CN106250048A - The method and device of management storage array - Google Patents
The method and device of management storage array Download PDFInfo
- Publication number
- CN106250048A CN106250048A CN201510307041.XA CN201510307041A CN106250048A CN 106250048 A CN106250048 A CN 106250048A CN 201510307041 A CN201510307041 A CN 201510307041A CN 106250048 A CN106250048 A CN 106250048A
- Authority
- CN
- China
- Prior art keywords
- controller
- storage array
- lun
- running status
- storage
- 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 embodiment of the present invention provides a kind of method and device managing storage array.The method includes: when the controller in the first storage array receives to the request of LUN write data, and the running status of the person that writes lock rights management of LUN is when being online, controller in first storage array passes through communication link, sends the application request writing locked authority to the person that writes lock rights management of LUN;When the controller in the first storage array obtain that the person that writes lock rights management provides write locked authority after, write described data to LUN.In embodiments of the present invention, when locally stored array to opposite end storage array application to LUN write locked authority after write data again, the concordance of data can be kept.
Description
Technical field
The present embodiments relate to communication technology, particularly relate to a kind of method and device managing storage array.
Background technology
Dual-active data center, is that the double WriteMode of employing is simultaneously written locally stored system data and far-end is deposited
Storage system, it is ensured that data in remote storage system with the real-time synchronization in locally stored system.Work as this locality
During the storage system failure, the business of locally stored system bearing can switch in remote storage system, from
RPO) and recovery time target and (Recovery Point Objective is called for short: to realize return contact target
(Recovery Time Objective, is called for short: RTO) be all a solution of 0.
(Active-Active is called for short: AA) data center's solution includes dual master control device mould dual-active
Formula.In AA pattern, locally stored system and remote storage system provide LUN (logic simultaneously
Unit number, is called for short: read-write LUN), when the locally stored system failure, and remote storage system
Can seamlessly carry the business in locally stored system.
In prior art, AA pattern regards locally stored system and remote storage system as an entirety,
Each storage system arranges one or more cluster.Cluster includes one or more engine, passes through between engine
Optical-fibre channel (fibre channel, be called for short: FC) or Internet protocol (Internet Protocol, letter
Claim: IP) Network Synchronization data.But in the prior art, locally stored system cannot know remote storage
The state of storage control in system, and when local LUN is by the controller lock of remote storage system
Time, if local controller write data, the concordance of data can be destroyed.
Summary of the invention
The embodiment of the present invention provides a kind of method and device managing storage array, when locally stored array exists
To opposite end storage array application to LUN write locked authority after write data again, the one of data can be kept
Cause property.
First aspect, the embodiment of the present invention provides a kind of method managing storage array, is applied to first and deposits
In the controller of storage array, described first storage array and the second storage array pass through communication link,
Controller storage in described first storage array has controller running state information and writes lock authority letter
Breath, wherein, described controller running state information includes the fortune of each controller in described second storage array
Row state, described in write that lock authority information includes LUN LUN in described first storage array write lock
Rights manager, described LUN writes lock rights management person is the controller in described second storage array,
Described method includes:
When the controller in described first storage array receives to the request of described LUN write data,
And when the running status of the person that writes lock rights management of described LUN is online, in described first storage array
Controller by described communication link, send to the person that writes lock rights management of described LUN and write locked authority
Application request;
What the person that writes lock rights management described in obtaining when the controller in described first storage array provided writes locking
After authority, write described data to described LUN.
In conjunction with first aspect, in the first possible implementation of first aspect, described first storage
Controller in array passes through described communication link, sends to the lock rights holder of described LUN and writes locking
After the application request of authority, described method also includes:
What the person that writes lock rights management described in obtaining when the controller in described first storage array provided writes locking
After authority, send to described second storage array in the mirror image LUN of described LUN, write described data
Instruction, described mirror image LUN is by the controller management in described second storage array.
In conjunction with the first possible implementation of first aspect, the reality that the second in first aspect is possible
In existing mode, described method also includes:
When described LUN has write, and receive the described mirror image LUN that described second storage array sends
After the response message write, described in the controller release in described first storage array, write locked authority.
In conjunction with first aspect, first aspect the first in the possible implementation of the second any one,
In the third possible implementation of first aspect, described method also includes:
Before writing locked authority described in controller release in described first storage array, described first deposits
Controller in storage array receives the another one request to described LUN write data, and described
When the running status of the person that writes lock rights management of LUN is online, by described communication link, to described
The person that writes lock rights management of LUN sends the application request writing locked authority;
Controller in described first storage array receive described in the person that writes lock rights management return refusal and write
The response message of locked authority.
In conjunction with first aspect, in the 4th kind of possible implementation of first aspect, described method is also wrapped
Include:
The master controller that master controller in described first storage array receives in described second storage array is sent out
The pending event sent, described pending event occurs in described second storage array, wherein, described
Master controller in first storage array is in the controller in described first storage array, described
Master controller in second storage array is in the controller in described second storage array;
Master controller in described first storage array, according to described pending event, updates described controller
The running status of the controller in described second storage array that running state information comprises;
Master controller in described first storage array sends the controller running state information after updating to the
Other controllers in one storage array, each other controller updates the described controller fortune of oneself storage
Row status information.
In conjunction with the 4th kind of possible implementation of first aspect, in the 5th kind of possible reality of first aspect
In existing mode, when described pending event is controller failure event, in described first storage array
Master controller according to described pending event, update that described controller running state information comprises described
The running status of the controller in two storage arrays, including:
The controller of described controller failure event will be there is in the master controller in described first storage array
Running status removes from described controller running state information, or
In described controller running state information, the master controller in described first storage array will occur
The running status of the controller of described controller failure event is updated to off-line or fault.
In conjunction with first aspect, in the 6th kind of possible implementation of first aspect, described method is also wrapped
Include:
The master controller that master controller in described first storage array receives in described second storage array is sent out
The recovery request sent, described recovery request carries the running status of each controller in described second storage array;
Master controller in described first storage array is by the operation of each controller in described second storage array
State is added to described controller running state information, wherein, each controller in described second storage array
Running status be online, or, the running status of each controller in described second storage array is updated to
Online;
Master controller in described first storage array sends the first response message to described second storage array
In master controller, described first response message carries the operation of each controller in described first storage array
State;
Wherein, the master controller in described first storage array is the controller in described first storage array
In one, the master controller in described second storage array is the controller in described second storage array
In one.
In conjunction with first aspect, in the 7th kind of possible implementation of first aspect, described method is also wrapped
Include:
When the master controller in described first storage array detects that communication link recovers event, described the
Master controller in one storage array generates recovery request, and described recovery request carries described first storage battle array
The running status of each controller in row;
Master controller in described first storage array sends described recovery request to described second storage array
In master controller;
The master controller that master controller in described first storage array receives in described second storage array is sent out
The second response message sent, described second response message carries each controller in described second storage array
Running status;
Master controller in described first storage array resolves described second response message, obtains described second
The running status of each controller in storage array;
Master controller in described first storage array is by the operation of each controller in described second storage array
State is added to described controller running state information, wherein, each controller in described second storage array
Running status be online, or, the running status of each controller in described second storage array is updated to
Online;
Wherein, the master controller in described first storage array is the controller in described first storage array
In one, the master controller in described second storage array is the controller in described second storage array
In one.
In conjunction with first aspect, first aspect the first in the 7th kind of possible implementation any one,
In the 8th kind of possible implementation of first aspect, described controller running status includes: online,
Any one in off-line and fault.
