block of block.
Failed disk after quick formatting is performed complete disk consistency check, it is thus achieved that new metadata
Information.
Specifically, by the block of each data block in the data field of the failed disk retained after quick formatting
Interior description information is compared with corresponding new metadata information；
When describing the word in (i.e. Description) information in the block of any one data block (i.e. Block)
When section is inconsistent with the corresponding field in corresponding new metadata information, according to describing letter in the block of data block
Breath, is updated the newest metadata information.
As shown in table 4, owing to after quick formatting, the data of data field yet suffer from, therefore, each
Block Data structure has Description information, and, the corresponding number of each data block information
According to block metadata information (metadata information now is the most initialized), will describe in the block of a data block
Information is compared with corresponding data block metadata information, if inconsistent, then retouches according in the block of data block
Information of stating correspondingly writes the fast metadata information of data, it is thus achieved that the new metadata letter that each data block is corresponding
Breath, and new metadata information is preserved.
Step 120: reset recovery time stamp (i.e. Recycle TS) in new metadata information, by new
Metadata information is saved on the disk after quick formatting.
During recovery time stamp in resetting new metadata information, particularly as follows:
In original disk information according to failed disk, record is ultimately written block ID, after quick formatting
Disk positions corresponding data block；
Recovery time in new metadata information corresponding for this corresponding data block is stabbed and is set to zero, should
The current location of corresponding data block is updated to again write original position during data；
The data block of zero it is set to, successively by new unit corresponding for other all data blocks based on recovery time stamp
The recovery time stamp of data message resets in order.
Specifically, from CQFS running log obtain this failed disk original be ultimately written block ID, base
It is ultimately written, in this, the data block that block ID location failed disk is corresponding, this data block is examined in disk concordance
Recovery time stamp in the new metadata information determined after looking into resets, so so that it is reclaim spatial order symbol
Close original order, say, that, it is ensured that after having repaired disk, will not be by when re-writing data
The data cover that originally disk is not destroyed, all can normally read, and can proceed video recording storage meter
Drawing, the Video data recovery order of whole disk reclaims also according to the principle of first in first out.
Such as, whole disk is divided into 100 data blocks, and each data block is allocated a block and uniquely marks
Know (i.e. Block ID), Block ID is respectively 0,1,2 ..., 99, initially carrying out disc format
During change, the metadata information of each data block arranges a recovery time and stabs (such as the Recycle in table 5
TS), now this recovery time stamp is consistent with corresponding Block ID, i.e. the 0th to the 99th in order
Individual data block recovery time stamp be respectively 0,1,2,3 ..., 99, when carrying out the write of data,
Data block recycles, and from the 0th data BOB(beginning of block) distribution, often one data block of distribution is used for recording data
Time, the recovery time in this data block metadata information is stabbed and is set to current maximum recovery time stamp
Numerical value adds 1, such as, when recording data, from the 0th data block (i.e. Block ID is the data block of 0)
Start write, then Block ID be 0 data block metadata information in recovery time stamp be set to current maximum
Recovery time stamp (i.e. 99) add 1, be i.e. set to 100, by Block during the 1st data block write data
ID is that the recovery time stamp in the data block metadata information of 1 is set to current maximum recovery time stamp (i.e.
100) add 1, be i.e. recorded as being 101 ..., it is assumed that the 30th is had been written into when disk failures
(Block ID is 30) data block, the recovery during i.e. this Block ID is the data block metadata information of 30
Timestamp has been set to 130.
So, during disk is repaired, after quick formatting and disk consistency check, former
The metadata begun is cleared, then the stamp of the recovery time in the 0th to the 99th data block metadata information quilt
Again be designated as 0,1,2,3 ..., 99, in this case, when re-writing data, also will
From the 0th data BOB(beginning of block) write, but before disk failures, recorded the number that Block ID is 30
According to block, it should proceed by this position and re-write, in order to avoid this inconsistent situation occurs,
Recovery spatial order is made to meet original order, by the recovery in the new metadata information of these 100 data blocks
Timestamp resets.
The process of concrete replacement is: according to being ultimately written block ID, by the 30th (i.e. Block ID is 30)
In data block, the stamp of the recovery time in metadata information resets to 0, by the 31st metadata that data block is new letter
Recovery time stamp in breath resets to 1, by that analogy, by returning in metadata information new for all data blocks
Receiving timestamp all to reset, the recovery time stamp in the 29th metadata information that data block is new will reset
It is 99.
