CN104407806B - The amending method and device of RAID group hard disc information - Google Patents
The amending method and device of RAID group hard disc information Download PDFInfo
- Publication number
- CN104407806B CN104407806B CN201410527853.0A CN201410527853A CN104407806B CN 104407806 B CN104407806 B CN 104407806B CN 201410527853 A CN201410527853 A CN 201410527853A CN 104407806 B CN104407806 B CN 104407806B
- Authority
- CN
- China
- Prior art keywords
- request
- raid group
- communication equipment
- group hard
- disk
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0683—Plurality of storage devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/2017—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where memory access, memory control or I/O control functionality is redundant
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0629—Configuration or reconfiguration of storage systems
Abstract
The embodiment of the present invention provides a kind of amending method and device of RAID group hard disc information.This method includes:The first I/O Request that communication equipment is issued according to terminal accesses the first RAID group hard discs;First RAID group hard discs carry the first reference count of the number of times for identifying the first I/O Request that communication equipment is received;When communication equipment judges any hard disk failure in the first RAID group hard discs or failed in preset time, communication equipment changes the first RAID group hard disc information, obtains the 2nd RAID group hard discs, and do not start the first operation;The second I/O Request that communication equipment receiving terminal is issued, and the 2nd RAID group hard discs are accessed according to the second I/O Request, until all first I/O Requests of the first reference count correspondence number of times are returned after result, communication device initiated first operation.Method provided in an embodiment of the present invention, the continuity of terminal traffic is ensure that in the case where data are not lost.
Description
Technical field
The present embodiments relate to the communication technology, more particularly to a kind of modification of RAID group hard disc information
Method and apparatus.
Background technology
RAID (Redundant Arrays of Inexpensive Disks, RAID) group hard disc is
The redundant array that a kind of lower-cost hard disk different by polylith is constituted, under the operating system that main frame and communication equipment are constituted
It is to occur as an independent large storage facilities.Assuming that RAID group hard discs include tri- hard disks of A, B, C, communication equipment exists
When receiving the write request of external host transmission, the partial data in the data that will be carried in the write request is write in disk C, and
Another part data are write on other at least one hard disks in RAID group hard discs (assuming that writing on hard disk B), so made
Some hard disk in RAID group hard discs when breaking down or failing, communication equipment can be reconstructed or in advance by RAID group hard discs
The method of copy ensures that communication equipment will not be influenceed by hard disk corruptions or failure.By taking the reconstruct of RAID group hard discs as an example, that is, lead to
Equipment is believed when the disk C in detecting RAID group hard discs breaks down, and communication equipment modification RAID group hard disc information communicate
Equipment reselects a new hard disk (being assumed to be hard disk D) and is associated as HotSpare disk, and by the HotSpare disk and faulty disk
So that HotSpare disk is added into RAID group hard discs;After communication equipment changes RAID group hard discs information, communication equipment is read on hard disk B
The another part data of storage, the data for obtaining faulty disk C are calculated using redundancy, and will calculate obtained faulty disk C
Data be re-written on HotSpare disk (now RAID group hard discs include other disks, faulty disk and HotSpare disk);Finally, communicate
Equipment replaces faulty disk with HotSpare disk, and faulty disk is deleted from RAID group hard discs.
But, the modification of RAID group hard disc information is relate in the process of the reconstruct of above-mentioned RAID group hard discs or pre-copy, and
Communication equipment is also required to during reconstruct (or pre-copy) while handling host service, i.e., while needing processing main frame to issue
Main frame input and output (Input/Output, hereinafter referred to as IO) request (request can be write request), it is therefore desirable to ensure
The correctness for the RAID group hard disc information that main frame is accessed.Because if communication equipment is while modification RAID groups, one side is again to master
The main frame IO processing that eedle is issued to RAID groups, the data that main frame IO may be caused to carry can not be correctly written into corresponding hard
On disk, cause hard disc data slitting inconsistent;Or so that main frame can not correctly read the data on hard disk.
Prior art is to solve the above problems, employ following scheme:Communication equipment is determining modification RAID group hard disc letters
When breath, new main frame IO is hung up and (i.e. during RAID group hard disc information is changed, suspends main to new-comer
Machine IO processing so that new main frame IO is waited for), until being handed down to the main frame IO of RAID group hard discs before
Return to result, the main frame IO of communication equipment ability start to process suspension.
But, prior art during hanging the main frame IO of the RAID group hard discs in it can cause host service of short duration
It is disconnected, and if while communication equipment changes RAID group hard disc information, the main frame IO that main frame is issued in RAID groups is more,
The time for so waiting main frame IO to return will be longer, and the time of host computer service interruption also can be longer.
The content of the invention
The embodiment of the present invention provides a kind of amending method and device of RAID group hard disc information, to solve
The technical problem that the terminal traffic that certainly prior art is caused during RAID group hard disc information is changed is interrupted.
In a first aspect, the present invention provides a kind of amending method of RAID group hard disc information, including:
The first input and output I/O Request that communication equipment is issued according to terminal accesses the first RAID RAID
Group hard disc;Wherein, the first RAID group hard discs carry the first reference count, and first reference count is used to identify described lead to
The number of times for the first I/O Request that letter equipment is received;
When the communication equipment judges any hard disk failure in the first RAID group hard discs or the mistake in preset time
During effect, the communication equipment changes the first RAID group hard disc information, obtains the 2nd RAID group hard discs, and do not start the first behaviour
Make;Wherein, first operation is that reconstructed operation or pre-copy are operated, and the 2nd RAID group hard discs include the first RAID
Group hard disc and HotSpare disk;
The communication equipment receives the second I/O Request that the terminal is issued, and according to being accessed second I/O Request
2nd RAID group hard discs, until all first I/O Requests of first reference count correspondence number of times return to result
Afterwards, communication device initiated first operation;Wherein, second I/O Request is communication equipment modification described first
The I/O Request received after RAID group hard disc information.
With reference in a first aspect, in the first possible embodiment of first aspect, if first I/O Request and institute
The second I/O Request is stated for write request, then first operation is reconstructed operation;When the communication equipment judges the first RAID
Any hard disk failure in group hard disc or when being failed in preset time, the communication equipment changes the first RAID group hard discs
Information, obtains the 2nd RAID group hard discs, and does not start the first operation, specifically includes:
When the communication equipment judges any hard disk failure in the first RAID group hard discs, the communication equipment will
The HotSpare disk is associated with the hard disk, obtains the 2nd RAID group hard discs, and do not start the reconstructed operation;
The first data carried in first I/O Request are write the first RAID group hard discs by the communication equipment.
With reference to the first possible embodiment of first aspect, in second of possible embodiment of first aspect
In, the communication equipment receives the second I/O Request that the terminal is issued, and accesses described second according to second I/O Request
RAID group hard discs, until all first I/O Requests of first reference count correspondence number of times are returned after result, institute
Communication device initiated first operation is stated, is specifically included:
The communication equipment receives the second I/O Request that the terminal is issued, and by carried in second I/O Request the
Two data write the 2nd RAID group hard discs;
After all first I/O Requests of first reference count correspondence number of times return to result, the communication
Equipment is by the data reconstruction of the hard disk to the HotSpare disk;
The hard disk is replaced with the HotSpare disk by the communication equipment, and by the hard disk from the 2nd RAID group hard discs
Remove, obtain the 3rd RAID group hard discs;
The communication equipment receives the 3rd I/O Request, and the 3rd data carried in the 3rd I/O Request write-in is described
3rd RAID group hard discs;Wherein, the 3rd request is obtaining being followed by for the 3rd RAID group hard discs for the communication equipment
The I/O Request received.