Second aspect, the embodiment of the present invention provides a kind of device managing storage array, is integrated in first and deposits
In the controller of storage array, described first storage array and the second storage array pass through communication link,
Controller storage in described first storage array has controller running state information and writes lock authority letter
Breath, wherein, described controller running state information includes the fortune of each controller in described second storage array
Row state, described in write that lock authority information includes LUN LUN in described first storage array write lock
Rights manager, described LUN writes lock rights management person is the controller in described second storage array,
Described device includes receiver module, sending module, detection module and processing module;
Described receiver module, for receiving the request to described LUN write data;
Described detection module, for detecting the running status of the person that writes lock rights management of described LUN;
Described sending module, for receiving described to described LUN write data when described receiver module
During request, and described detection module detects that the running status of the person that writes lock rights management of described LUN is
Time online, by described communication link, send to the person that writes lock rights management of described LUN and write locked authority
Application request;
Described receiver module, be additionally operable to that the person that writes lock rights management described in receiving provides write locked authority should
Answer message;
Described processing module, after what the person that writes lock rights management described in obtain provided writes locked authority, to
Described LUN writes described data.
In conjunction with second aspect, in the implementation that the second of second aspect is possible, described processing module
It is additionally operable to:
The person that writes lock rights management described in obtaining provide write locked authority after, trigger described sending module to
Described second storage array sends the instruction writing described data in the mirror image LUN of described LUN, institute
State mirror image LUN by the controller management in described second storage array.
In conjunction with the implementation that the second of second aspect is possible, in the third possible reality of second aspect
In existing mode, described receiver module, it is additionally operable to receive the described mirror image LUN that described second storage array sends
The response message write;
Described processing module, is additionally operable to when described LUN has write, and described receiver module receives institute
After stating the response message that mirror image LUN has write, described in release, write locked authority.
In conjunction with second aspect, second aspect the first in the possible implementation of the second any one,
In the third possible implementation of second aspect, described sending module, it is additionally operable in described process
Before writing locked authority described in module release, described receiver module receives another one and writes to described LUN
Enter the request of data, and described detection module detects the operation of the person that writes lock rights management of described LUN
When state is online, by described communication link, send to the person that writes lock rights management of described LUN and write lock
Determine the application request of authority;
Described receiver module, be additionally operable to receive described in the person that writes lock rights management return refusal and write locking power
The response message of limit.
In conjunction with second aspect, in the 4th kind of possible implementation of second aspect, integrated described device
Controller be the master controller in described first storage array, the main control in described first storage array
Device is in the controller in described first storage array;
Described receiver module, what the master controller being additionally operable to receive in described second storage array sent waits to locate
Director's part, described pending event occurs in described second storage array, wherein, described second storage
Master controller in array is in the controller in described second storage array;
Described processing module, is additionally operable to, according to described pending event, update described controller running status
The running status of the controller in described second storage array that information comprises;
Described sending module, is additionally operable to send the controller running state information after updating to the first storage battle array
Other controllers in row, each other controller updates the described controller running status letter of oneself storage
Breath.
In conjunction with the 4th kind of possible implementation of second aspect, in the 5th kind of possible reality of second aspect
In existing mode, when described pending event is controller failure event, described processing module performs basis
Described pending event, updates in described second storage array that described controller running state information comprises
The running status of controller time, particularly as follows:
The running status of the controller of described controller failure event will be occurred from described controller running status
Information removes, or
In described controller running state information, the controller of described controller failure event will be there is
Running status is updated to off-line or fault.
In conjunction with second aspect, in the 6th kind of possible implementation of second aspect, integrated described device
Controller be the master controller in described first storage array, the main control in described first storage array
Device is in the controller in described first storage array;
Described receiver module, is additionally operable to the recovery of the master controller transmission receiving in described second storage array
Request, described recovery request carries the running status of each controller in described second storage array;
Described processing module, is additionally operable to add the running status of each controller in described second storage array
To described controller running state information, wherein, the operation shape of each controller in described second storage array
State is online, or, the running status of each controller in described second storage array is updated to online;
Described sending module, is additionally operable to send the first response message to the master control in described second storage array
Device processed, described first response message carries the running status of each controller in described first storage array;
Wherein, the master controller in described second storage array is the controller in described second storage array
In one.
In conjunction with second aspect, in the 7th kind of possible implementation of second aspect, integrated described device
Controller be the master controller in described first storage array, the main control in described first storage array
Device is in the controller in described first storage array;
Described detection module, is additionally operable to detect communication link and recovers event;
Described processing module, is additionally operable to when described detection module detects that described communication link recovers event,
Generating recovery request, described recovery request carries the running status of each controller in described first storage array;
Described sending module, is additionally operable to send described recovery request to the master control in described second storage array
Device processed;
Described receiver module, is additionally operable to the second of the master controller transmission receiving in described second storage array
Response message, described second response message carries the running status of each controller in described second storage array;
Described processing module, is additionally operable to resolve described second response message, obtains described second storage array
In the running status of each controller;And, the running status of each controller in described second storage array is added
Add to described controller running state information, wherein, the operation of each controller in described second storage array
State is online, or, the running status of each controller in described second storage array is updated to online;
Wherein, the master controller in described second storage array is the controller in described second storage array
In one.
In conjunction with second aspect, second aspect the first in the 7th kind of possible implementation any one,
In the 8th kind of possible implementation of second aspect, described controller running status includes: online,
Any one in off-line and fault.
The method and device of embodiment of the present invention management storage array, by the control in the first storage array
Device storage processed has controller running state information and writes lock authority information, wherein, and controller running status
Information includes the running status of each controller in the second storage array, writes lock authority information and includes the first storage
The person that writes lock rights management of LUN in array, LUN writes lock rights management person is in the second storage array
Controller, the first storage array and the second storage array pass through communication link, it is achieved first storage
The array monitoring to the running status of the controller in the second storage array, say, that the first storage battle array
Row can directly know the existence of the second storage array and the generation of fault;It addition, at the first storage array
In controller receive to LUN write data request time, and LUN write lock rights management person's
When running status is online, the controller in the first storage array sends to the person that writes lock rights management of LUN
Write the application request of locked authority, thus obtain that the person that writes lock rights management provides write locked authority, to LUN
Write data, can keep the concordance of data.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, below will be to embodiment or existing skill
Art in describing the required accompanying drawing used do one and introduce simply, it should be apparent that, attached in describing below
Figure is some embodiments of the present invention, for those of ordinary skill in the art, is not paying creativeness
On the premise of laborious, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is Application Scenarios-Example figure of the present invention;
Fig. 2 is the flow chart that the present invention manages the embodiment of the method one of storage array;
Fig. 3 is that the present invention manages the method for storage array to be applied to the work of application scenarios as shown in Figure 1 former
Reason exemplary plot;
Fig. 4 is the flow chart that the present invention manages the embodiment of the method two of storage array;
Fig. 5 is the exemplary plot that the present invention manages the embodiment of the method three of storage array;
Fig. 6 is the flow chart that the present invention manages the embodiment of the method four of storage array;
Fig. 7 is the exemplary plot that the present invention manages the embodiment of the method five of storage array;
Fig. 8 is the exemplary plot that the present invention manages the embodiment of the method six of storage array;
Fig. 9 is the flow chart that the present invention manages the embodiment of the method seven of storage array;
Figure 10 is the flow chart that the present invention manages the embodiment of the method eight of storage array;
Figure 11 is the structural representation that the present invention manages the device embodiment one of storage array;
Figure 12 is the structural representation that the present invention manages the device embodiment two of storage array.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with this
Accompanying drawing in bright embodiment, is clearly and completely described the technical scheme in the embodiment of the present invention,
Obviously, described embodiment is a part of embodiment of the present invention rather than whole embodiments.Based on
Embodiment in the present invention, those of ordinary skill in the art are obtained under not making creative work premise
The every other embodiment obtained, broadly falls into the scope of protection of the invention.
