CN108156203A - A kind of storage system and storage node management method - Google Patents
A kind of storage system and storage node management method Download PDFInfo
- Publication number
- CN108156203A CN108156203A CN201611109234.5A CN201611109234A CN108156203A CN 108156203 A CN108156203 A CN 108156203A CN 201611109234 A CN201611109234 A CN 201611109234A CN 108156203 A CN108156203 A CN 108156203A
- Authority
- CN
- China
- Prior art keywords
- information
- memory node
- database
- manager
- storage system
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
- G06F16/2308—Concurrency control
- G06F16/2315—Optimistic concurrency control
- G06F16/2329—Optimistic concurrency control using versioning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a kind of storage system and storage node management method, system includes at least two memory nodes：Role Information is the first memory node of manager and Role Information is not the second memory node of manager；Operational management service in every memory node；After second memory node detects that the first memory node breaks down, by the election mechanism under the management service that is run in each memory node, judge whether self-role information is manager；If so, judging itself whether be configured with the address information for accessing storage system, if be not configured, simultaneously configuration address information is obtained.Using this programme, after breaking down as the memory node of manager, utilize the election mechanism under the management service run in each memory node, each memory node is likely to become new manager, the address information of new manager's configuration access storage system, the storage system can continue to provide storage service；This improves the reliabilities of storage system.
Description
Technical field
The present invention relates to technical field of data processing, more particularly to a kind of storage system and storage node management method.
Background technology
Existing storage system generally comprises management node and memory node, and memory node is physics Jie for storing data
Matter, management node are managed the data of memory node storage each in system.
In the smaller storage system of some storage data quantities, independent management node is usually not provided with, but selectes one
As manager, manager carries out pipe to the data of memory node storage each in storage system for platform or several memory nodes
Reason.
In said program, manager's negligible amounts, if manager is broken down, storage system paralysis, it is impossible to
Continue to provide storage service, that is to say, that the reliability of the storage system is poor.
Invention content
The embodiment of the present invention is designed to provide a kind of storage system and storage node management method, improves storage system
Reliability.
In order to achieve the above objectives, the embodiment of the invention discloses a kind of storage system, including at least two memory nodes, often
Operational management service in platform memory node；The first storage Role Information for manager is included at least two memory nodes
Node and Role Information are not the second memory node of manager；
Second memory node after detecting that first memory node breaks down, passes through each memory node
Election mechanism under the management service of middle operation judges whether self-role information is manager；If so, whether judge itself
The address information for accessing the storage system is configured with, if be not configured, obtains and described address information is configured.
Optionally, second memory node can be also used for through the election mechanism, judge whether itself is interim
Manager：
If so, initiating election, according to the vote information that each election participant sends, determine each in the storage system
The Role Information of memory node, and using the management service run in each memory node, to the angle of each memory node
Color information is identified, so that each memory node determines self-role information according to the mark；
If not, judging whether itself is election participant, if election participant, initiate to elect in caretaker manager
Afterwards, vote information is sent to caretaker manager.
Optionally, second memory node can be also used for judging self-role information not for the situation of manager
Under, judge itself whether be configured with the address information for accessing the storage system, if configured, delete described address letter
Breath.
Optionally, second memory node can be also used for the management service by self-operating, receive fault cues
Information, the fault cues information represent that first memory node breaks down；
Second memory node, after can be also used for election, according to what is identified in the management service of self-operating
Role Information, judge self-role information whether manager.
Optionally, second memory node can be also used for the management service by self-operating, receive data change
Information；According to the data modification information, the database of itself storage is updated.
Optionally, it is stored in the database in the storage system and stores the index information of data and the storage
The nodal information of each memory node in system；
Second memory node can be also used for according to the index of change data included in the data modification information
Information updates the index information in the database；According to the section of change memory node included in the data modification information
Point information, updates the nodal information in the database.
Optionally, it is stored in the database in the storage system and stores the index information of data and the storage
The nodal information of each memory node in system；
Second memory node can be also used for reading index information and the institute of each memory node storage data
State the nodal information of each memory node；
The index information read and the index information in the database are compared, obtain the first comparison knot
Fruit；
The nodal information read and the nodal information in the database are compared, obtain the second comparison knot
Fruit；
According to first comparing result and second comparing result, the database is updated.