With reference in a first aspect, in the third possible embodiment of first aspect, if first I/O Request and institute
The second I/O Request is stated for read request, then first operation is pre-copy operation;When the communication equipment judges described first
Any hard disk failure in RAID group hard discs or when being failed in preset time, the communication equipment changes the first RAID groups
Hard disk information, obtains the 2nd RAID group hard discs, and does not start the first operation, specifically includes:
When any hard disk that the communication equipment is judged in the first RAID group hard discs fails in preset time, institute
State communication equipment to be associated the HotSpare disk with the hard disk, obtain the 2nd RAID group hard discs, and do not start described
Pre-copy is operated;
The communication equipment reads the data in the first RAID group hard discs according to first I/O Request.
With reference to the third possible embodiment of first aspect, in the 4th kind of possible embodiment of first aspect
In, the communication equipment receives the second I/O Request that the terminal is issued, and accesses described second according to second I/O Request
RAID group hard discs, until all first I/O Requests of first reference count correspondence number of times are returned after result, institute
Communication device initiated first operation is stated, is specifically included:
The communication equipment receives the second I/O Request that the terminal is issued, and according to being read second I/O Request
Data in 2nd RAID group hard discs;
After all first I/O Requests of first reference count correspondence number of times return to result, the communication
Equipment is by the data copy of the hard disk to the HotSpare disk;
The hard disk is replaced with the HotSpare disk by the communication equipment, and the hard disk is hard from the 2nd RAID groups
Removed in disk, obtain the 4th RAID group hard discs;
The communication equipment receives the 4th I/O Request, and reads the 4th RAID group hard discs according to the 4th I/O Request
In data;Wherein, the 4th I/O Request is that the communication equipment is received after the 4th RAID group hard discs are obtained
I/O Request.
Second aspect, the present invention provides a kind of modification device of RAID group hard disc information, including:
First access modules, the first input and output I/O Request for being issued according to terminal accesses the first independent disk redundancy
Array RAID group hard discs;Wherein, the first RAID group hard discs carry the first reference count, and first reference count is used to mark
Know the number of times for the first I/O Request that the communication equipment is received;
Processing module, for judge when the communication equipment any hard disk failure in the first RAID group hard discs or
When being failed in preset time, the first RAID group hard disc information is changed, the 2nd RAID group hard discs are obtained, and do not start the first behaviour
Make;Wherein, first operation is that reconstructed operation or pre-copy are operated, and the 2nd RAID group hard discs include the first RAID
Group hard disc and HotSpare disk;
Second access modules, are visited for receiving the second I/O Request that the terminal is issued, and according to second I/O Request
The 2nd RAID group hard discs are asked, until all first I/O Request return processing of first reference count correspondence number of times
As a result after, first operation is started;Wherein, second I/O Request is that the processing module changes the first RAID groups firmly
The I/O Request received after disk information.
With reference to second aspect, in the first possible embodiment of second aspect, if first I/O Request and institute
The second I/O Request is stated for write request, then first operation is reconstructed operation;The then processing module, including:
First acquisition unit, for judging any hard disk failure in the first RAID group hard discs when the communication equipment
When, the HotSpare disk is associated with the hard disk, the 2nd RAID group hard discs are obtained, and do not start the reconstruct behaviour
Make;
First writing unit, it is hard that the first data for will be carried in first I/O Request write the first RAID groups
Disk.
With reference to the first possible embodiment of second aspect, in second of possible embodiment of second aspect
In, second access modules, including:
Second writing unit, for receiving the second I/O Request that the terminal is issued, and will take in second I/O Request
Second data of band write the 2nd RAID group hard discs;
Reconfiguration unit, for returning to processing knot when all first I/O Requests of first reference count correspondence number of times
After fruit, by the data reconstruction of the hard disk to the HotSpare disk;
First replaces and removes unit, for the hard disk to be replaced with into the HotSpare disk, and by the hard disk from second
Removed in RAID group hard discs, obtain the 3rd RAID group hard discs;
3rd writing unit, for receiving the 3rd I/O Request, and the 3rd data carried in the 3rd I/O Request is write
Enter the 3rd RAID group hard discs;Wherein, the 3rd request removes unit in acquisition the 3rd RAID groups for described replace
The I/O Request received after hard disk.
With reference to second aspect, in the third possible embodiment of second aspect, if first I/O Request and institute
The second I/O Request is stated for read request, then first operation is pre-copy operation;The processing module, including:
Second acquisition unit, for judging any hard disk in the first RAID group hard discs pre- when the communication equipment
If failed in the time, the HotSpare disk is associated with the hard disk, the 2nd RAID group hard discs are obtained, and do not start
The pre-copy operation;
First reading unit, for reading the data in the first RAID group hard discs according to first I/O Request.
With reference to the third possible embodiment of second aspect, in the 4th kind of possible embodiment of second aspect
In, second access modules, including:
Second reading unit, reads for receiving the second I/O Request that the terminal is issued, and according to second I/O Request
Take the data in the 2nd RAID group hard discs;
Copy cell, for returning to processing knot when all first I/O Requests of first reference count correspondence number of times
After fruit, by the data copy of the hard disk to the HotSpare disk;
Second replaces and removes unit, for the hard disk to be replaced with into the HotSpare disk, and by the hard disk from described the
Removed in two RAID group hard discs, obtain the 4th RAID group hard discs;
3rd reading unit, the 4th RAID groups are read firmly for receiving the 4th I/O Request, and according to the 4th I/O Request
Data in disk;Wherein, the 4th I/O Request is that the communication equipment is received after the 4th RAID group hard discs are obtained
The I/O Request arrived.