Although should be understood that and term first, second etc. can being used herein to describe various elements or event
Or message, but these elements or event or message should not be restricted by these terms and limit.These terms only in order to
Distinguish an element or event or message and another element or event or message.For example, without departing from
In the case of the application protection domain, the first storage array is referred to alternatively as the second storage array, and similarly,
Second storage array is referred to alternatively as the first storage array.Wherein, storage array includes controller and disk,
After controller receives user data, it is sent to disk and stores.
Wherein, RTO is the time span of enterprise's tolerable service disruption.Such as there is afternoon in disaster
The most just needing to recover, RTO value is exactly 12 hours.RPO refers to when, after service recovery, recovery gets
Point between time corresponding to data.If carrying out backing up once during current enterprise morning every day zero, when service is extensive
After Fu, store in system can be only data during that morning zero before nearest disaster occurs.More than according to
Two simple principles, enterprise not only can obtain optimal disaster recovery solution to existing data system,
Can also require to obtain optimal disaster recovery solution according to set RTO and RPO.
In embodiments of the present invention, using AA pattern dual-active data center solution as storage system
One characteristic.The embodiment of the present invention can realize in the case of need not add the equipment such as such as engine
RTO and the RPO target equal to 0.The embodiment of the present invention can be generalized to such as network attached storage
NAS), the dual-active of the storage system composition such as object storage (Network Attached Storage, is called for short:
In data center's solution.
In the embodiment of the present invention, there is controller running status by the controller storage in the first storage array
Information, wherein, this controller running state information includes the operation shape of each controller in the second storage array
State, invents one by the controller cluster in two composition dual-active data centers, independent storage arrays
Individual entirety, the controller in each storage array has a controller running state information.Controller is transported
Row status information can not only make fault in the existence of storage array perception opposite end and opposite end occur, it addition, the
Lock authority information is write in controller storage in one storage array, and this is write lock authority information and includes the first storage
The person that writes lock rights management of LUN in array, this LUN writes lock rights management person is the second storage array
In controller, thus provide unified management for distributed lock.
As it is shown in figure 1, two storage arrays, respectively storage array A and storage array B, wherein,
The first storage array that storage array A can be interpreted as in the embodiment of the present invention by those skilled in the art,
The second storage array that storage array B is in the embodiment of the present invention.And similarly, it is also possible to will storage
The first storage array that array B is interpreted as in the embodiment of the present invention, storage array A is the present invention to be implemented
The second storage array in example.
In following example, using storage array A as the first storage array, storage array B is as second
Illustrate as a example by storage array.
Storage array A includes controller A1, controller A2, controller A3 and controller A4.Wherein,
Controller A1 is the master controller in storage array A, the master controller in the i.e. first storage array, its
More than be in the first storage array from controller.Storage array B include controller B1, controller B2,
Controller B3 and controller B4.Wherein, controller B1 is the master controller in storage array B, i.e.
Master controller in second storage array, remaining be in the second storage array from controller.It should be noted that
It is that the number from controller from controller and the second storage array in the first storage array is not limited to
In as shown in Figure 13.
In the embodiment of the present invention, each storage array is safeguarded a controller running state information and writes lock
Authority information, say, that controller running state information and write lock authority information be stored in storage array
In controller in.Form storage array being stored to above-mentioned information is unrestricted, such as with view
Form stores, etc..When using the form storage of view, view includes the controller in storage array B
Mapping in storage array A, alternatively, also includes that the local of the controller in storage array A maps.
It is referred to as calling node here it is possible to mapped by the controller in storage array B.Therefore, the present invention implements
The view that in example, storage array A and storage array B each sees is different.
As it is shown in figure 1, the view of storage array A includes controller A1, controller A2, controller A3
Local with controller A4 maps A1, A2, A3 and A4, also include controller B1, controller B2,
Controller B3 and controller B4 maps and obtains calling node B1 ', B2 ', B3 ' and B4 '.Accordingly,
The view of storage array B includes controller B1, controller B2, controller B3 and the basis of controller B4
Ground Mapping B 1, B2, B3 and B4, also include by controller A1, controller A2, controller A3 and
Controller A4 maps another name node A1 ', A2 ', A3 ' and the A4 ' obtained.Storage array A and storage battle array
The communication between the controller in storage array is processed by switch 10 and switch 20 between row B,
Each switch serves a different attachment structure.
Fig. 2 is the flow chart that the present invention manages the embodiment of the method one of storage array.The embodiment of the present invention carries
For a kind of method managing storage array, it is applied in the controller of the first storage array, this first storage
Array and the second storage array pass through communication link.Controller storage in first storage array has control
Device running state information processed and write lock authority information, wherein, controller running state information includes second
The running status of each controller in storage array, writes lock authority information and includes LUN in the first storage array
The person that writes lock rights management, LUN writes lock rights management person is the controller in the second storage array.
The method includes:
S201, when the controller in the first storage array receive to LUN write data request time, and
And the running status of the person that writes lock rights management of LUN is when being online, the controller in the first storage array leads to
Cross communication link, send the application request writing locked authority to the person that writes lock rights management of LUN.
S202, write locking power when what the controller in the first storage array obtained that the person that writes lock rights management provides
After limit, write data to LUN.
Wherein, the applicant writing locked authority is controller, and specifically controller is data application,
These data are written to the authority of LUN.
Specifically, dual-active data center based on distributed lock solution main frame input and output
(Input/Output is called for short: IO) can be written and read two storage arrays simultaneously.When main frame is to
When the LUN of the controller in one storage array writes, write data in the second storage array simultaneously
In the LUN of controller, it is achieved that the real-time synchronization of the LUN of the controller in two storage arrays, its
In, LUN refers to the block memory element that storage array presents to main frame.Owing to main frame can be simultaneously to first
The LUN of the controller in the LUN of the controller in storage array and the second storage array is written and read,
Therefore, main frame must add the mutual exclusion lock between storage array when accessing LUN.Between storage array, mutual exclusion lock is individual point
Cloth is locked, the principle of distributed lock be the person that writes lock rights management can be any one in two storage arrays
Individual controller, and other controllers write lock authority to application all must be to the person that writes lock rights management application.
Controller in first storage array and the controller in the second storage array map and obtain view, with
Time view between storage array mutual exclusion lock provide across storage array controllers view.The embodiment of the present invention is by group
Two storage arrays becoming dual-active data center are managed as a cluster, simplify between storage array
The deployment of mutual exclusion lock, reduces the complexity of dual-active data center IO, improves IO performance.
Due to controller running status may include that online, any one in off-line and fault.Cause
This, the controller in the first storage array receives when the request of LUN write data, the first storage
Whether the running status of the controller detection person that writes lock rights management in array is online, if writing lock authority pipe
When the running status of reason person is online, the controller in the first storage array passes through communication link, to LUN
The person that writes lock rights management send the application request writing locked authority, and person carries to write lock rights management in acquisition
Confession write locked authority after, to LUN write data.
The method and device of embodiment of the present invention management storage array, by the control in the first storage array
Device storage processed has controller running state information and writes lock authority information, wherein, and controller running status
Information includes the running status of each controller in the second storage array, writes lock authority information and includes the first storage
The person that writes lock rights management of LUN in array, LUN writes lock rights management person is in the second storage array
Controller, the first storage array and the second storage array pass through communication link, it is achieved first storage
The array monitoring to the running status of the controller in the second storage array, say, that the first storage battle array
Row can directly know the existence of the second storage array and the generation of fault;It addition, at the first storage array
In controller receive to LUN write data request time, and LUN write lock rights management person's
When running status is online, the controller in the first storage array sends to the person that writes lock rights management of LUN
Write the application request of locked authority, thus obtain that the person that writes lock rights management provides write locked authority, to LUN
Write data, can keep the concordance of data.