So, when re-writing data, according to the instruction order of timestamp, will be from the 30th data block
(recovery time stamp is 0) starts write, and is stabbed by the recovery time in the 30th data block metadata information
It is set to 100, when continuing to write to data, will write from the 31st data block, and by the 31st data block
Recovery time stamp in metadata information is set to 101.In this, recovery spatial order can be made to meet original
Sequentially, the most not destroyed Data Area data all can normally read, and continues video recording storage simultaneously, can be normal
Write
Finally, new metadata information is saved on the disk after quick formatting.So, according to renewal
After metadata information can read all not destroyed data of the data field before fault, so far, disk is repaiied
Multiple process is complete.
Based on above-described embodiment, refering to shown in Fig. 2, in the embodiment of the present invention, CQFS system includes quickly
Formatting unit 200, metadata recovery unit 210 and storage unit 220, wherein:
Quick formatting unit 200, in time there is failed disk in determining CQFS, to failed disk
Execution quick formatting operates；
Metadata recovery unit 210, for data field based on the failed disk retained after quick formatting
Description information in block, determines the new metadata information of correspondence, and metadata information is for describing the base of data block
Describe information one_to_one corresponding in this information, with the block of data block, and reset returning in new metadata information
Receive timestamp；
Storage unit 220, the disk after new metadata information is saved in quick formatting.
So, it is possible to provide data in magnetic disk restorative procedure general during a kind of CQFS system disk fault, at magnetic
After dish reparation completes, all not destroyed Data Area data all can normally read, and just can proceed with
Often storage.
It is preferred that farther include:
Acquiring unit 230, for, before failed disk is performed quick formatting operation, transporting from CQFS
The original disk information of the failed disk obtained in row daily record, original disk information at least includes: before fault
Disk path, disk UUID and be ultimately written Block ID.
It is preferred that after performing quick formatting operation, before determining new metadata information, quick lattice
Formula unit 200 is further used for:
Disk UUID in the CQFS head of the disk after quick formatting is defined as the disk of failed disk
UUID。
In such manner, it is possible to ensure that the disk that user identifies after repair terminates is same.
It is preferred that describe information in the block of the data field of the failed disk retained after based on quick formatting,
When determining the new metadata information of correspondence, metadata recovery unit 210 specifically for:
To describe in the block of each data block in the data field of the failed disk retained after quick formatting
Information is compared with corresponding new metadata information；
When the field in description information in the block of any one data block with in corresponding new metadata information
Corresponding field inconsistent time, according to describing information in the block of data block, by the newest metadata information
It is updated.
During it is preferred that the recovery time in resetting new metadata information stabs, metadata recovery unit 210
Specifically for:
In original disk information according to failed disk, record is ultimately written Block ID, in quick formatting
After disk in position corresponding data block；
Recovery time in new metadata information corresponding for corresponding data block is stabbed and is set to zero, so may be used
The current location of corresponding data block is updated to again write original position during data；
The data block of zero it is set to, successively by new unit corresponding for other all data blocks based on recovery time stamp
The recovery time stamp of data message resets in order.
So, recovery spatial order can be made to meet original order.
In sum, in the embodiment of the present invention, determine time CQFS system exists failed disk, to this event
Barrier disk performs quick formatting operation, the block of data field based on the failed disk retained after quick formatting
Interior description information, determines the new metadata information of correspondence, and this metadata information is for describing the base of data block
Information one_to_one corresponding is described, when resetting the recovery in new metadata information in this information, with the block of data block
Between stab, new metadata information is saved on the disk after quick formatting.So, it is possible to provide a kind of
Data in magnetic disk restorative procedure general during CQFS system disk fault, after disk reparation completes, all not
Destroyed Data Area data all can normally read, and can proceed with normal storage.
Those skilled in the art are it should be appreciated that embodiments of the invention can be provided as method, system or meter
Calculation machine program product.Therefore, the present invention can use complete hardware embodiment, complete software implementation or knot
The form of the embodiment in terms of conjunction software and hardware.And, the present invention can use and wherein wrap one or more
Computer-usable storage medium containing computer usable program code (include but not limited to disk memory,
CD-ROM, optical memory etc.) form of the upper computer program implemented.
The present invention is with reference to method, equipment (system) and computer program product according to embodiments of the present invention
The flow chart of product and/or block diagram describe.It should be understood that can by computer program instructions flowchart and
/ or block diagram in each flow process and/or flow process in square frame and flow chart and/or block diagram and/
Or the combination of square frame.These computer program instructions can be provided to general purpose computer, special-purpose computer, embedding
The processor of formula datatron or other programmable data processing device is to produce a machine so that by calculating
The instruction that the processor of machine or other programmable data processing device performs produces for realizing at flow chart one
The device of the function specified in individual flow process or multiple flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions may be alternatively stored in and computer or the process of other programmable datas can be guided to set
In the standby computer-readable memory worked in a specific way so that be stored in this computer-readable memory
Instruction produce and include the manufacture of command device, this command device realizes in one flow process or multiple of flow chart
The function specified in flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions also can be loaded in computer or other programmable data processing device, makes
Sequence of operations step must be performed to produce computer implemented place on computer or other programmable devices
Reason, thus the instruction performed on computer or other programmable devices provides for realizing flow chart one
The step of the function specified in flow process or multiple flow process and/or one square frame of block diagram or multiple square frame.