The amending method and device of RAID group hard disc information provided in an embodiment of the present invention, communication equipment
The first request issued according to terminal accesses the first RAID group hard discs for carrying the first reference count;When the communication equipment judges
Any hard disk failure in first RAID group hard discs or when being failed in preset time, communication equipment changes the first RAID groups firmly
Disk information, obtains the 2nd RAID group hard discs, and do not start the first operation;When communication equipment changes above-mentioned first RAID group hard discs letter
The second I/O Request that terminal is issued is received after breath, the 2nd RAID group hard discs are accessed according to second I/O Request, until above-mentioned the
All first I/O Requests of the first reference count correspondence number of times of one RAID group hard discs are returned after result, communication device initiated
First operation.Method provided in an embodiment of the present invention, either communication equipment is when changing the first RAID group hard disc information, still
Change after the first RAID group hard disc information, the I/O Request that communication equipment is sent to terminal can be handled, and can't be interrupted
The business that terminal is issued is stated, the continuity of terminal traffic is ensure that in the case where data are not lost.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are this hairs
Some bright embodiments, for those of ordinary skill in the art, without having to pay creative labor, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
The flow of the amending method embodiment one for the RAID group hard disc information that Fig. 1 provides for the present invention is shown
It is intended to;
Fig. 2 reconstructs schematic diagram for the first RAID group hard discs that the present invention is provided;
The flow of the amending method embodiment two for the RAID group hard disc information that Fig. 3 provides for the present invention is shown
It is intended to;
The flow of the amending method embodiment three for the RAID group hard disc information that Fig. 4 provides for the present invention is shown
It is intended to;
The structure of the modification device embodiment one for the RAID group hard disc information that Fig. 5 provides for the present invention is shown
It is intended to;
The structure of the modification device embodiment two for the RAID group hard disc information that Fig. 6 provides for the present invention is shown
It is intended to;
The structure of the modification device embodiment three for the RAID group hard disc information that Fig. 7 provides for the present invention is shown
It is intended to.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
A part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
The flow of the amending method embodiment one for the RAID group hard disc information that Fig. 1 provides for the present invention is shown
It is intended to.The executive agent of this method is communication equipment, and the communication equipment can be server, can be computer, can also be
Common controller entity, the embodiment of the present invention is not limited to the form of communication equipment, as long as hard provided with RAID groups in it
Disk.As shown in figure 1, this method includes:
S101:The first I/O Request that communication equipment is issued according to terminal accesses the first RAID group hard discs;Wherein, described first
RAID group hard discs carry the first reference count, and first reference count is used to identify the first IO that the communication equipment is received
The number of times of request.
Specifically, terminal is set up with communication equipment after being connected, terminal continuously sends the first IO to communication equipment
Request, first I/O Request is used to access the data in the first RAID group hard discs, and first I/O Request can be read request,
It can be write request.Optionally, terminal here can be user equipment, can also be held for user orienteds such as main frames, for user
The equipment of row operation.The embodiment of the present invention is not limited to the form of terminal, as long as it can be communicated i.e. with communication equipment
Can.First RAID group hard discs can include multiple different hard disks, and communication equipment can inquire about the after the first I/O Request is received
One RAID group hard disc information, determines which hard disk can write data, then by the data carried in the first I/O Request write-in institute really
In fixed hard disk, or, determine which hard disk is the terminal data to be read be located on according to the first I/O Request, then reading should
The data of hard disk.
S102:When communication equipment judges any hard disk failure in the first RAID group hard discs or the mistake in preset time
During effect, the communication equipment changes the first RAID group hard disc information, obtains the 2nd RAID group hard discs, and do not start the first behaviour
Make;Wherein, first operation is that reconstructed operation or pre-copy are operated, and the 2nd RAID group hard discs include the first RAID
Group hard disc and HotSpare disk.
Specifically, when communication equipment judges any hard disk failure in the first RAID group hard discs or in preset time
During failure, communication equipment selects a new hard disk as HotSpare disk within the storage system, and by the HotSpare disk with breaking down
Hard disk or associated with the hard disk that can be failed in preset time, i.e., " association " herein is that is, HotSpare disk is added
Enter the first RAID group hard discs, so as to have modified the first RAID group hard disc information, obtain the 2nd RAID group hard discs, and for second
RAID group hard discs do not start the first operation.2nd RAID group hard discs can include all hard in above-mentioned first RAID group hard discs
Disk and above-mentioned HotSpare disk.
In order to better illustrate the technical scheme of the embodiment of the present invention, cite a plain example herein, following first
It is write request that I/O Request, which is assumed,:As shown in Fig. 2 the first RAID group hard discs include tetra- hard disks of A, B, C, D, terminal constantly to
Communication equipment sends the first I/O Request (being numbered respectively with 1#, 2#, 3# etc.).Assuming that terminal sends 1# first to communication equipment
I/O Request, to ask to write the data " abcd " carried in the I/O Requests of 1# first in A and B hard disks (RAID groups when writing data,
Deblocking can typically be write at least two hard disks), communication equipment is when receiving first I/O Requests of 1#, by the first RAID
The reference count of group hard disc is designated as 1, and inquires about the first RAID group hard disc information, determines that 1#, 2# hard disk can be write, then by " ab "
A hard disks are write, " cd " is write into B hard disks, and (how communication equipment is write data " abcd " on A and B hard disks, by communication equipment certainly
The algorithm of body determines, herein simply simple example);When communication equipment handles the first I/O Requests of 1#, communication equipment
The I/O Requests of 2# first are have received, communication equipment now Jia 1 in the original reference count of the first RAID group hard discs, communicated simultaneously
Equipment also handles the I/O Requests of 2# first according to the processing procedure to the I/O Requests of 1# first, the like.Assuming that communication equipment is received
The I/O Requests of 9# first, its ask by data " efg " write C and D hard disks (the now reference count of the first RAID group hard discs is 9, and
8 the first I/O Requests that communication equipment is just being received before treatment), communication equipment inquiry hard disk D failures, then communication equipment will
" ef " writes C hard disks, and g does not write D hard disks;Meanwhile, while data are write disk by communication equipment, communication equipment is also repaiied
Change the first RAID group hard disc information, HotSpare disk (hard disk E) is added into the first RAID group hard discs, form the 2nd RAID group hard discs, but
It is that communication equipment (does not start weight to the operation of the 2nd RAID group hard discs startup first in the example shown in Fig. 2
Structure), i.e., the data on hard disk D are not write on HotSpare disk.
S103:Communication equipment receives the second I/O Request that the terminal is issued, and accesses institute according to second I/O Request
The 2nd RAID group hard discs are stated, until all first I/O Requests of first reference count correspondence number of times return to result
Afterwards, communication device initiated first operation;Wherein, second I/O Request is communication equipment modification described first
The I/O Request received after RAID group hard disc information.
Specifically, with continued reference to the example shown in Fig. 2, during above-mentioned communication equipment changes the first RAID group hard discs
(not forming the 2nd RAID group hard discs also), if communication equipment receives other first I/O Requests and (is assumed to be the IO of 10# the first
Request), then communication equipment still handles the I/O Requests of 10# first according to the processing procedure before to the first I/O Request, that is, remains unchanged
Data are write into the first RAID group hard discs.But, after communication equipment obtains the 2nd RAID group hard discs, communication equipment is received
To I/O Request be the second I/O Request, communication equipment writes data into the 2nd RAID group hard discs according to second I/O Request, by
Faulty disk (hard disk D) is associated with HotSpare disk (hard disk E) in communication equipment, therefore can will write the data of past faulty disk
Write in HotSpare disk;In addition, 10 foregoing the first I/O Requests are remained and write data into the first RAID group hard discs.Namely
Say, either communication equipment communicates when changing the first RAID group hard disc information, or after the first RAID group hard disc information of modification
The I/O Request that equipment is sent to terminal can be handled, and can't interrupt the business that above-mentioned terminal is issued, it is ensured that terminal industry
The continuity of business.
When communication equipment according to the first reference count knows that foregoing 10 the first I/O Requests are processed into that is, this 10
First I/O Request can return to result, and communication equipment need to only check that result is returned several times, then be quoted with first
Number of times in counting is contrasted, if the two is equal, and 10 the first I/O Requests before communication equipment is determined have been handled
Into.Afterwards, communication device initiated first operation (starting the reconstruct of the 2nd RAID group hard discs in the example), be specially:Communication is set
It is standby by reading hard disk A, hard disk B, the data in disk C, by redundant computation, the data on hard disk D are obtained, so that this be counted
Obtained data are write on HotSpare disk.To sum up, the data on HotSpare disk not only include the sheet after the first RAID group hard discs of modification
This is write toward before the data in the second I/O Request of faulty disk, in addition to the first RAID group hard disc information of modification and modification first
Should write toward the data in the first I/O Request of faulty disk, the i.e. scheme of the embodiment of the present invention not during RAID group hard discs
The loss of data can be caused.