On the basis of above-described embodiment, the controller in the first storage array passes through communication link, to LUN
Lock rights holder send the application request writing locked authority after, the method can also include: when the
After what controller in one storage array obtained that the person that writes lock rights management provides writes locked authority, deposit to second
Storage array sends the instruction writing data in the mirror image LUN of LUN, and this mirror image LUN is by the second storage
Controller management in array.Wherein, LUN and mirror image LUN can have identical LUN identification
(identification is called for short: ID).The data of LUN and mirror image LUN storage keep synchronizing.Therefore,
Data protection is formed between LUN and mirror image LUN, when the controller managing LUN sends fault, can
Obtaining data from its mirror image LUN is that, it is to avoid loss of data.LUN and mirror image LUN is positioned at not
In same storage array, each self-corresponding controller it is managed.
Storage array in the embodiment of the present invention has 2, and main frame and 2 storage arrays connect, and work as main frame
Send when the instruction of certain LUN write data, the LUN indicated by LUN ID carried in instruction
Being referred to as dual-active LUN, the set of the two storage array is properly termed as dual-active storage array.Dual-active LUN
It is virtual, dual-active LUN corresponding 2 physics LUN, the most aforesaid LUN and mirror image LUN,
The LUN ID of dual-active LUN and physics LUN three can be identical.Therefore, the two storage array in
Now give only one of which LUN (dual-active LUN) of user, but that actual storage data is 2 physics LUN
(LUN and mirror image LUN).When writing data in certain LUN, data are actually written 2
In LUN, be equivalent to data and backed up.Locally stored system can directly know remote storage system
Exist and the generation of fault, thus the seamless pipe of data write cannot be realized.When writing data, double
Live 2 physics LUN corresponding to LUN all in can the state of movable (data can be write), because of
This it be referred to as double LUN.
Running status includes: any one in online, off-line and fault.Online controller is permissible
Normal work, the controller in off-line and fault cannot normally work.
If to LUN write data, other controllers also write other data in LUN, or
Person writes other data in its mirror image LUN, and the data that all can cause LUN and mirror image LUN are inconsistent,
Namely destroy the data consistency between LUN and mirror image LUN.Therefore, the embodiment of the present invention makes
With writing locked authority and avoid the generation of this situation, write the authority that locked authority is a kind of mutual exclusion.Only Shen
Please to the right writing data in the controller writing locked authority this LUN oriented;Additionally, application is to writing
The controller of locked authority has the right writing identical data in mirror image LUN.Locking power is write in release
Before limit, other controllers in other arrays all do not have authority to write in LUN or mirror image LUN
Data.
Write locked authority and the data write are bound, say, that write locked authority and define and can only write Shen
Please write data specified during locked authority.Even if controller has writes locked authority, also cannot be write it
His data.In other words, the applicant writing locked authority is controller, be controller be described data Shen
Please, described data are written to the authority of described LUN.
It addition, the method can also include: when LUN has write, and receive the second storage array and send out
After the response message that the mirror image LUN sent has write, locking is write in the controller release in the first storage array
Authority.This embodiment ensures that writing locked authority is being obtained by different controllers the most in the same time.
On the basis of the above, the method can also include: the controller release in the first storage array is write
Before locked authority, the controller in the first storage array receives another one to LUN write data
Request, the and when running status of the person that writes lock rights management of LUN is online, by communication link, to
The person that writes lock rights management of LUN sends the application request writing locked authority;Control in first storage array
Device reception writes lock rights management, and person returns the response message refusing to write locked authority.It is to say, write
Locked authority can only be obtained by a controller at synchronization, does not discharge it there being controller to obtain it
Before, remaining controller waits.
Here by specifically illustrating above-described embodiment.As it is shown on figure 3, Fig. 3 illustrates as shown in Figure 1
The fundamental diagram of example.With reference to Fig. 3, user passes through controller A2 and the storage array of storage array A
The controller B2 of B accesses a dual-active LUN, this dual-active LUN and includes thing corresponding to controller A2
The reason LUN1 physics LUN2 corresponding with controller B2, and this dual-active LUN write lock rights management
Person is controller B1.What the controller at storage array B was seen write lock rights management, and person is controller B1,
And see on storage array A write lock rights management person be the controller B1 view at storage array A
In another name node B1 '.When user issues the request of write data from the controller B2 of storage array B,
Controller B2 writes locked authority, the most respectively to physics LUN2 and physics LUN1 to controller B1 application
Writing data, after writing, locked authority is write in controller B2 release.If applied for writing at controller B2
During locked authority writes locked authority to release, user issues additionally from the controller A2 of storage array A
One to physics LUN1 write data request, controller A2 to controller B1 storage array B's
Calling node B1 ' application in view and write locked authority, controller A2 judges representated by another name node B1 '
Controller physical node is the controller B1 of storage array B, and controller A2 will apply for that authority request sends
To the controller B1 of storage array B.Controller B1 find write locked authority held by controller B2 and
Not release, then return to controller A2 and apply for unsuccessfully, the request etc. of the write data on controller A2
Locked authority to be written.Controller B1 waits that controller B2 release is reallocated after writing locked authority and writes locking power
Limiting to controller A2, on controller A2, the request of write data continues executing with number after obtaining and writing locked authority
According to write operation.And read data and only read data in the storage array this locality issued, it is not related to across storage battle array
The mutual exclusion of row, it is not necessary to read lock determines authority.
The controller storing the controller in aforementioned first storage array is described below by way of specific embodiment
The management of running state information.
Fig. 4 is the flow chart that the present invention manages the embodiment of the method two of storage array.Implementing shown in Fig. 2
On the basis of example, as shown in Figure 4, the method can also include:
Master controller in S401, the first storage array receives the master controller in the second storage array and sends
Pending event.
Specifically, pending event occurs in the second storage array.Wherein, in the first storage array
Master controller is one in the controller in the first storage array, the master controller in the second storage array
For one in the controller in described second storage array.
Master controller in S402, the first storage array, according to pending event, updates controller and runs shape
The running status of the controller in the second storage array that state information comprises;
Master controller in S403, the first storage array send the controller running state information after updating to
Other controllers in first storage array, each other controller updates the controller of oneself storage and runs
Status information.
Ensured that by the embodiment of the present invention controller running status of the controller storage in the first storage battle array is believed
Cease consistent with the controller running state information of the controller of the second storage array kind storage.But institute of the present invention
Refer to is not unanimously strong consistency, and is to ensure that final consistency.Strong consistency is to ensure that each moment
It is all consistent.And the most do not require consistent constantly, but configuration or event handling flow process is waited to tie
It is consistent after bundle.
Alternatively, pending event can include controller failure event, newly-increased controller event, first
Between storage array and the second storage array in communication link fails event and communication link recovery event extremely
A few type.Pending event refers to cause number and/or the fortune of the controller in the first storage array
The event that row state changes, and/or, the number of the controller in the second storage array and/or operation shape
The event that state changes.
The controller running state information that controller in first storage array is stored by different pending events
Impact different, describe in detail below by way of several specific implementations.
In the first specific implementation, pending event is controller failure event.In this implementation,
S402 may include that the master controller in the first storage array is by the controller of generator controller event of failure
Running status remove from controller running state information;Or, in controller running state information,
The running status of the controller of generator controller event of failure is updated by the master controller in the first storage array
For off-line or fault.
In the first optional mode, the event that the controller in the first storage array is broken down, also
When being exactly local controller event of failure, the master controller in the first storage array is by generator controller fault
The running status of the controller of event removes from controller running state information;Or, run at controller
In status information, the master controller in the first storage array is by the controller of generator controller event of failure
Running status is updated to off-line or fault.Reference Fig. 5, the master controller in the first storage array, i.e.
When controller A1 detects controller A4 fault, mapping corresponding for controller A4 is removed from view,
Or, the running status of controller A4 is updated to off-line or fault.