Optionally, the corresponding first version number of every index information and every node letter are also stored in the database
Cease the corresponding second edition number；
Second memory node can be also used for, according to first comparing result, updating the rope in the database
Fuse ceases and its corresponding first version number；
According to second comparing result, the nodal information in the database and its corresponding second edition number are updated.
In order to achieve the above objectives, the embodiment of the invention also discloses a kind of storage node management methods, are applied to storage
Memory node in system, operational management service in each memory node of the storage system；The method includes：
After detecting the memory node failure that Role Information is manager, pass through the management run in each memory node
Election mechanism under service, judges whether self-role information is manager；
If so, judging itself whether be configured with the address information for accessing the storage system, if be not configured, obtain
And described address information is configured.
Optionally, after detecting the memory node failure that Role Information is manager, the method can also wrap
It includes：
By the election mechanism, judge whether itself is caretaker manager：
If so, initiating election, according to the vote information that each election participant sends, determine each in the storage system
The Role Information of memory node, and using the management service run in each memory node, to the angle of each memory node
Color information is identified；
If not, judging whether itself is election participant, if election participant, initiate to elect in caretaker manager
Afterwards, vote information is sent to caretaker manager.
Optionally, in the case where judging self-role information not for manager, the method further includes：
Judge itself whether be configured with the address information for accessing the storage system, if configured, delete describedly
Location information.
Optionally, operational management service in the memory node in the storage system；It is manager to detect Role Information
Memory node break down the step of can include：
By the management service of self-operating, fault cues information is received, the fault cues information represents Role Information
Memory node for manager breaks down；
The election mechanism by under the management service that is run in each memory node judges whether self-role information manages
The step of reason person, can include：
After election, according to the Role Information identified in the management service of self-operating, judge that self-role information is
No manager.
Optionally, the method can also include：
By the management service of self-operating, data modification information is received；
According to the data modification information, the database of itself storage is updated.
Optionally, it is stored in the database in the storage system and stores the index information of data and the storage
The nodal information of each memory node in system；
In the case of index information comprising change data in the data modification information, itself storage of the update
The step of database, can include：
According to the index information of change data included in the data modification information, the index in the database is updated
Information；
In the case of nodal information comprising change memory node in the data modification information, the update itself is deposited
The step of database of storage, can include：
According to the nodal information of change memory node included in the data modification information, update in the database
Nodal information.
Optionally, it is stored in the database in the storage system and stores the index information of data and the storage
The nodal information of each memory node in system；The step of database of itself storage of the update, includes：
Read the index information of each memory node storage data and the nodal information of each memory node；
The index information read and the index information in the database are compared, obtain the first comparison knot
Fruit；
The nodal information read and the nodal information in the database are compared, obtain the second comparison knot
Fruit；
According to first comparing result and second comparing result, the database is updated.
Optionally, the corresponding first version number of every index information and every node letter are also stored in the database
Cease the corresponding second edition number, it is described according to first comparing result and second comparing result, update the database
The step of can include：
According to first comparing result, the index information in the database and its corresponding first version number are updated；
According to second comparing result, the nodal information in the database and its corresponding second edition number are updated.
It can be seen that using the embodiment of the present invention, after breaking down as the memory node of manager, each storage is utilized
Election mechanism under the management service run in node, each memory node are likely to become new manager, new manager
The address information of the configuration access storage system, the storage system can continue to provide storage service；This improves storages to be
The reliability of system.
Certainly, it implements any of the products of the present invention or method must be not necessarily required to reach all the above excellent simultaneously
Point.
Description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention, for those of ordinary skill in the art, without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of structure diagram of storage system provided in an embodiment of the present invention；
Fig. 2 is a kind of flow diagram of storage node management method provided in an embodiment of the present invention.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other without creative efforts
Embodiment shall fall within the protection scope of the present invention.
In order to solve the above-mentioned technical problem, an embodiment of the present invention provides a kind of storage system and memory node managers
Method, this method can be applied to the memory node in the storage system.It is to storage provided in an embodiment of the present invention first below
System is described in detail.
Storage system can be with as shown in Figure 1, include at least two memory nodes：Memory node 1, memory node 2 ... are deposited
Node N is stored up, alternatively, memory node 1 and memory node 2 can also be included only, is not limited specifically.It is run in every memory node
Management service；The first memory node and Role Information that Role Information is manager are included at least two memory nodes
It is not the second memory node of manager.