In the prior art, with continued reference to the example shown in Fig. 2, communication equipment detects hard disk D and broken down, therefore selection
One HotSpare disk, and the HotSpare disk is added the first RAID group hard discs to change RAID group hard disc information;Communication equipment is opened afterwards
Starting weight structure hard disk D data;But during the first RAID group hard disc information is changed, terminal issues IO to communication equipment please
Ask, thus communication equipment now inquire about should be reconstruct before the first RAID group hard discs information (i.e. HotSpare disk now also not add
Enter the first RAID group hard discs), therefore communication equipment will not write the data carried on above-mentioned I/O Request on HotSpare disk (certainly
It will not write on faulty disk, the partial data in the data that communication equipment can only carry the I/O Request is write other normal hard
On disk).Therefore, because reconstruct and write the process of disk and carry out simultaneously, on faulty disk before data reconstructed crossed (i.e.
Write on HotSpare disk), therefore cause the data of above-mentioned I/O Request not to be reconfigured to again on HotSpare disk.When finally from the first RAID
When deleting hard disk D in group hard disc, the data on HotSpare disk only include the data before hard disk D, cause to write on HotSpare disk
The loss of data that carries of above-mentioned I/O Request, there is the problem of slitting is inconsistent, and in order to avoid the problem, prior art is used
During the first RAID group hard disc information is changed, for new I/O Request, pause processing, until the first RAID groups is hard
After the completion of the modification of disk information completes and handles the request having been received by before, the IO of the above-mentioned pause of ability start to process please
The problem of asking, therefore cause the service disconnection of terminal, and if the IO received before the first RAID group hard disc information is changed
Request is more, then the time of the business pause of terminal is longer.
But, in this application, communication equipment is changing the first RAID group hard discs information and there is no second
During RAID group hard discs, if receiving new first I/O Request, it will continue according to new first I/O Request to the
One RAID group hard discs are handled;If communication equipment has obtained the 2nd RAID group hard discs, the first operation is not started, if now
When receiving I/O Request (I/O Request is the second I/O Request), then the 2nd RAID group hard discs are carried out according to second I/O Request
Handle (data write toward faulty disk now can be just write direct on HotSpare disk), until the first I/O Request that communication equipment is received
The number of times of the result of return is equal with the reference count of the first RAID group hard discs, and communication equipment just starts the first operation, will
The data of faulty disk are by reconstructing in write-in HotSpare disk.That is, the data on HotSpare disk not only include the first RAID of modification
Should write toward before the data in the second I/O Request of faulty disk, in addition to the first RAID group hard disc information of modification after group hard disc
With should write toward the data in the first I/O Request of faulty disk, the i.e. embodiment of the present invention during the first RAID group hard discs of modification
Scheme can't cause the loss of data, on the premise of ensureing that data do not lose, it is ensured that the continuity of business.
The amending method of RAID group hard disc information provided in an embodiment of the present invention, communication equipment is according to end
The first request issued is held to access the first RAID group hard discs for carrying the first reference count;When the communication equipment judge this first
Any hard disk failure in RAID group hard discs or when being failed in preset time, communication equipment changes the first RAID group hard discs letter
Breath, obtains the 2nd RAID group hard discs, and do not start the first operation;After communication equipment changes above-mentioned first RAID group hard disc information
The second I/O Request that terminal is issued is received, the 2nd RAID group hard discs are accessed according to second I/O Request, until above-mentioned first
All first I/O Requests of the first reference count correspondence number of times of RAID group hard discs are returned after result, and communication device initiated the
One operation.Method provided in an embodiment of the present invention, either communication equipment change the first RAID group hard disc information when, still repair
Change after the first RAID group hard disc information, the I/O Request that communication equipment is sent to terminal can be handled, and can't be interrupted above-mentioned
The business that terminal is issued, the continuity of terminal traffic is ensure that in the case where data are not lost.
The flow of the amending method embodiment two for the RAID group hard disc information that Fig. 3 provides for the present invention is shown
It is intended to.On the basis of above-described embodiment, the present embodiment is referred to when the I/O Request that communication equipment is received is write request,
Communication equipment ensures the continuous method of terminal traffic while the first RAID group hard disc information is changed.As shown in figure 3, this method
Including:
S201:The first I/O Request that communication equipment is issued according to terminal accesses the first RAID group hard discs;Wherein, described first
RAID group hard discs carry the first reference count, and first reference count is used to identify the first IO that the communication equipment is received
The number of times of request.
Specifically, S201 implementation procedure may refer to above-mentioned S101, it will not be repeated here.It should be noted that this reality
It is write request to apply the first I/O Request in example, and the first operation is reconstructed operation.
S202:When the communication equipment judges any hard disk failure in the first RAID group hard discs, the communication
The HotSpare disk is associated by equipment with the hard disk, obtains the 2nd RAID group hard discs.
S203:The first data carried in first I/O Request are write the first RAID group hard discs by communication equipment.
Specifically, S202 and S203 can be carried out simultaneously.I.e. when communication equipment is judged in the first RAID group hard discs
Any hard disk failure, the HotSpare disk selected within the storage system is associated with faulty disk, obtains the 2nd RAID group hard discs, and do not open
Dynamic reconstructed operation.But, (now there is no the 2nd RAID group hard discs) during the 2nd RAID group hard discs are obtained,
Communication equipment is still constantly receiving the first I/O Request, and the first data carried in the first I/O Request are write into the first RAID
In group hard disc.Exemplified by shown in above-mentioned Fig. 2, it is assumed that communication equipment is received while the first RAID group hard disc information is changed
Some first I/O Request, it asks to write first data into disk C and hard disk D (the now reference meter of the first RAID group hard discs
Several values is the number of times of the first I/O Request that the first RAID group hard discs are received, and communication equipment is also just being received before treatment
The first I/O Request), communication equipment inquiry hard disk D failures, then communication equipment by the data of part first write C hard disks;Together
When, while data are write disk by communication equipment, communication equipment also changes the first RAID group hard disc information, and HotSpare disk is added into the
One RAID group hard discs, form the 2nd RAID group hard discs, but communication equipment does not start reconstruct behaviour to the 2nd RAID group hard discs
Make, i.e., do not write the data on hard disk D on HotSpare disk.
That is, above-mentioned communication equipment is during the first RAID group informations are changed, the first I/O Request received
In the first data still write the first RAID group hard discs.
S204:Communication equipment receives the second I/O Request that the terminal is issued, and will be carried in second I/O Request
Second data write the 2nd RAID group hard discs, and second I/O Request is write request.
S205:After all first I/O Requests of first reference count correspondence number of times return to result, institute
Communication equipment is stated by the data reconstruction of the hard disk to the HotSpare disk.