In the optional mode of the second, if generator controller event of failure for the master control in the first storage array
Device processed, the first storage array determines the master controller in the first new storage array;The first new storage battle array
The running status of the controller of generator controller event of failure is run shape from controller by the master controller in row
State information removes;Or, in controller running state information, the new master control in the first storage array
The running status of the controller of generator controller event of failure is updated to off-line or fault by device processed.
The controller of generator controller event of failure is probably the controller in the second storage array, first deposits
Master controller in storage array or other controllers.If the controller plc in the second storage array,
The first specific implementation is used to update controller running state information;If in the first storage array
Master controller, uses the optional mode of the second to update controller running state information;If the first storage battle array
Other controllers in row, use the first optional mode to update controller running state information.
In the second specific implementation, pending event is newly-increased controller event.In this implementation,
S402 may include that the master controller in the first storage array is by controller newly-increased for newly-increased controller event
Add to controller running state information.
Wherein, the handling process of newly-increased controller event is similar with the handling process of controller failure event.
Newly-increased controller is joined in the way of another name node the controller running status letter of the first storage array
Breath.If newly-increased controller event is event in the first storage array, the master in the first storage array
Controller just this newly-increased controller joins controller running state information.
Fig. 6 and Fig. 7 illustrates the master controller in the first storage array and the master in the second storage array
The controller handling process to controller failure event.
As shown in Figure 6, the method may include that
Master controller in S601, the first storage array detects the controller failure in the first storage array
Event.
Master controller in S602, the first storage array runs according to controller failure event update controller
Status information.
Controller running state information after master controller in S603, the first storage array will update synchronizes
Other controllers in the first storage array.
Master controller in S604, the first storage array is by controller failure event forwarding to second storage battle array
Master controller in row.
Hereafter, the master controller in the first storage array terminates handling process.
Master controller in second storage array receives this controller failure event, updates the second storage battle array
The controller running state information of the controller storage in row;The control in the second storage array after updating
The controller running state information of device processed storage is synchronized on other controllers in the second storage array;Knot
Bundle handling process.
With reference to Fig. 7,1. represent that the master controller in the first storage array updates the control in the first storage array
Device running state information processed, 2. represents that controller failure event is sent out by the master controller in the first storage array
Give the master controller in the second storage array, 3. represent that master controller in the second storage array updates the
Controller running state information in two storage arrays.
Master controller in first storage array can process above-mentioned first storage array and the second storage array
Controller number and/or run generation state change event.
In the third specific implementation, pending event is communication link fails event.This implementation
In, the master controller in the first storage array by the running status of the controller in the second storage array from control
Device running state information processed removes;Or, the master controller in the first storage array is by the second storage array
In running status in controller running state information of the running status of controller be updated to off-line or therefore
Barrier.
Communication link between the first storage array and the second storage array breaks down, the first storage array with
Second storage array cannot communicate.The another name of the other side is each saved by the first storage array and the second storage array
Point removes or its running status is updated to off-line or fault, such as Fig. 8 from controller running state information
Shown in.
In 4th kind of specific implementation, pending event is that communication link recovers event, and pending thing
Part is event in the first storage array.
Fig. 9 is the flow chart that the present invention manages the embodiment of the method seven of storage array.Implementing shown in Fig. 2
On the basis of example, as it is shown in figure 9, the method can also include:
Master controller in S901, the first storage array receives the master controller in described second storage array
The recovery request sent, this recovery request carries the running status of each controller in the second storage array.
Master controller in S902, the first storage array is by the operation shape of controller each in the second storage array
State is added to controller running state information, wherein, and the running status of each controller in the second storage array
For online, or, the running status of controller each in the second storage array is updated to online.
Master controller in S903, the first storage array sends the first response message in the second storage array
Master controller, this first response message carries the running status of each controller in the first storage array.
Wherein, the master controller in the first storage array is one in the controller in the first storage array,
Master controller in second storage array is one in the controller in the second storage array.
In 5th kind of specific implementation, pending event is link-recovery event, and pending event is
The event that second storage array side generates.Figure 10 is the embodiment of the method eight that the present invention manages storage array
Flow chart.On the basis of embodiment illustrated in fig. 2, as shown in Figure 10, the method can also include:
S110, when the master controller in the first storage array detect communication link recover event time, first
Master controller in storage array generates recovery request, and this recovery request is carried in the first storage array and respectively controlled
The running status of device processed.
Master controller in S120, the first storage array sends recovery request to the master in the second storage array
Controller.
Master controller in S130, the first storage array receives the master controller in the second storage array and sends
The second response message, this second response message carries the running status of each controller in the second storage array.
Master controller in S140, the first storage array resolves the second response message, obtains the second storage battle array
The running status of each controller in row.
Master controller in S150, the first storage array is by the operation shape of controller each in the second storage array
State is added to controller running state information, wherein, and the running status of each controller in the second storage array
For online, or, the running status of controller each in the second storage array is updated to online.
Wherein, the master controller in the first storage array is one in the controller in the first storage array,
Master controller in second storage array is one in the controller in the second storage array.
It follows that the 4th kind of specific implementation of unified explanation and the 5th kind of specific implementation.
Communication link when between the first storage array and the second storage array recovers normal, the first storage array
With the second storage array by the master controller in the first storage array and the main control in the second storage array
Device is controlled the smooth configuration of device running state information.For smooth configuration.Those skilled in the art are permissible
It is interpreted as: when the communication link between two storage arrays recovers normal, is run shape by initiating establishment controller
One direction of state information jointly creates the opposing party of this controller running state information and inquires about controller information also
Change into another name node to be written in the controller running state information of this locality.If it is to say, creating control
Device running state information processed is to be initiated by the first storage array in the embodiment of the present invention, then the present invention is real
Executing the first storage array in example uses the 5th kind of specific implementation to update controller running state information;
If creating controller running state information is to be initiated, then by the second storage array in the embodiment of the present invention
The first storage array in the embodiment of the present invention uses the 4th kind of specific implementation to update controller and runs shape
State information.
Above-described embodiment illustrates how to update controller running state information by specific implementation, and more
The premise of new controller running state information is intended to create controller running state information.
Create controller running state information may include that
1) master controller in the first storage array receives the life for creating controller running state information
Order.
User sends for creating the order of controller running state information to the first storage array, by first
Master controller in storage array performs this for the order creating controller running state information.
2) master controller in the first storage array obtains the operation shape of the controller in the first storage array
State.
3) master controller in the first storage array sends and creates order to the main control in the second storage array
Device, this establishment order includes the running status of the controller in the first storage array.
First storage array, according to the running status of the controller in the first storage array, generates and creates order
And it is sent to the master controller in the second storage array.
Correspondingly, after the master controller in the second storage array receives establishment order, first obtain this locality
The running status of controller;Then, the master controller in the second storage array is according to creating what order included
The operation shape of the controller in the running status of the controller in the first storage array and the second storage array
State, configures the controller running state information of the second storage array;After configuration is good, the second storage battle array
Master controller in row the controller running state information configured is synchronized in the second storage array its
In his controller;Master controller in second storage array is by the fortune of the controller in this second storage array
Row state, establishment perform result and carry the master controller returned in response message in the first storage array.
4) master controller in the first storage array receives answering of the master controller transmission in the second storage array
Answering message, this response message includes the running status of the controller in the second storage array.
5) master controller in the first storage array is according to the running status of the controller in the first storage array
With the running status of the controller in the second storage array, the controller creating the first storage array runs shape
State information.
To sum up, after the first storage array receives the order for creating controller running state information, will
The running status of the controller in this storage array is sent to the second storage array, and waits the second storage battle array
The execution result of row;Then according to the running status of the controller of two storage arrays, local control is created
Device running state information processed.
Finally, local controller running state information is synchronized to by the master controller in the first storage array
In other controllers in each first storage array.Alternatively, the master controller in the first storage array returns
Return result to user.