Second memory node can be used for after detecting that first memory node breaks down, pass through each storage
Election mechanism under the management service run in node judges whether self-role information is manager；If so, judge itself
The address information for accessing the storage system whether is configured with, if be not configured, obtains and described address information is configured.
It should be noted that the management service run in each memory node can communicate, for example, the management service can
To include zookeeper services, the event notification mechanism that each memory node is serviced by zookeeper can detect role
Information breaks down for the memory node of manager.
Specifically, the second memory node can be serviced by the zookeeper of self-operating, fault cues information is received,
If memory node receives fault cues information, then it represents that Role Information is the memory node (the first memory node) of manager
It breaks down.
In this case, the election mechanism under being serviced by zookeeper, elects new manager.As one
Kind embodiment, second memory node can judge whether itself is caretaker manager by the election mechanism：
If so, initiating election, according to the vote information that each election participant sends, determine each in the storage system
The Role Information of memory node, and using the management service run in each memory node, to the angle of each memory node
Color information is identified；
If not, judging whether itself is election participant, if election participant, initiate to elect in caretaker manager
Afterwards, vote information is sent to caretaker manager.
Specifically, it is serviced by zookeeper, a caretaker manager can be generated, which can send out
Play election.The caretaker manager can specify other memory nodes as election participant：It can specify all in addition to itself
Memory node is election participant, can also be arbitrarily designated odd number memory node as election participant, not limit specifically.
Vote information is sent to caretaker manager by the memory node for being designated as election participant.Caretaker manager according to
The vote information that each election participant received sends, determines that the Role Information of each memory node (namely determines management
Person and non-management person).Caretaker manager utilizes the zookeeper run in each memory node to service, to the angle of each memory node
Color information is identified.In this way, each memory node can service after election according to the zookeeper of self-operating
The Role Information of middle mark judges whether self-role information is manager.
Likewise, second memory node can judge whether self-role information is manager, such as after election
Fruit is then to judge itself whether be configured with the address information for accessing the storage system, if be not configured, obtains and institute is configured
State address information.
In the present embodiment, if the second memory node judges self-role information not for manager, second storage
Node may determine that the address information for itself whether being configured with and accessing storage system, if configured, delete the address information.
The address information can be understood as virtual ip address, that is, the virtual ip address that storage system externally provides, and use
Family can access the storage system by the virtual ip address.Only have manager that the virtual ip address is configured in storage system, because
This, if former manager is broken down, after certain memory node becomes new management person, new management person should be configured with address letter
Breath.Specifically, the address information can be directly acquired by order line.
In addition, if non-management person is configured with the address information, which will appear unknown error, therefore, when depositing
Storage node judges self-role information not for manager and when being configured with the address information, which should believe the address
Breath is deleted from itself.
As one embodiment of the present invention, database can be stored in each memory node, memory node passes through certainly
The management service of body operation after receiving data modification information, can update the database of itself storage.
Specifically, if data change occurs for the memory node in storage system, which can be by the data
Altering event notifies each memory node by the zookeeper event notification mechanisms serviced.Each memory node passes through self-operating
Zookeeper services can receive data modification information.
Data change the change that can include index information, can also include the change of nodal information.For example, when depositing
In storage system when increase, deletion or mobile storage data, the index information of data changes；When certain in storage system stores
When node increases newly, deletes child node, nodal information changes.Index information can be not only included in above-mentioned database, but also is included
Nodal information.
As a kind of embodiment, the index information of change data can be included in the data modification information, in this way, can be with
According to the index information of change data included in data modification information, the index information in library is updated the data.
As a kind of embodiment, the nodal information of change memory node can also be included in the data modification information, this
Sample can update the data the nodal information in library according to the nodal information of change memory node included in data modification information.
As a kind of embodiment, index information can not be included in the data modification information, also not comprising nodal information,
Only play suggesting effect.Memory node reads the index of each memory node storage data after the data modification information is received
The nodal information of information and each memory node；Index information in the index information and database that are read is compared,
Obtain the first comparing result；Nodal information in the nodal information and database that are read is compared, obtains second pair
Compare result；According to the first comparing result and the second comparing result, library is updated the data.
For the convenience of description, index information is referred to as recording by content with nodal information below.
Specifically, the record in every memory node can be directed to, by its (being assumed to be A) and the record B of own database
It is compared, is not present in B if there is some contents A1 exists only in A, then A1 can be inserted into own database.