Specifically, referring to above-mentioned Fig. 2, the I/O Request received after the 2nd RAID group hard discs is obtained in above-mentioned communication equipment
For the second I/O Request, communication equipment is according to second I/O Request, and the second data that the second I/O Request is carried write the 2nd RAID
Group hard disc, because communication equipment associates faulty disk (hard disk D) with HotSpare disk (hard disk E), therefore can be directly by this
It should write into the data of the part second write-in HotSpare disk of faulty disk;In addition, not obtaining the 2nd RAID for communication equipment
The first I/O Request received before group, communication equipment is remained the first data carried in these the first I/O Requests write-in the
In one RAID group hard discs.That is, either communication equipment change the first RAID group hard disc information when, or modification first
After RAID group hard disc information, the I/O Request that communication equipment is sent to terminal can be handled, and can't be interrupted under above-mentioned terminal
The business of hair, it is ensured that the continuity of terminal traffic.
Further, communication equipment can know that communication equipment is not obtaining the 2nd RAID according to above-mentioned first reference count
The first I/O Request received before group is processed into that is, this I/O Request of part first can return to result, communication
Equipment need to only check that result is returned several times, is then contrasted with the number of times in the first reference count, if the two is equal,
These first I/O Requests before then communication equipment is determined have handled completion.Afterwards, communication device initiated 2nd RAID groups are hard
Disk reconstructed operation, i.e.,:Communication equipment passes through redundancy meter by reading the data of part first on the hard disk of other normal works
Calculate, the data of another part first being stored on hard disk D are obtained, so that the data that the calculating is obtained are write on HotSpare disk.It is comprehensive
On, the data on HotSpare disk not only include should write into the second I/O Request of faulty disk after the first RAID group hard discs of modification
The data of part second, in addition to modification the first RAID group hard disc information before and modification the first RAID group hard discs during sheet
This writes toward the data of part first in the first I/O Request of faulty disk English, i.e. the scheme of the embodiment of the present invention can't cause data
Loss.
S206:The hard disk is replaced with the HotSpare disk by communication equipment, and by the hard disk from the 2nd RAID group hard discs
It is middle to remove, obtain the 3rd RAID group hard discs.
Specifically, when communication equipment by redundant computation by the data recovery of faulty disk to HotSpare disk after, communication equipment
Faulty disk is replaced with into HotSpare disk, and faulty disk is removed from the 2nd RAID group hard discs, the 3rd RAID group hard discs are obtained, this
Three RAID group hard discs include the hard disk and HotSpare disk of normal work in the 2nd RAID group hard discs.
S207:Communication equipment receives the 3rd I/O Request, and the 3rd data carried in the 3rd I/O Request are write into institute
State the 3rd RAID group hard discs;Wherein, the 3rd request is the communication equipment after the 3rd RAID group hard discs are obtained
The I/O Request received.
Specifically, the request that receives is the 3rd I/O Request after communication equipment obtains the 3rd RAID group hard discs, this
Three I/O Requests are write request.For the 3rd I/O Request, communication equipment is after the 3rd I/O Request is received, by the 3rd IO
Data in request are write in the 3rd RAID group hard discs on corresponding hard disk.
The amending method of RAID group hard disc information provided in an embodiment of the present invention, communication equipment is according to end
The first request issued is held to access the first RAID group hard discs for carrying the first reference count;When communication equipment judges the first RAID
During any hard disk failure in group hard disc, communication equipment changes the first RAID group hard disc information, obtains the 2nd RAID group hard discs, and
Do not start reconstructed operation;The 2nd IO that terminal issues is received after communication equipment changes above-mentioned first RAID group hard disc information please
Ask, the 2nd RAID group hard discs are accessed according to second I/O Request, until the first reference count pair of above-mentioned first RAID group hard discs
All first I/O Requests of number of times are answered to return after result, communication device initiated reconstructed operation.It is provided in an embodiment of the present invention
Method, either communication equipment is when changing the first RAID group hard disc information, or changes after the first RAID group hard disc information, leads to
The I/O Request that letter equipment is sent to terminal can be handled, and can't be interrupted the business that above-mentioned terminal is issued, not lost in data
The continuity of terminal traffic is ensure that in the case of mistake.
The flow of the amending method embodiment three for the RAID group hard disc information that Fig. 4 provides for the present invention is shown
It is intended to.On the basis of above-mentioned embodiment illustrated in fig. 1, the I/O Request that the present embodiment refers to receive when communication equipment is reading
During request, communication equipment ensures the continuous method of terminal traffic while the first RAID group hard disc information is changed.
As shown in figure 4, this method includes:
S301:The first I/O Request that communication equipment is issued according to terminal accesses the first RAID group hard discs;Wherein, described first
RAID group hard discs carry the first reference count, and first reference count is used to identify the first IO that the communication equipment is received
The number of times of request.
Specifically, S301 implementation procedure may refer to above-mentioned S101, it will not be repeated here.It should be noted that this reality
It is read request to apply the first I/O Request in example, and the first operation is pre-copy operation.
S302:When the communication equipment judges that any hard disk in the first RAID group hard discs fails in preset time
When, the HotSpare disk is associated by communication equipment with the hard disk, obtains the 2nd RAID group hard discs, and do not start described
Pre-copy is operated.
S303:Communication equipment reads the data in the first RAID group hard discs according to first I/O Request.
Specifically, S302 and S303 can be carried out simultaneously.I.e. when communication equipment is judged in the first RAID group hard discs
When any hard disk fails in preset time (when that is will fail, the following referred to as failure discs of the hard disk), it will deposit
The HotSpare disk selected in storage system is associated with failure disc, obtains the 2nd RAID group hard discs, and do not start pre-copy operation.But,
(now there is no the 2nd RAID group hard discs) during the 2nd RAID group hard discs are obtained, communication equipment is not still
The disconnected I/O Request of reception first, it is assumed that these first I/O Requests ask to read the data of failure disc.
In the prior art, can be from storage when some hard disk during communication equipment judges RAID group hard discs will fail
A HotSpare disk is selected in system, and by the data copy of the failure disc to HotSpare disk, in this copy procedure, communication equipment
The I/O Request for carrying out self terminal can be still received, to ask to read the data on failure disc.Assuming that communication equipment is asked according to the IO
During seeking the data on reading failure disc, the data complete copy on some moment, failure disc communicates to HotSpare disk
Equipment can remove the failure disc from RAID group hard discs, but now above-mentioned I/O Request is reading the failure disc data, therefore lead
Above-mentioned I/O Request is caused can not correctly to read the data on failure disc.The problem of in order to avoid there is digital independent error, existing skill
Art is used during modification RAID group hard discs information and pre-copy completion, at the I/O Request first pause issued for terminal
Reason, wait completes pre-copy, and failure disc is replaced with after HotSpare disk, is further continued for the I/O Request suspended before processing, therefore make
Into the interruption of terminal traffic.