In above-described embodiment, bitmap can be used to represent the running status of the controller in the first storage array,
And/or, the running status of the controller in the second storage array.Such as, be expressed as 1 online, fault or
Off-line is expressed as 0, etc., the embodiment of the present invention is not limited.
Figure 11 is the structural representation that the present invention manages the device embodiment one of storage array.The present invention implements
Example provides a kind of device managing storage array, and this device is integrated in the controller of the first storage array,
First storage array and the second storage array pass through communication link.Controller in first storage array
Storage has controller running state information and writes lock authority information, wherein, and controller running state information
Running status including controller each in the second storage array;Write lock authority information and include the first storage array
The person that writes lock rights management of middle LUN, LUN writes lock rights management person is the control in the second storage array
Device processed.
As shown in figure 11, the device 100 of management storage array includes: receiver module 11, sending module
22, detection module 33 and processing module 44.
Wherein, receiver module 11 is for receiving the request to LUN write data.Detection module 33 is used for
The running status of the person that writes lock rights management of detection LUN.Sending module 22 is for connecing when receiver module 11
Receive when the request of described LUN write data, and detection module 33 detects that LUN's writes lock power
Limit gerentocratic running status when being online, by communication link, writing lock rights management person send out to LUN
Send the application request writing locked authority.Receiver module 11 is additionally operable to receive that the person that writes lock rights management provides writes
The response message of locked authority.Processing module 44 writes locking power for obtain that the person that writes lock rights management provides
After limit, write data to described LUN.
The device of the present embodiment, may be used for performing the technical scheme of above-mentioned any means embodiment, in fact
Existing principle is similar with technique effect, and here is omitted.
On the basis of the above, processing module 44 can be also used for: is obtaining what the person that writes lock rights management provided
After writing locked authority, trigger sending module 22 and send in the mirror image LUN of LUN to the second storage array
Writing the instruction of described data, this mirror image LUN is by the controller management in the second storage array.
Additionally, receiver module 11 can be also used for receiving the mirror image LUN write that the second storage array sends
The response message completed.Processing module 44 can be also used for when LUN has write, and receiver module
After 11 receive the response message that mirror image LUN has write, locked authority is write in release.
Further, sending module 22 can be also used for before locked authority is write in processing module 44 release,
Receiver module 11 receives the another one request to LUN write data, and detection module 33 detects
When the running status of the person that writes lock rights management to LUN is online, by communication link, to LUN's
Person sends the application request writing locked authority to write lock rights management.Receiver module 11 can be also used for reception and writes
Lock rights management person returns refusal and writes the response message of locked authority.
In a kind of implementation, the controller of the device 100 of integrated management storage array is the first storage battle array
Master controller in row, the master controller in the first storage array is in the controller in the first storage array
One.
In this implementation, receiver module 11 can be also used for receiving the main control in the second storage array
The pending event that device sends, pending event occurs in the second storage array, wherein, the second storage
Master controller in array is one in the controller in the second storage array.Processing module 44 is all right
For according to pending event, updating the control in the second storage array that controller running state information comprises
The running status of device processed.Sending module 22 can be also used for sending the controller running state information after updating
To other controllers in the first storage array, each other controller updates the described control of oneself storage
Device running state information.
Alternatively, when pending event is controller failure event, processing module 44 performs according to waiting to locate
Director's part, the operation shape of the controller in the second storage array that renewal controller running state information comprises
During state, can be particularly as follows: the running status of the controller of generator controller event of failure be transported from controller
Row status information removes, or, in controller running state information, by generator controller event of failure
The running status of controller be updated to off-line or fault.
In another kind of implementation, the controller of the device 100 of integrated management storage array is the first storage
Master controller in array, the master controller in the first storage array is the controller in the first storage array
In one.
In this implementation, receiver module 11 can be also used for receiving the main control in the second storage array
The recovery request that device sends, this recovery request carries the running status of each controller in the second storage array.
Processing module 44 can be also used for adding to controlling the running status of controller each in the second storage array
Device running state information, wherein, in the second storage array, the running status of each controller is online, or,
The running status of controller each in the second storage array is updated to online.Sending module 22 can be also used for
Sending the first response message to the master controller in the second storage array, the first response message carries first and deposits
The running status of each controller in storage array.Wherein, the master controller in the second storage array is second to deposit
In storage array one in controller.
In another implementation, the controller of the device 100 of integrated management storage array is the first storage
Master controller in array, the master controller in the first storage array is the controller in the first storage array
In one.
In this implementation, detection module 33 can be also used for detecting communication link and recovers event.Process
Module 44 can be also used for, when detection module 33 detects that communication link recovers event, generating and recovering to ask
Asking, this recovery request carries the running status of each controller in the first storage array.Sending module 22 also may be used
For sending recovery request to the master controller in the second storage array.Receiver module 11 can be also used for
Receiving the second response message that the master controller in the second storage array sends, this second response message carries
The running status of each controller in second storage array.Processing module 44 can be also used for resolving the second response
Message, obtains the running status of each controller in the second storage array;And, by each in the second storage array
The running status of controller is added to controller running state information, wherein, respectively controls in the second storage array
The running status of device processed is online, or, the running status of controller each in the second storage array is updated to
Online.Wherein, the master controller in the second storage array is in the controller in the second storage array
Individual.
Need explanation, in above-mentioned any embodiment, controller running status may include that online,
Any one in off-line and fault.It addition, for newly-increased controller event and communication link fails thing
Part and establishment controller running state information etc., do not repeat them here, specifically describe and refer to above-mentioned side
Method embodiment.
Figure 12 is the structural representation that the present invention manages the device embodiment two of storage array.Such as Figure 12 institute
Showing, the device 200 of the management storage array that the present embodiment provides includes processor 210 and memorizer 220.
Wherein, memorizer 220 storage performs instruction, when the device 200 managing storage array runs, processes
Communicating between device 210 with memorizer 220, processor 210 calls and performs instruction in memorizer 220,
For the method performing above-mentioned management storage array, it is similar with technique effect that it realizes principle, the most no longer
Repeat.
In several embodiments provided herein, it should be understood that disclosed apparatus and method,
Can realize by another way.Such as, apparatus embodiments described above is only schematically,
Such as, described unit or the division of module, it is only a kind of logic function and divides, actual permissible when realizing
Have other dividing mode, the most multiple unit or module can in conjunction with or be desirably integrated into another and be
Unite, or some features can be ignored, or do not perform.Another point, shown or discussed each other
Coupling or direct-coupling or communication connection can be the INDIRECT COUPLING by some interfaces, equipment or module or
Communication connection, can be electrical, machinery or other form.
The described module illustrated as separating component can be or may not be physically separate, as
The parts that module shows can be or may not be physical module, i.e. may be located at a place, or
Can also be distributed on multiple NE.Can select therein some or all of according to the actual needs
Module realizes the purpose of the present embodiment scheme.
One of ordinary skill in the art will appreciate that: realize all or part of step of above-mentioned each method embodiment
Can be completed by the hardware that programmed instruction is relevant.Aforesaid program can be stored in an embodied on computer readable
In storage medium.This program upon execution, performs to include the step of above-mentioned each method embodiment;And it is aforesaid
Storage medium includes: the various media that can store program code such as ROM, RAM, magnetic disc or CD.
Last it is noted that various embodiments above is only in order to illustrate technical scheme, rather than to it
Limit;Although the present invention being described in detail with reference to foregoing embodiments, the ordinary skill of this area
Personnel it is understood that the technical scheme described in foregoing embodiments still can be modified by it, or
The most some or all of technical characteristic is carried out equivalent;And these amendments or replacement, do not make phase
The essence answering technical scheme departs from the scope of various embodiments of the present invention technical scheme.