A is not present in, and in the child node of the corresponding memory nodes of B1 not if there is some contents B1 exists only in B
There are B1, then can delete B1 from local data base.
A is not present in, and there is no son sections under the corresponding memory nodes of B2 if there is some contents B2 exists only in B
Point, it may be considered that loss of data in A, in this case, record that can be to B2 in corresponding memory node are updated.
In the present embodiment, the corresponding first version number of every index information and every can also be stored in database
The corresponding second edition number of nodal information, therefore, can update the rope in the database according to first comparing result
Fuse ceases and its corresponding first version number；According to second comparing result, update nodal information in the database and
Its corresponding second edition number.
Specifically, every record is updated once, corresponding version number can add 1.
Due to storing the version number of every record, what the record in above-mentioned memory node corresponding to B2 was updated
In the process, it in the record in each memory node, can determine the record of version number's maximum, that is, determine newest note
Record, according to newest record, the record to B2 in corresponding memory node is updated.In addition, other memory nodes can also
According to the newest record update own database.
It, can also be according to respectively being deposited in storage system it should be noted that if memory node is reached the standard grade again after breaking down
The database of node is stored up, own database is updated.
A kind of specific embodiment is described below：
All operational management services on each memory node in storage system are serviced comprising zookeeper in the management service.It deposits
Manager's (the first memory node) in storage system is broken down, other memory nodes (the second memory node) pass through
The event notification mechanism of zookeeper services detects that the manager is broken down.
Election mechanism under being serviced by zookeeper, the person that can elect new management.Specifically, it is taken by zookeeper
Business, can generate a caretaker manager, which can initiate to elect.The caretaker manager can specify other to deposit
Node is stored up as election participant, specifically, it is possible to specify all memory nodes in addition to itself are election participant, can also
Odd number memory node is arbitrarily designated as election participant, is not limited specifically.
Vote information is sent to caretaker manager by the memory node for being designated as election participant.Caretaker manager according to
The vote information that each election participant received sends, determines that the Role Information of each memory node (namely determines management
Person and non-management person).Caretaker manager utilizes the zookeeper run in each memory node to service, to the angle of each memory node
Color information is identified.In this way, the role that each memory node identifies in being serviced according to the zookeeper of self-operating believes
Breath, judges whether self-role information is manager.
The address information should be configured with as the memory node of new management person, can just provide a user storage system in this way
Management service.Specifically, the address information can be directly acquired by order line.
In addition, if non-management person is configured with the address information, which will appear unknown error, therefore, when depositing
Storage node judges self-role information not for manager and when being configured with the address information, which should believe the address
Breath is deleted from itself.
It should be noted that this programme can also be applied by being provided with the storage system of management node：The system can have two
Kind of operating mode, under the first operating mode, management node and memory node normal operation, under second of operating mode, management
Node does not run or is run as memory node, and the manager in memory node carries out data management.The system can be two
It is switched under kind operating mode.
Specifically, the management service run in memory node can determine the work at present of storage system according to configuration file
Pattern if current operation mode is second of operating mode, can perform this programme.That is, it is provided with management section
The storage system of point can flexibly choose whether to perform this programme according to actual conditions.
Using illustrated embodiment of the present invention, after breaking down as the memory node of manager, each memory node is utilized
Election mechanism under the management service of middle operation, each memory node are likely to become new manager, new manager configuration
The address information of the storage system is accessed, which can continue to provide storage service；This improves storage systems
Reliability.
Fig. 2 is a kind of flow diagram of storage node management method provided in an embodiment of the present invention, shown in Fig. 2 of the present invention
Embodiment can be applied to the memory node in storage system, operational management service in each memory node of the storage system；
Embodiment illustrated in fig. 2 of the present invention can also be applied to the second memory node in embodiment illustrated in fig. 1 of the present invention.Fig. 2 includes：
S201：After detecting the memory node failure that Role Information is manager, by being run in each memory node
Management service under election mechanism, judge whether self-role information is manager；If so, perform S202.
It, can be with operational management service, in each memory node in the memory node in storage system as a kind of embodiment
The management service of operation can communicate, and be serviced for example, the management service can include zookeeper, each memory node leads to
The event notification mechanism of zookeeper services is crossed, can detect that the memory node that Role Information is manager breaks down.