But, in embodiments of the present invention, exemplified by shown in above-mentioned Fig. 2, it is assumed that communication equipment is changing the first RAID groups
During hard disk information, some first I/O Request is received, it asks communication equipment to read mistake in the first RAID group hard discs
(now the value of the reference count of the first RAID group hard discs is that the first IO that the first RAID group hard discs are received is asked to data on effect disk
The number of times asked, and the first I/O Request for also just receiving before treatment of communication equipment), then communication equipment can correctly read mistake
Imitate the data (simply changing the first RAID group hard disc information, copy is not started to also) on disk;When communication equipment obtains second
After RAID group hard discs, communication equipment does not start pre-copy operation, i.e., not the data copy on failure disc to HotSpare disk, then right
In communication equipment obtain the 2nd RAID group hard discs before (i.e. change the first RAID group hard disc information before and modification the first RAID
During group hard disc information) for all first I/O Requests for receiving, do not copied in advance because communication equipment now starts
Shellfish operate, therefore communication equipment can according to these first I/O Requests correctly read the first RAID group hard discs data (assuming that
These first I/O Requests ask to read the data on failure disc, then communication equipment can be correct according to these first I/O Requests
Read the data on failure disc).
S304:Communication equipment receives the second I/O Request that the terminal is issued, and reads institute according to second I/O Request
State the data in the 2nd RAID group hard discs.
S305:After all first I/O Requests of first reference count correspondence number of times return to result, institute
Communication equipment is stated by the data copy of the hard disk to the HotSpare disk.
Specifically, the I/O Request received after communication equipment obtains the 2nd RAID group hard discs is the second I/O Request, communication
Equipment reads the data of the 2nd RAID group hard discs according to second I/O Request after the second I/O Request is received.Now second
RAID group hard discs include the first RAID group hard discs and HotSpare disk.If second I/O Request is directed to the data of failure disc, by
Do not start pre-copy operation now in communication equipment, therefore communication equipment still can correctly read the number of failure disc
According to.
Further, communication equipment can know that communication equipment is not obtaining the 2nd RAID according to above-mentioned first reference count
The first I/O Request received before group is processed into that is, this I/O Request of part first can return to result, communication
Equipment need to only check that result is returned several times, is then contrasted with the number of times in the first reference count, if the two is equal,
These first I/O Requests before then communication equipment is determined have handled completion.Afterwards, communication equipment just starts the 2nd RAID groups
Hard disk pre-copy is operated, i.e., by the data copy of failure disc into HotSpare disk.
That is, either communication equipment change the first RAID group hard disc information when, or modification the first RAID groups
After hard disk information, the I/O Request that communication equipment is sent to terminal can be handled, and can't interrupt the industry that above-mentioned terminal is issued
Business, it is ensured that the continuity of terminal traffic, and read failure disc data when, due to and not actuated pre-copy operate, therefore protect
It is correct to have demonstrate,proved the failure disc data that the first I/O Request received during pre-copy read.
S306:The hard disk is replaced with the HotSpare disk by communication equipment, and by the hard disk from the 2nd RAID group hard discs
It is middle to remove, obtain the 4th RAID group hard discs.
Specifically, after communication equipment is by the data copy of failure disc to HotSpare disk, communication equipment replaces failure disc
For HotSpare disk, and failure disc is removed from the 2nd RAID group hard discs, obtain the 4th RAID group hard discs, the 4th RAID group hard discs
Include the hard disk and HotSpare disk of normal work in the 2nd RAID group hard discs.
S307:Communication equipment receives the 4th I/O Request, and reads the 4th RAID groups firmly according to the 4th I/O Request
Data in disk;Wherein, the 4th I/O Request is that the communication equipment is received after the 4th RAID group hard discs are obtained
The I/O Request arrived.
Specifically, the request that receives is the 4th I/O Request after communication equipment obtains the 4th RAID group hard discs, this
Four I/O Requests are read request.For the 4th I/O Request, communication equipment after the 4th I/O Request is received, according to this
Four I/O Requests directly read the data on the 4th RAID group hard discs.
The amending method of RAID group hard disc information provided in an embodiment of the present invention, communication equipment is according to end
The first request issued is held to access the first RAID group hard discs for carrying the first reference count;When the communication equipment judge this first
When any hard disk in RAID group hard discs fails in preset time, communication equipment changes the first RAID group hard disc information, obtains
2nd RAID group hard discs, and do not start pre-copy operation;Received after communication equipment changes above-mentioned first RAID group hard disc information
The second I/O Request issued to terminal, the 2nd RAID group hard discs are accessed according to second I/O Request, until above-mentioned first RAID groups
All first I/O Requests of the first reference count correspondence number of times of hard disk are returned after result, communication device initiated reconstruct behaviour
Make.Method provided in an embodiment of the present invention, either communication equipment is when changing the first RAID group hard disc information, or modification the
After one RAID group hard disc information, the I/O Request that communication equipment is sent to terminal can be handled, and can't interrupt above-mentioned terminal
The business issued, it is ensured that the continuity of terminal traffic.
One of ordinary skill in the art will appreciate that:Realizing all or part of step of above-mentioned each method embodiment can lead to
The related hardware of programmed instruction is crossed to complete.Foregoing program can be stored in a computer read/write memory medium.The journey
Sequence upon execution, performs the step of including above-mentioned each method embodiment;And foregoing storage medium includes:ROM, RAM, magnetic disc or
Person's CD etc. is various can be with the medium of store program codes.
The structure of the modification device embodiment one for the RAID group hard disc information that Fig. 5 provides for the present invention is shown
It is intended to.The device can with it is integrated in a communications device, or communication equipment.As shown in figure 5, the device includes:First visits
Ask module 10, the access modules 12 of processing module 11 and second.Wherein, the first access modules 10, for issued according to terminal
One input and output I/O Request accesses the first RAID RAID group hard discs;Wherein, the first RAID group hard discs are taken
The reference count of band first, first reference count is used for the number of times for identifying the first I/O Request that the communication equipment is received;
Processing module 11, for judging any hard disk failure in the first RAID group hard discs when the communication equipment or when default
During interior failure, the first RAID group hard disc information is changed, the 2nd RAID group hard discs are obtained, and do not start the first operation;Its
In, first operation is that reconstructed operation or pre-copy are operated, and it is hard that the 2nd RAID group hard discs include the first RAID groups
Disk and HotSpare disk;Second access modules 12, for receiving the second I/O Request that the terminal is issued, and please according to the 2nd IO
Access the 2nd RAID group hard discs are sought, until all first I/O Requests of first reference count correspondence number of times are returned
After result, start first operation;Wherein, second I/O Request is that the processing module changes the first RAID
The I/O Request received after group hard disc information.
The modification device of RAID group hard disc information provided in an embodiment of the present invention, can perform above-mentioned side
Method embodiment, its implementing principle and technical effect are similar, will not be repeated here.
The structure of the modification device embodiment two for the RAID group hard disc information that Fig. 6 provides for the present invention is shown
It is intended to.On the basis of above-mentioned embodiment illustrated in fig. 5, if first I/O Request and second I/O Request are write request,
First operation is reconstructed operation;Then further, above-mentioned processing module 11, including:First acquisition unit 111, for working as
When the communication equipment judges any hard disk failure in the first RAID group hard discs, the HotSpare disk is entered with the hard disk
Row association, obtains the 2nd RAID group hard discs, and do not start the reconstructed operation;First writing unit 112, for by described in
The first data carried in first I/O Request write the first RAID group hard discs.