Claims (18)
1. the method managing storage array, it is characterised in that be applied to the control of the first storage array
In device, described first storage array and the second storage array are by communication link, described first storage
Controller storage in array has controller running state information and writes lock authority information, wherein, described
Controller running state information includes the running status of each controller in described second storage array, described in write
Lock authority information includes write lock rights management person, the institute of LUN LUN in described first storage array
Stating writing lock rights management person being the controller in described second storage array of LUN, described method includes:
When the controller in described first storage array receives to the request of described LUN write data,
And when the running status of the person that writes lock rights management of described LUN is online, in described first storage array
Controller by described communication link, send to the person that writes lock rights management of described LUN and write locked authority
Application request;
What the person that writes lock rights management described in obtaining when the controller in described first storage array provided writes locking
After authority, write described data to described LUN.
Method the most according to claim 1, it is characterised in that the control in described first storage array
Device processed passes through described communication link, sends the application writing locked authority to the lock rights holder of described LUN
After request, described method also includes:
What the person that writes lock rights management described in obtaining when the controller in described first storage array provided writes locking
After authority, send to described second storage array in the mirror image LUN of described LUN, write described data
Instruction, described mirror image LUN is by the controller management in described second storage array.
Method the most according to claim 2, it is characterised in that described method also includes:
When described LUN has write, and receive the described mirror image LUN that described second storage array sends
After the response message write, described in the controller release in described first storage array, write locked authority.
4. according to the method described in any one of claims 1 to 3, it is characterised in that described method is also wrapped
Include:
Before writing locked authority described in controller release in described first storage array, described first deposits
Controller in storage array receives the another one request to described LUN write data, and described
When the running status of the person that writes lock rights management of LUN is online, by described communication link, to described
The person that writes lock rights management of LUN sends the application request writing locked authority;
Controller in described first storage array receive described in the person that writes lock rights management return refusal and write
The response message of locked authority.
Method the most according to claim 1, it is characterised in that described method also includes:
The master controller that master controller in described first storage array receives in described second storage array is sent out
The pending event sent, described pending event occurs in described second storage array, wherein, described
Master controller in first storage array is in the controller in described first storage array, described
Master controller in second storage array is in the controller in described second storage array;
Master controller in described first storage array, according to described pending event, updates described controller
The running status of the controller in described second storage array that running state information comprises;
Master controller in described first storage array sends the controller running state information after updating to the
Other controllers in one storage array, each other controller updates the described controller fortune of oneself storage
Row status information.
Method the most according to claim 5, it is characterised in that when described pending event is for controlling
During device event of failure, the master controller in described first storage array, according to described pending event, updates
The running status of the controller in described second storage array that described controller running state information comprises,
Including:
The controller of described controller failure event will be there is in the master controller in described first storage array
Running status removes from described controller running state information, or
In described controller running state information, the master controller in described first storage array will occur
The running status of the controller of described controller failure event is updated to off-line or fault.
Method the most according to claim 1, it is characterised in that described method also includes:
The master controller that master controller in described first storage array receives in described second storage array is sent out
The recovery request sent, described recovery request carries the running status of each controller in described second storage array;
Master controller in described first storage array is by the operation of each controller in described second storage array
State is added to described controller running state information, wherein, each controller in described second storage array
Running status be online, or, the running status of each controller in described second storage array is updated to
Online;
Master controller in described first storage array sends the first response message to described second storage array
In master controller, described first response message carries the operation of each controller in described first storage array
State;
Wherein, the master controller in described first storage array is the controller in described first storage array
In one, the master controller in described second storage array is the controller in described second storage array
In one.
Method the most according to claim 1, it is characterised in that described method also includes:
When the master controller in described first storage array detects that communication link recovers event, described the
Master controller in one storage array generates recovery request, and described recovery request carries described first storage battle array
The running status of each controller in row;
Master controller in described first storage array sends described recovery request to described second storage array
In master controller;
The master controller that master controller in described first storage array receives in described second storage array is sent out
The second response message sent, described second response message carries each controller in described second storage array
Running status;
Master controller in described first storage array resolves described second response message, obtains described second
The running status of each controller in storage array;
Master controller in described first storage array is by the operation of each controller in described second storage array
State is added to described controller running state information, wherein, each controller in described second storage array
Running status be online, or, the running status of each controller in described second storage array is updated to
Online;
Wherein, the master controller in described first storage array is the controller in described first storage array
In one, the master controller in described second storage array is the controller in described second storage array
In one.
9. according to the method described in any one of claim 1 to 8, it is characterised in that described controller is transported
Row state includes: any one in online, off-line and fault.
10. the device managing storage array, it is characterised in that be integrated in the control of the first storage array
In device processed, described first storage array and the second storage array are by communication link, and described first deposits
Controller storage in storage array has controller running state information and writes lock authority information, wherein, and institute
State controller running state information and include the running status of each controller in described second storage array, described
Write lock authority information and include the person that writes lock rights management of LUN LUN in described first storage array,
Described LUN writes lock rights management person is the controller in described second storage array, and described device includes
Receiver module, sending module, detection module and processing module;
Described receiver module, for receiving the request to described LUN write data;
Described detection module, for detecting the running status of the person that writes lock rights management of described LUN;
Described sending module, for receiving described to described LUN write data when described receiver module
During request, and described detection module detects that the running status of the person that writes lock rights management of described LUN is
Time online, by described communication link, send to the person that writes lock rights management of described LUN and write locked authority
Application request;
Described receiver module, be additionally operable to that the person that writes lock rights management described in receiving provides write locked authority should
Answer message;
Described processing module, after what the person that writes lock rights management described in obtain provided writes locked authority, to
Described LUN writes described data.
11. devices according to claim 10, it is characterised in that described processing module is additionally operable to:
The person that writes lock rights management described in obtaining provide write locked authority after, trigger described sending module to
Described second storage array sends the instruction writing described data in the mirror image LUN of described LUN, institute
State mirror image LUN by the controller management in described second storage array.
12. devices according to claim 11, it is characterised in that
Described receiver module, is additionally operable to receive the described mirror image LUN write that described second storage array sends
The response message completed;
Described processing module, is additionally operable to when described LUN has write, and described receiver module receives institute
After stating the response message that mirror image LUN has write, described in release, write locked authority.
13. according to the device described in any one of claim 10 to 12, it is characterised in that
Described sending module, is additionally operable to before writing locked authority described in the release of described processing module, described
Receiver module receives the another one request to described LUN write data, and the inspection of described detection module
When the running status of the person that writes lock rights management measuring described LUN is online, by described communication link,
The application request writing locked authority is sent to the person that writes lock rights management of described LUN;
Described receiver module, be additionally operable to receive described in the person that writes lock rights management return refusal and write locking power
The response message of limit.
14. devices according to claim 10, it is characterised in that the controller of integrated described device
For the master controller in described first storage array, the master controller in described first storage array is described
In controller in first storage array one;
Described receiver module, what the master controller being additionally operable to receive in described second storage array sent waits to locate
Director's part, described pending event occurs in described second storage array, wherein, described second storage
Master controller in array is in the controller in described second storage array;
Described processing module, is additionally operable to, according to described pending event, update described controller running status
The running status of the controller in described second storage array that information comprises;
Described sending module, is additionally operable to send the controller running state information after updating to the first storage battle array
Other controllers in row, each other controller updates the described controller running status letter of oneself storage
Breath.
15. devices according to claim 14, it is characterised in that when described pending event is control
During device event of failure processed, described processing module performs according to described pending event, updates described controller
During the running status of the controller in described second storage array that running state information comprises, particularly as follows:
The running status of the controller of described controller failure event will be occurred from described controller running status
Information removes, or
In described controller running state information, the controller of described controller failure event will be there is
Running status is updated to off-line or fault.