Specifically, memory node can be serviced by the zookeeper of self-operating, fault cues information is received, if
Memory node receives fault cues information, then it represents that Role Information breaks down for the memory node of manager.
In this case, the election mechanism under being serviced by zookeeper, elects new manager.As one
Kind embodiment, the memory node can judge whether itself is caretaker manager by the election mechanism：
If so, initiating election, according to the vote information that each election participant sends, determine each in the storage system
The Role Information of memory node, and using the management service run in each memory node, to the angle of each memory node
Color information is identified；
If not, judging whether itself is election participant, if election participant, initiate to elect in caretaker manager
Afterwards, vote information is sent to caretaker manager.
Specifically, it is serviced by zookeeper, a caretaker manager can be generated, which can send out
Play election.The caretaker manager can specify other memory nodes as election participant, specifically, it is possible to specify in addition to itself
All memory nodes for election participant, can also be arbitrarily designated odd number memory node for election participant, do not do specifically
It limits.
Vote information is sent to caretaker manager by the memory node for being designated as election participant.Caretaker manager according to
The vote information that each election participant received sends, determines that the Role Information of each memory node (namely determines management
Person and non-management person).Caretaker manager utilizes the zookeeper run in each memory node to service, to the angle of each memory node
Color information is identified.In this way, each memory node can service after election according to the zookeeper of self-operating
The Role Information of middle mark judges whether self-role information is manager.
S202：Judge itself whether be configured with the address information for accessing the storage system, if not, performing S203.
S203：It obtains and described address information is configured.
In the present embodiment, if memory node judges self-role information not for manager, which may determine that
The address information for accessing storage system itself whether is configured with, if configured, deletes the address information.
The address information can be understood as virtual ip address, that is, the virtual ip address that storage system externally provides, and use
Family can access the storage system by the virtual ip address.Only have manager that the virtual ip address is configured in storage system, because
This, if former manager is broken down, after certain memory node becomes new management person, new management person should be configured with address letter
Breath.Specifically, the address information can be directly acquired by order line.
In addition, if non-management person is configured with the address information, which will appear unknown error, therefore, when depositing
Storage node judges self-role information not for manager and when being configured with the address information, which should believe the address
Breath is deleted from itself.
Using embodiment illustrated in fig. 2 of the present invention, after breaking down as the memory node of manager, saved using each storage
Election mechanism under the management service run in point, each memory node are likely to become new manager, and new manager matches
The address information for accessing the storage system is put, which can continue to provide storage service；This improves storage systems
Reliability.
As one embodiment of the present invention, database can be stored in each memory node, memory node passes through certainly
The management service of body operation after receiving data modification information, can update the database of itself storage.
Specifically, if data change occurs for the memory node in storage system, which can be by the data
Altering event notifies each memory node by the zookeeper event notification mechanisms serviced.Each memory node passes through self-operating
Zookeeper services can receive data modification information.
Data change the change that can include index information, can also include the change of nodal information.For example, when depositing
In storage system when increase, deletion or mobile storage data, the index information of data changes；When certain in storage system stores
When node increases newly, deletes child node, nodal information changes.Index information can be not only included in above-mentioned database, but also is included
Nodal information.
As a kind of embodiment, the index information of change data can be included in the data modification information, in this way, can be with
According to the index information of change data included in data modification information, the index information in library is updated the data.
As a kind of embodiment, the nodal information of change memory node can also be included in the data modification information, this
Sample can update the data the nodal information in library according to the nodal information of change memory node included in data modification information.
As a kind of embodiment, index information can not be included in the data modification information, also not comprising nodal information,
Only play suggesting effect.Memory node reads the index of each memory node storage data after the data modification information is received
The nodal information of information and each memory node；Index information in the index information and database that are read is compared,
Obtain the first comparing result；Nodal information in the nodal information and database that are read is compared, obtains second pair
Compare result；According to the first comparing result and the second comparing result, library is updated the data.
For the convenience of description, index information is referred to as recording by content with nodal information below.
Specifically, the record in every memory node can be directed to, by its (being assumed to be A) and the record B of own database
It is compared, is not present in B if there is some contents A1 exists only in A, then A1 can be inserted into own database.
A is not present in, and in the child node of the corresponding memory nodes of B1 not if there is some contents B1 exists only in B
There are B1, then can delete B1 from local data base.