Above-mentioned second access modules 12, including:Second writing unit 121, for receiving the 2nd IO that the terminal is issued
Request, and the second data carried in second I/O Request are write into the 2nd RAID group hard discs;Reconfiguration unit 122, is used
In after all first I/O Requests of first reference count correspondence number of times return to result, by the number of the hard disk
According to reconstruct to the HotSpare disk;First replaces and removes unit 123, for the hard disk to be replaced with into the HotSpare disk, and by institute
State hard disk to remove from the 2nd RAID group hard discs, obtain the 3rd RAID group hard discs;3rd writing unit 124, for receiving the 3rd
I/O Request, and the 3rd data carried in the 3rd I/O Request are write into the 3rd RAID group hard discs;Wherein, the described 3rd
Ask to remove the I/O Request that unit is received after the 3rd RAID group hard discs are obtained for described replace.
The modification device of RAID group hard disc information provided in an embodiment of the present invention, can perform above-mentioned side
Method embodiment, its implementing principle and technical effect are similar, will not be repeated here.
The structure of the modification device embodiment three for the RAID group hard disc information that Fig. 7 provides for the present invention is shown
It is intended to.On the basis of above-mentioned embodiment illustrated in fig. 5, if first I/O Request and second I/O Request are read request,
First operation is pre-copy operation;Then above-mentioned processing module 11, including:Second acquisition unit 113, for when the communication
When equipment judges that any hard disk in the first RAID group hard discs fails in preset time, by the HotSpare disk with it is described hard
Disk is associated, and obtains the 2nd RAID group hard discs, and do not start the pre-copy operation;First reading unit 114, is used for
Data in the first RAID group hard discs are read according to first I/O Request.
Above-mentioned second access modules 12, including:Second reading unit 125, for receiving the 2nd IO that the terminal is issued
Request, and the data in the 2nd RAID group hard discs are read according to second I/O Request;Copy cell 126, for working as
All first I/O Requests for stating the first reference count correspondence number of times are returned after result, by the data copy of the hard disk
To the HotSpare disk;Second replaces and removes unit 127, for the hard disk to be replaced with into the HotSpare disk, and by the hard disk
Removed from the 2nd RAID group hard discs, obtain the 4th RAID group hard discs;3rd reading unit 128, for receiving the 4th I/O Request,
And the data in the 4th RAID group hard discs are read according to the 4th I/O Request;Wherein, the 4th I/O Request is the communication
The I/O Request that equipment is received after the 4th RAID group hard discs are obtained.
The modification device of RAID group hard disc information provided in an embodiment of the present invention, can perform above-mentioned side
Method embodiment, its implementing principle and technical effect are similar, will not be repeated here.
Finally it should be noted that:Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations;To the greatest extent
The present invention is described in detail with reference to foregoing embodiments for pipe, it will be understood by those within the art that:Its according to
The technical scheme described in foregoing embodiments can so be modified, or which part or all technical characteristic are entered
Row equivalent substitution;And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology
The scope of scheme.
Claims (10)
1. a kind of amending method of RAID group hard disc information, it is characterised in that including:
It is hard that the first input and output I/O Request that communication equipment is issued according to terminal accesses the first RAID RAID groups
Disk;Wherein, the first RAID group hard discs carry the first reference count, and first reference count is set for identifying the communication
The number of times of standby the first I/O Request received;
When the communication equipment judges any hard disk failure in the first RAID group hard discs or failed in preset time,
The communication equipment changes the first RAID group hard disc information, obtains the 2nd RAID group hard discs, and do not start the first operation;Its
In, first operation is that reconstructed operation or pre-copy are operated, and it is hard that the 2nd RAID group hard discs include the first RAID groups
Disk and HotSpare disk;
The communication equipment receives the second I/O Request that the terminal is issued, and accesses described second according to second I/O Request
RAID group hard discs, until all first I/O Requests of first reference count correspondence number of times are returned after result, institute
State communication device initiated first operation;Wherein, second I/O Request is that the communication equipment changes the first RAID
The I/O Request received after group hard disc information.
2. according to the method described in claim 1, it is characterised in that if first I/O Request and second I/O Request are to write
Request, then first operation is reconstructed operation;When the communication equipment judges any hard in the first RAID group hard discs
Disk failure or when being failed in preset time, the communication equipment changes the first RAID group hard disc information, obtains second
RAID group hard discs, and do not start the first operation, specifically include:
When the communication equipment judges any hard disk failure in the first RAID group hard discs, the communication equipment will be described
HotSpare disk is associated with the hard disk, obtains the 2nd RAID group hard discs, and do not start the reconstructed operation;
The first data carried in first I/O Request are write the first RAID group hard discs by the communication equipment.
3. method according to claim 2, it is characterised in that the communication equipment receives the 2nd IO that the terminal is issued
Request, and the 2nd RAID group hard discs are accessed according to second I/O Request, until first reference count correspondence number of times
All first I/O Requests return after results, communication device initiated first operation is specifically included:
The communication equipment receives the second I/O Request that the terminal is issued, and carried in second I/O Request second is counted
According to write-in the 2nd RAID group hard discs;
After all first I/O Requests of first reference count correspondence number of times return to result, the communication equipment
By the data reconstruction of the hard disk to the HotSpare disk;
The hard disk is replaced with the HotSpare disk by the communication equipment, and the hard disk is moved from the 2nd RAID group hard discs
Remove, obtain the 3rd RAID group hard discs;
The communication equipment receives the 3rd I/O Request, and the 3rd data carried in the 3rd I/O Request are write into the described 3rd
RAID group hard discs;Wherein, the 3rd I/O Request is that the communication equipment is received after the 3rd RAID group hard discs are obtained
The I/O Request arrived.
4. according to the method described in claim 1, it is characterised in that if first I/O Request and second I/O Request are reading
Request, then first operation is pre-copy operation;When the communication equipment judges any in the first RAID group hard discs
Hard disk failure or when being failed in preset time, the communication equipment changes the first RAID group hard disc information, obtains second
RAID group hard discs, and do not start the first operation, specifically include:
It is described logical when any hard disk that the communication equipment is judged in the first RAID group hard discs fails in preset time
The HotSpare disk is associated by letter equipment with the hard disk, acquisition the 2nd RAID group hard discs, and does not start described pre- copy
Shellfish operates;
The communication equipment reads the data in the first RAID group hard discs according to first I/O Request.
5. method according to claim 4, it is characterised in that the communication equipment receives the 2nd IO that the terminal is issued
Request, and the 2nd RAID group hard discs are accessed according to second I/O Request, until first reference count correspondence number of times
All first I/O Requests return after results, communication device initiated first operation is specifically included:
The communication equipment receives the second I/O Request that the terminal is issued, and reads described second according to second I/O Request
Data in RAID group hard discs;
After all first I/O Requests of first reference count correspondence number of times return to result, the communication equipment
By the data copy of the hard disk to the HotSpare disk;
The hard disk is replaced with the HotSpare disk by the communication equipment, and by the hard disk from the 2nd RAID group hard discs
Remove, obtain the 4th RAID group hard discs;
The communication equipment receives the 4th I/O Request, and is read according to the 4th I/O Request in the 4th RAID group hard discs
Data;Wherein, the 4th I/O Request is the IO that the communication equipment is received after the 4th RAID group hard discs are obtained
Request.