Although preferred embodiments of the present invention have been described, but those skilled in the art once know base
This creativeness concept, then can make other change and amendment to these embodiments.So, appended right is wanted
Ask and be intended to be construed to include preferred embodiment and fall into all changes and the amendment of the scope of the invention.
Obviously, those skilled in the art can carry out various change and modification to the embodiment of the present invention and not take off
Spirit and scope from the embodiment of the present invention.So, if these amendments of the embodiment of the present invention and modification belong to
Within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention be also intended to comprise these change and
Including modification.
Claims (10)
1. a data in magnetic disk restorative procedure, it is characterised in that including:
Determine time round-robin queue's file system CQFS exists failed disk, described failed disk is performed fast
Speed format manipulation；
Describing information in the block of data field based on the described failed disk retained after quick formatting, it is right to determine
The new metadata information answered, described metadata information is for describing the essential information of data block, with data block
Block in description information one_to_one corresponding；
Reset the recovery time stamp in described new metadata information, described new metadata information is saved in
On disk after quick formatting.
2. the method for claim 1, it is characterised in that farther include:
Before described failed disk is performed quick formatting operation, obtain from CQFS running log
The original disk information of described failed disk, described original disk information at least includes: the disk before fault
Path, disk globally unique identifier character string UUID and be ultimately written block and uniquely identify Block ID.
3. method as claimed in claim 2, it is characterised in that after performing quick formatting operation,
Before determining new metadata information, farther include:
Disk UUID in the CQFS head of the disk after quick formatting is defined as described failed disk
Disk UUID.
4. method as claimed in claim 2, it is characterised in that based on the institute retained after quick formatting
Description information in the block of the data field stating failed disk, determines the new metadata information of correspondence, specifically includes:
By in the block of each data block in the data field of the described failed disk retained after quick formatting
Description information is compared with corresponding new metadata information；
When the field in description information in the block of any one data block with in corresponding new metadata information
Corresponding field inconsistent time, in the block according to described data block, information is described, by the newest metadata
Information is updated.
5. the method as described in claim 2,3 or 4, it is characterised in that reset described new metadata
Recovery time stamp in information, specifically includes:
In original disk information according to described failed disk, record is ultimately written Block ID, at quick lattice
Disk after formula positions corresponding data block；
Recovery time in new metadata information corresponding for corresponding data block is stabbed and is set to zero；
Be set to the data block of zero based on recovery time stamp, successively by corresponding for other all data blocks described newly
Metadata information recovery time stamp reset in order.
6. a data in magnetic disk prosthetic device, it is characterised in that including:
Quick formatting unit, in time there is failed disk in determining CQFS, to described failed disk
Execution quick formatting operates；
Metadata recovery unit, for data field based on the described failed disk retained after quick formatting
Description information in block, determines the new metadata information of correspondence, and described metadata information is used for describing data block
Essential information, describe information one_to_one corresponding in the block of data block, and reset described new metadata letter
Recovery time stamp in breath；
Storage unit, the disk after described new metadata information is saved in quick formatting.
7. device as claimed in claim 6, it is characterised in that farther include:
Acquiring unit, for, before described failed disk is performed quick formatting operation, transporting from CQFS
The original disk information of the described failed disk obtained in row daily record, described original disk information at least includes:
Disk path, disk UUID before fault and be ultimately written Block ID.
8. device as claimed in claim 7, it is characterised in that after performing quick formatting operation,
Before determining new metadata information, described quick formatting unit is further used for:
Disk UUID in the CQFS head of the disk after quick formatting is defined as described failed disk
Disk UUID.
9. device as claimed in claim 7, it is characterised in that retain after based on quick formatting
Description information in the block of the data field of described failed disk, when determining the new metadata information of correspondence, described
Metadata recovery unit specifically for:
By in the block of each data block in the data field of the described failed disk retained after quick formatting
Description information is compared with corresponding new metadata information；
When the field in description information in the block of any one data block with in corresponding new metadata information
Corresponding field inconsistent time, in the block according to described data block, information is described, by the newest metadata
Information is updated.