A is not present in, and there is no son sections under the corresponding memory nodes of B2 if there is some contents B2 exists only in B
Point, it may be considered that loss of data in A, in this case, record that can be to B2 in corresponding memory node are updated.
In the present embodiment, the corresponding first version number of every index information and every can also be stored in database
The corresponding second edition number of nodal information, therefore, can update the rope in the database according to first comparing result
Fuse ceases and its corresponding first version number；According to second comparing result, update nodal information in the database and
Its corresponding second edition number.
Specifically, every record is updated once, corresponding version number can add 1.
Due to storing the version number of every record, what the record in above-mentioned memory node corresponding to B2 was updated
In the process, it in the record in each memory node, can determine the record of version number's maximum, that is, determine newest note
Record, according to newest record, the record to B2 in corresponding memory node is updated.In addition, other memory nodes can also
According to the newest record update own database.
It, can also be according to respectively being deposited in storage system it should be noted that if memory node is reached the standard grade again after breaking down
The database of node is stored up, own database is updated.
It should be noted that herein, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any this practical relationship or sequence.Moreover, term " comprising ", "comprising" or its any other variant are intended to
Non-exclusive inclusion, so that process, method, article or equipment including a series of elements not only will including those
Element, but also including other elements that are not explicitly listed or further include as this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
Also there are other identical elements in process, method, article or equipment including the element.
Each embodiment in this specification is described using relevant mode, identical similar portion between each embodiment
Point just to refer each other, and the highlights of each of the examples are difference from other examples.Especially for device reality
For applying example, since it is substantially similar to embodiment of the method, so description is fairly simple, related part is referring to embodiment of the method
Part explanation.
One of ordinary skill in the art will appreciate that all or part of step in realization above method embodiment is can
It is completed with instructing relevant hardware by program, the program can be stored in computer read/write memory medium,
The storage medium designated herein obtained, such as：ROM/RAM, magnetic disc, CD etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent replacement, improvement and so within the spirit and principles in the present invention, are all contained in protection scope of the present invention
It is interior.
Claims (16)
1. a kind of storage system, which is characterized in that including at least two memory nodes, in every memory node operational management take
Business；It is the first memory node of manager comprising Role Information at least two memory nodes and Role Information is not pipe
The second memory node of reason person；
Second memory node, after detecting that first memory node breaks down, by being transported in each memory node
Election mechanism under capable management service judges whether self-role information is manager；If so, judging whether itself is configured
There is the address information for accessing the storage system, if be not configured, obtain and described address information is configured.
2. system according to claim 1, which is characterized in that
Second memory node is additionally operable to through the election mechanism, judges whether itself is caretaker manager：
If so, initiating election, according to the vote information that each election participant sends, determine respectively to store in the storage system
The Role Information of node, and using the management service run in each memory node, the role of each memory node is believed
Breath is identified, so that each memory node determines self-role information according to the mark；
If not, judge whether itself is election participant, and if election participant, after caretaker manager initiates election,
Vote information is sent to caretaker manager.
3. system according to claim 1, which is characterized in that
Second memory node is additionally operable to, in the case where judging self-role information not for manager, whether judge itself
The address information for accessing the storage system is configured with, if configured, deletes described address information.
4. system according to claim 2, which is characterized in that
Second memory node is additionally operable to the management service by self-operating, receives fault cues information, and the failure carries
Show that information represents that first memory node breaks down；
Second memory node, after being additionally operable to election, according to the Role Information identified in the management service of self-operating,
Judge self-role information whether manager.
5. system according to claim 1, which is characterized in that
Second memory node is additionally operable to the management service by self-operating, receives data modification information；According to the number
According to modification information, the database of itself storage is updated.
6. system according to claim 5, which is characterized in that be stored in the storage system and store in the database
The nodal information of each memory node in the index information of data and the storage system；
Second memory node is additionally operable to according to the index information of change data included in the data modification information, more
Index information in the new database；According to included in the data modification information change memory node nodal information,
Update the nodal information in the database.
7. system according to claim 5, which is characterized in that be stored in the storage system and store in the database
The nodal information of each memory node in the index information of data and the storage system；
Second memory node is additionally operable to read the index information of each memory node storage data and each storage
The nodal information of node；The index information read and the index information in the database are compared, obtain first
Comparing result；The nodal information read and the nodal information in the database are compared, obtain the second comparison knot
Fruit；According to first comparing result and second comparing result, the database is updated.