6. a kind of modification device of RAID group hard disc information, it is characterised in that including:
First access modules, the first input and output I/O Request for being issued according to terminal accesses the first RAID
RAID group hard discs;Wherein, the first RAID group hard discs carry the first reference count, and first reference count is logical for identifying
The number of times for the first I/O Request that letter equipment is received;
Processing module, for judging any hard disk failure in the first RAID group hard discs when the communication equipment or presetting
When being failed in the time, the first RAID group hard disc information is changed, the 2nd RAID group hard discs are obtained, and do not start the first operation;
Wherein, first operation is that reconstructed operation or pre-copy are operated, and the 2nd RAID group hard discs include the first RAID groups
Hard disk and HotSpare disk;
Second access modules, institute is accessed for receiving the second I/O Request that the terminal is issued, and according to second I/O Request
The 2nd RAID group hard discs are stated, until all first I/O Requests of first reference count correspondence number of times return to result
Afterwards, first operation is started;Wherein, second I/O Request is that the processing module changes the first RAID group hard discs letter
The I/O Request received after breath.
7. device according to claim 6, it is characterised in that if first I/O Request and second I/O Request are to write
Request, then first operation is reconstructed operation;The then processing module, including:
First acquisition unit, for judging when the communication equipment during any hard disk failure in the first RAID group hard discs,
The HotSpare disk is associated with the hard disk, the 2nd RAID group hard discs are obtained, and do not start the reconstructed operation;
First writing unit, the first data for will be carried in first I/O Request write the first RAID group hard discs.
8. device according to claim 7, it is characterised in that second access modules, including:
Second writing unit, for receiving the second I/O Request that the terminal is issued, and will be carried in second I/O Request
Second data write the 2nd RAID group hard discs;
Reconfiguration unit, for returning to result when all first I/O Requests of first reference count correspondence number of times
Afterwards, by the data reconstruction of the hard disk to the HotSpare disk;
First replaces and removes unit, for the hard disk to be replaced with into the HotSpare disk, and by the hard disk from the 2nd RAID groups
Removed in hard disk, obtain the 3rd RAID group hard discs;
3rd writing unit, institute is write for receiving the 3rd I/O Request, and by the 3rd data carried in the 3rd I/O Request
State the 3rd RAID group hard discs;Wherein, the 3rd I/O Request is that the described first replacement removes unit in acquisition the 3rd RAID
The I/O Request received after group hard disc.
9. device according to claim 6, it is characterised in that if first I/O Request and second I/O Request are reading
Request, then first operation is pre-copy operation;The processing module, including:
Second acquisition unit, for judging any hard disk in the first RAID group hard discs when default when the communication equipment
During interior failure, the HotSpare disk is associated with the hard disk, the 2nd RAID group hard discs are obtained, and do not start described
Pre-copy is operated;
First reading unit, for reading the data in the first RAID group hard discs according to first I/O Request.
10. device according to claim 9, it is characterised in that second access modules, including:
Second reading unit, institute is read for receiving the second I/O Request that the terminal is issued, and according to second I/O Request
State the data in the 2nd RAID group hard discs;
Copy cell, for returning to result when all first I/O Requests of first reference count correspondence number of times
Afterwards, by the data copy of the hard disk to the HotSpare disk;
Second replaces and removes unit, for the hard disk to be replaced with into the HotSpare disk, and by the hard disk from described second
Removed in RAID group hard discs, obtain the 4th RAID group hard discs;
3rd reading unit, reads in the 4th RAID group hard discs for receiving the 4th I/O Request, and according to the 4th I/O Request
Data;Wherein, the 4th I/O Request communication equipment is received after the 4th RAID group hard discs are obtained
I/O Request.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410527853.0A CN104407806B (en) | 2014-10-09 | 2014-10-09 | The amending method and device of RAID group hard disc information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410527853.0A CN104407806B (en) | 2014-10-09 | 2014-10-09 | The amending method and device of RAID group hard disc information |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104407806A CN104407806A (en) | 2015-03-11 |
CN104407806B true CN104407806B (en) | 2017-10-10 |
Family
ID=52645439
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410527853.0A Active CN104407806B (en) | 2014-10-09 | 2014-10-09 | The amending method and device of RAID group hard disc information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104407806B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106610788B (en) * | 2015-10-26 | 2019-09-03 | 华为技术有限公司 | Hard disk array control method and device |
CN106648468A (en) * | 2016-12-29 | 2017-05-10 | 郑州云海信息技术有限公司 | Control method and system for IO request during data migration process |
CN108459824B (en) * | 2017-12-19 | 2021-05-04 | 西安华为技术有限公司 | Data modification and rewriting method and device |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103345450A (en) * | 2013-06-26 | 2013-10-09 | 浙江宇视科技有限公司 | Device and method for dynamically regulating redundant RAID array reconstruction |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010009442A (en) * | 2008-06-30 | 2010-01-14 | Fujitsu Ltd | Disk array system, disk controller, and its reconstruction processing method |
US20140281689A1 (en) * | 2013-03-15 | 2014-09-18 | Teradata Corporation | Method and system for rebuilding data following a disk failure within a raid storage system |
-
2014
- 2014-10-09 CN CN201410527853.0A patent/CN104407806B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103345450A (en) * | 2013-06-26 | 2013-10-09 | 浙江宇视科技有限公司 | Device and method for dynamically regulating redundant RAID array reconstruction |
Also Published As
Publication number | Publication date |
---|---|
CN104407806A (en) | 2015-03-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106201338B (en) | Date storage method and device | |
US11099953B2 (en) | Automatic data healing using a storage controller | |
US9372743B1 (en) | System and method for storage management | |
CN110807064B (en) | Data recovery device in RAC distributed database cluster system | |
CN104583930B (en) | Method, controller and the data migration device of Data Migration | |
WO2011110542A1 (en) | Buffer disk in flashcopy cascade | |
CN104407814B (en) | Double method and apparatus for writing data | |
CN106850315B (en) | Automatic disaster recovery system | |
CN106155943B (en) | A kind of method and device of the power down protection of dual control storage equipment | |
US20150278050A1 (en) | Communication of conditions at a primary storage controller to a host | |
CN106776147A (en) | A kind of variance data backup method and variance data back-up device | |
CN106557143A (en) | For the apparatus and method of data storage device | |
US11226746B2 (en) | Automatic data healing by I/O | |
CN104407806B (en) | The amending method and device of RAID group hard disc information | |
WO2021088423A1 (en) | Memory management method and system for raid io, terminal and storage medium | |
CN103678517B (en) | Method for fixing logical errors of hard disk and controller | |
CN109196458A (en) | Storage system active volume calculation method and device | |
WO2021253855A1 (en) | Information recording method, apparatus, and device, and readable storage medium | |
CN104572374B (en) | Processing method, device and the storage device of storage | |
CN104484354B (en) | Ensure the Snapshot Method and storage device of data consistency | |
US9207883B2 (en) | Indication of a destructive write via a notification from a disk drive that emulates blocks of a first block size within blocks of a second block size | |
CN106708445A (en) | Link selection method and device | |
CN104170307A (en) | Failure switching method, device and system | |
CN116501259A (en) | Disk group dual-activity synchronization method and device, computer equipment and storage medium | |
CN111857560A (en) | Method, apparatus and computer program product for managing data |
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 |