16. devices according to claim 10, it is characterised in that the controller of integrated described device
For the master controller in described first storage array, the master controller in described first storage array is described
In controller in first storage array one;
Described receiver module, is additionally operable to the recovery of the master controller transmission receiving in described second storage array
Request, described recovery request carries the running status of each controller in described second storage array;
Described processing module, is additionally operable to add the running status of each controller in described second storage array
To described controller running state information, wherein, the operation shape of each controller in described second storage array
State is online, or, the running status of each controller in described second storage array is updated to online;
Described sending module, is additionally operable to send the first response message to the master control in described second storage array
Device processed, described first response message carries the running status of each controller in described first storage array;
Wherein, the master controller in described second storage array is the controller in described second storage array
In one.
17. devices according to claim 10, it is characterised in that the controller of integrated described device
For the master controller in described first storage array, the master controller in described first storage array is described
In controller in first storage array one;
Described detection module, is additionally operable to detect communication link and recovers event;
Described processing module, is additionally operable to when described detection module detects that described communication link recovers event,
Generating recovery request, described recovery request carries the running status of each controller in described first storage array;
Described sending module, is additionally operable to send described recovery request to the master control in described second storage array
Device processed;
Described receiver module, is additionally operable to the second of the master controller transmission receiving in described second storage array
Response message, described second response message carries the running status of each controller in described second storage array;
Described processing module, is additionally operable to resolve described second response message, obtains described second storage array
In the running status of each controller;And, the running status of each controller in described second storage array is added
Add to described controller running state information, wherein, the operation of each controller in described second storage array
State is online, or, the running status of each controller in described second storage array is updated to online;
Wherein, the master controller in described second storage array is the controller in described second storage array
In one.
18. according to the device described in any one of claim 10 to 17, it is characterised in that described control
Device running status includes: any one in online, off-line and fault.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510307041.XA CN106250048B (en) | 2015-06-05 | 2015-06-05 | Manage the method and device of storage array |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510307041.XA CN106250048B (en) | 2015-06-05 | 2015-06-05 | Manage the method and device of storage array |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106250048A true CN106250048A (en) | 2016-12-21 |
CN106250048B CN106250048B (en) | 2019-06-28 |
Family
ID=57626470
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510307041.XA Active CN106250048B (en) | 2015-06-05 | 2015-06-05 | Manage the method and device of storage array |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106250048B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108345515A (en) * | 2017-01-22 | 2018-07-31 | 中国移动通信集团四川有限公司 | Storage method and device and its storage system |
WO2019080150A1 (en) * | 2017-10-25 | 2019-05-02 | 华为技术有限公司 | Dual active storage system and address allocation method |
CN109857341A (en) * | 2019-01-15 | 2019-06-07 | 新华三技术有限公司成都分公司 | Write the determination method and device that lock prefetches length |
CN110209641A (en) * | 2018-02-12 | 2019-09-06 | 杭州宏杉科技股份有限公司 | A kind of trunking service processing method and device applied in more controlled storage systems |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6073218A (en) * | 1996-12-23 | 2000-06-06 | Lsi Logic Corp. | Methods and apparatus for coordinating shared multiple raid controller access to common storage devices |
CN102541471A (en) * | 2011-12-28 | 2012-07-04 | 创新科软件技术(深圳)有限公司 | Storage system with multiple controllers |
CN103731485A (en) * | 2013-12-26 | 2014-04-16 | 华为技术有限公司 | Network equipment, cluster storage system and distributed lock management method |
CN104486319A (en) * | 2014-12-09 | 2015-04-01 | 上海爱数软件有限公司 | Real-time synchronization method and real-time synchronization system for configuration file applied to high-availability system |
CN104520845A (en) * | 2012-09-06 | 2015-04-15 | 惠普发展公司,有限责任合伙企业 | Scalable file system |
-
2015
- 2015-06-05 CN CN201510307041.XA patent/CN106250048B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6073218A (en) * | 1996-12-23 | 2000-06-06 | Lsi Logic Corp. | Methods and apparatus for coordinating shared multiple raid controller access to common storage devices |
CN102541471A (en) * | 2011-12-28 | 2012-07-04 | 创新科软件技术(深圳)有限公司 | Storage system with multiple controllers |
CN104520845A (en) * | 2012-09-06 | 2015-04-15 | 惠普发展公司,有限责任合伙企业 | Scalable file system |
CN103731485A (en) * | 2013-12-26 | 2014-04-16 | 华为技术有限公司 | Network equipment, cluster storage system and distributed lock management method |
CN104486319A (en) * | 2014-12-09 | 2015-04-01 | 上海爱数软件有限公司 | Real-time synchronization method and real-time synchronization system for configuration file applied to high-availability system |
Non-Patent Citations (2)
Title |
---|
严亮: "双控制器RAID系统的研究与实现", 《万方数据(学位)》 * |
赵旺: "分布式并行文件系统锁管理的研究与设计", 《中国优秀硕士学位论文全文数据库(信息科技辑)》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108345515A (en) * | 2017-01-22 | 2018-07-31 | 中国移动通信集团四川有限公司 | Storage method and device and its storage system |
WO2019080150A1 (en) * | 2017-10-25 | 2019-05-02 | 华为技术有限公司 | Dual active storage system and address allocation method |
US11507505B2 (en) | 2017-10-25 | 2022-11-22 | Huawei Technologies Co., Ltd. | Active-active storage system and address assignment method |
US11704243B2 (en) | 2017-10-25 | 2023-07-18 | Huawei Technologies Co., Ltd. | Active-active storage system and address assignment method |
CN110209641A (en) * | 2018-02-12 | 2019-09-06 | 杭州宏杉科技股份有限公司 | A kind of trunking service processing method and device applied in more controlled storage systems |
CN109857341A (en) * | 2019-01-15 | 2019-06-07 | 新华三技术有限公司成都分公司 | Write the determination method and device that lock prefetches length |
CN109857341B (en) * | 2019-01-15 | 2022-04-12 | 新华三技术有限公司成都分公司 | Method and device for determining write lock prefetch length |
Also Published As
Publication number | Publication date |
---|---|
CN106250048B (en) | 2019-06-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101547719B1 (en) | Maintaining data integrity in data servers across data centers | |
CN103777897B (en) | Method and system for copying data between primary and secondary storage locations | |
CN105577744B (en) | Cloud consistency technology | |
CN104994168B (en) | Distributed storage method and distributed memory system | |
US20070079171A1 (en) | No data loss it disaster recovery over extended distances | |
CN110807064B (en) | Data recovery device in RAC distributed database cluster system | |
CN106331166B (en) | A kind of access method and device of storage resource | |
EP1873645A1 (en) | Storage system and data replication method | |
CN102088490B (en) | Data storage method, device and system | |
CN104023246B (en) | A kind of video data private cloud storage system and video data private cloud storage method | |
CN103840961A (en) | Dual-computer hot backup system | |
CN108932338A (en) | Data-updating method, device, equipment and medium | |
CN112003716A (en) | Data center dual-activity implementation method | |
CN106250048A (en) | The method and device of management storage array | |
CN106874136A (en) | The fault handling method and device of a kind of storage system | |
CN105404565B (en) | A kind of dual-active data guard method and device | |
CN107135097A (en) | The disaster tolerance system and disaster recovery method filed based on bookkeeping | |
CN105988901A (en) | Data copying method and storage system | |
WO2021115043A1 (en) | Distributed database system and data disaster backup drilling method | |
US20150195167A1 (en) | Availability device, storage area network system with availability device and methods for operation thereof | |
CN110290163A (en) | A kind of data processing method and device | |
CN111600953B (en) | Method for realizing distributed deployment based on honeypot system | |
CN106027313B (en) | Network link disaster tolerance system and method | |
WO2015196692A1 (en) | Cloud computing system and processing method and apparatus for cloud computing system | |
EP3316114A1 (en) | Data reading and writing method and device |
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 |