10. the device as described in claim 7,8 or 9, it is characterised in that resetting described new unit
In data message recovery time stamp time, described metadata recovery unit specifically for:
In original disk information according to described failed disk, record is ultimately written Block ID, at quick lattice
Disk after formula positions corresponding data block；
Recovery time in new metadata information corresponding for corresponding data block is stabbed and is set to zero；
Be set to the data block of zero based on recovery time stamp, successively by corresponding for other all data blocks described newly
Metadata information recovery time stamp reset in order.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510060560.0A CN105988891B (en) | 2015-02-05 | 2015-02-05 | A kind of data in magnetic disk restorative procedure and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510060560.0A CN105988891B (en) | 2015-02-05 | 2015-02-05 | A kind of data in magnetic disk restorative procedure and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105988891A true CN105988891A (en) | 2016-10-05 |
| CN105988891B CN105988891B (en) | 2019-02-12 |
Family
ID=57037423
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510060560.0A Active CN105988891B (en) | 2015-02-05 | 2015-02-05 | A kind of data in magnetic disk restorative procedure and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105988891B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106991021A (en) * | 2017-03-31 | 2017-07-28 | 联想(北京)有限公司 | The method and system of new data file are built from available data file |
| CN110109773A (en) * | 2019-04-11 | 2019-08-09 | 苏州浪潮智能科技有限公司 | A kind of detection method and system of magnetic disk of virtual machine consistency |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1517906A (en) * | 2003-01-14 | 2004-08-04 | 联想(北京)有限公司 | Document system and document management method |
| US20070038915A1 (en) * | 2005-07-28 | 2007-02-15 | Prostor Systems, Inc. | Adaptive archival format |
| US7254677B1 (en) * | 2004-05-04 | 2007-08-07 | Xilinx, Inc. | First-in, first-out memory system with reduced cycle latency |
| CN104090829A (en) * | 2014-08-06 | 2014-10-08 | 浪潮电子信息产业股份有限公司 | Method for realizing logical volume metadata backup storage |
-
2015
- 2015-02-05 CN CN201510060560.0A patent/CN105988891B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1517906A (en) * | 2003-01-14 | 2004-08-04 | 联想(北京)有限公司 | Document system and document management method |
| US7254677B1 (en) * | 2004-05-04 | 2007-08-07 | Xilinx, Inc. | First-in, first-out memory system with reduced cycle latency |
| US20070038915A1 (en) * | 2005-07-28 | 2007-02-15 | Prostor Systems, Inc. | Adaptive archival format |
| CN104090829A (en) * | 2014-08-06 | 2014-10-08 | 浪潮电子信息产业股份有限公司 | Method for realizing logical volume metadata backup storage |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106991021A (en) * | 2017-03-31 | 2017-07-28 | 联想(北京)有限公司 | The method and system of new data file are built from available data file |
| CN106991021B (en) * | 2017-03-31 | 2021-02-19 | 联想(北京)有限公司 | Method and system for constructing new data file from existing data file |
| CN110109773A (en) * | 2019-04-11 | 2019-08-09 | 苏州浪潮智能科技有限公司 | A kind of detection method and system of magnetic disk of virtual machine consistency |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105988891B (en) | 2019-02-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106201338B (en) | Date storage method and device | |
| CN104199750B (en) | A kind of file access pattern method and device of Linux system | |
| US9542431B2 (en) | Cyclic commit transaction protocol | |
| US9183268B2 (en) | Partition level backup and restore of a massively parallel processing database | |
| CN106557266B (en) | Method and apparatus for redundant array of independent disks RAID | |
| CN101308471B (en) | Method and device for data restoration | |
| CN105205053A (en) | Method and system for analyzing database incremental logs | |
| CN106528071B (en) | The choosing method and device of object code | |
| CN108512726B (en) | Data monitoring method and equipment | |
| CN106919477B (en) | Virtual disk analysis method and system | |
| CN105022678A (en) | Data backup method and apparatus for virtual machine | |
| US8799716B2 (en) | Heap dump occurrence detection | |
| US8352439B1 (en) | Database verification following database write | |
| CN108009049B (en) | MYISAM storage engine deleted record offline recovery method and storage medium | |
| CN111367926A (en) | Data processing method and device for distributed system | |
| CN111125298A (en) | Method, equipment and storage medium for reconstructing NTFS file directory tree | |
| CN112088366A (en) | Host-aware update write | |
| CN105988891A (en) | Magnetic disk data repairing method and device | |
| CN106155838A (en) | A kind of database back-up data restoration methods and device | |
| CN106156076B (en) | The method and system of data processing | |
| CN105892954A (en) | Data storage method and device based on multiple copies | |
| CN108427675B (en) | Method and equipment for constructing index | |
| CN109542860B (en) | Service data management method based on HDFS and terminal equipment | |
| US8924773B1 (en) | Reducing file system data unavailability window by adapting hierarchical recovery framework | |
| CN102004761B (en) | Data storage method and system |
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 |