8. system according to claim 7, which is characterized in that every index information is also stored in the database and is corresponded to
First version number and the corresponding second edition number of every nodal information；
Second memory node, the index information for being additionally operable to, according to first comparing result, update in the database and
Its corresponding first version number；According to second comparing result, nodal information in the database and its corresponding is updated
The second edition number.
9. a kind of storage node management method, which is characterized in that applied to the memory node in storage system, the storage system
Each memory node in operational management service；The method includes：
After detecting the memory node failure that Role Information is manager, pass through the management service run in each memory node
Under election mechanism, judge whether self-role information is manager；
If so, judging itself whether be configured with the address information for accessing the storage system, if be not configured, obtain and match
Put described address information.
10. according to the method described in claim 9, it is characterized in that, detect that the memory node that Role Information is manager goes out
After existing failure, the method further includes：
By the election mechanism, judge whether itself is caretaker manager：
If so, initiating election, according to the vote information that each election participant sends, determine respectively to store in the storage system
The Role Information of node, and using the management service run in each memory node, the role of each memory node is believed
Breath is identified；
If not, judge whether itself is election participant, and if election participant, after caretaker manager initiates election,
Vote information is sent to caretaker manager.
11. according to the method described in claim 9, it is characterized in that, judging self-role information not for the situation of manager
Under, the method further includes：
Judge itself whether be configured with the address information for accessing the storage system, if configured, delete described address letter
Breath.
12. according to the method described in claim 10, it is characterized in that, detect that the memory node that Role Information is manager goes out
The step of existing failure, includes：
By the management service of self-operating, fault cues information is received, the fault cues information represents Role Information for pipe
The memory node of reason person breaks down；
The election mechanism by under the management service that is run in each memory node, judge self-role information whether manager
The step of include：
After election, according to the Role Information identified in the management service of self-operating, judge whether self-role information manages
Reason person.
13. according to the method described in claim 9, it is characterized in that, the method further includes：
By the management service of self-operating, data modification information is received；
According to the data modification information, the database of itself storage is updated.
14. according to the method for claim 13, which is characterized in that be stored in the storage system and deposit in the database
The nodal information of each memory node in the index information of storage data and the storage system；
In the case of index information comprising change data in the data modification information, the data of itself storage of the update
The step of library, includes：
According to the index information of change data included in the data modification information, the index letter in the database is updated
Breath；
In the case of nodal information comprising change memory node in the data modification information, itself storage of the update
The step of database, includes：
According to the nodal information of change memory node included in the data modification information, the node in the database is updated
Information.
15. according to the method for claim 13, which is characterized in that be stored in the storage system and deposit in the database
The nodal information of each memory node in the index information of storage data and the storage system；The data of itself storage of the update
The step of library, includes：
Read the index information of each memory node storage data and the nodal information of each memory node；
The index information read and the index information in the database are compared, obtain the first comparing result；
The nodal information read and the nodal information in the database are compared, obtain the second comparing result；
According to first comparing result and second comparing result, the database is updated.
16. according to the method for claim 15, which is characterized in that every index information pair is also stored in the database
The first version number answered and the corresponding second edition number of every nodal information；It is described according to first comparing result and described
Second comparing result, the step of updating the database, include：
According to first comparing result, the index information in the database and its corresponding first version number are updated；
According to second comparing result, the nodal information in the database and its corresponding second edition number are updated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611109234.5A CN108156203B (en) | 2016-12-06 | 2016-12-06 | Storage system and storage node management method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611109234.5A CN108156203B (en) | 2016-12-06 | 2016-12-06 | Storage system and storage node management method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108156203A true CN108156203A (en) | 2018-06-12 |
| CN108156203B CN108156203B (en) | 2021-01-08 |
Family
ID=62467661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611109234.5A Active CN108156203B (en) | 2016-12-06 | 2016-12-06 | Storage system and storage node management method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108156203B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109167660A (en) * | 2018-09-07 | 2019-01-08 | 腾讯科技（深圳）有限公司 | Elect a deputy node device method, apparatus, computer equipment and storage medium |
| CN109189854A (en) * | 2018-08-14 | 2019-01-11 | 新华三技术有限公司成都分公司 | The method and node device of sustained traffic are provided |
| US12052344B2 (en) | 2018-09-07 | 2024-07-30 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for electing representative node device, computer device, and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102929696A (en) * | 2012-09-28 | 2013-02-13 | 北京搜狐新媒体信息技术有限公司 | Method and apparatus for constructing, submitting and monitoring center node of distributed system |
| US20130117225A1 (en) * | 2011-11-03 | 2013-05-09 | Michael W. Dalton | Distributed storage medium management for heterogeneous storage media in high availability clusters |
| CN103118130A (en) * | 2013-02-25 | 2013-05-22 | 汉柏科技有限公司 | Cluster management method and cluster management system for distributed service |
| CN104753994A (en) * | 2013-12-27 | 2015-07-01 | 杭州海康威视***技术有限公司 | Method and device for data synchronization based on cluster server system |
-
2016
- 2016-12-06 CN CN201611109234.5A patent/CN108156203B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130117225A1 (en) * | 2011-11-03 | 2013-05-09 | Michael W. Dalton | Distributed storage medium management for heterogeneous storage media in high availability clusters |
| CN102929696A (en) * | 2012-09-28 | 2013-02-13 | 北京搜狐新媒体信息技术有限公司 | Method and apparatus for constructing, submitting and monitoring center node of distributed system |
| CN103118130A (en) * | 2013-02-25 | 2013-05-22 | 汉柏科技有限公司 | Cluster management method and cluster management system for distributed service |
| CN104753994A (en) * | 2013-12-27 | 2015-07-01 | 杭州海康威视***技术有限公司 | Method and device for data synchronization based on cluster server system |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109189854A (en) * | 2018-08-14 | 2019-01-11 | 新华三技术有限公司成都分公司 | The method and node device of sustained traffic are provided |
| CN109189854B (en) * | 2018-08-14 | 2021-06-08 | 新华三技术有限公司成都分公司 | Method and node equipment for providing continuous service |
| CN109167660A (en) * | 2018-09-07 | 2019-01-08 | 腾讯科技（深圳）有限公司 | Elect a deputy node device method, apparatus, computer equipment and storage medium |
| WO2020048427A1 (en) * | 2018-09-07 | 2020-03-12 | 腾讯科技（深圳）有限公司 | Method and apparatus for electing representative node device, and computer device and storage medium |
| KR20210005678A (en) * | 2018-09-07 | 2021-01-14 | 텐센트 테크놀로지(센젠) 컴퍼니 리미티드 | Method and apparatus for electing a representative node device, and computer device and storage medium |
| CN109167660B (en) * | 2018-09-07 | 2021-05-14 | 腾讯科技（深圳）有限公司 | Method and device for electing representative node equipment, computer equipment and storage medium |
| JP2021526676A (en) * | 2018-09-07 | 2021-10-07 | テンセント・テクノロジー・（シェンジェン）・カンパニー・リミテッド | Selection method of representative node equipment and its equipment, computer equipment and computer programs |
| KR102339194B1 (en) | 2018-09-07 | 2021-12-14 | 텐센트 테크놀로지(센젠) 컴퍼니 리미티드 | Method and apparatus for electing a representative node device, and computer device and storage medium |
| JP7112518B2 (en) | 2018-09-07 | 2022-08-03 | テンセント・テクノロジー・（シェンジェン）・カンパニー・リミテッド | Method of selecting representative node device and its device, computer device and computer program |
| US12052344B2 (en) | 2018-09-07 | 2024-07-30 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for electing representative node device, computer device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108156203B (en) | 2021-01-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8775387B2 (en) | Methods and systems for validating accessibility and currency of replicated data | |
| US20020112095A1 (en) | Method and apparatus for authorizing and reporting changes to device configurations | |
| US20060010337A1 (en) | Management system and management method | |
| US20060236053A1 (en) | Memory device system, storage device, and log recording method | |
| JP2005276177A (en) | Method, system, and program for checking and repairing network configuration | |
| US10841159B2 (en) | Robust service deployment | |
| CN104104582B (en) | A kind of data storage path management method, client and server | |
| CN108156203A (en) | A kind of storage system and storage node management method | |
| CN107885859A (en) | A kind of method, apparatus and computer-readable recording medium of file number quota | |
| CN112162761A (en) | Method, system and equipment for automatically deploying project to public cloud containerization platform | |
| CN106021027A (en) | Terminal data processing method and system | |
| JP6033189B2 (en) | COMMUNICATION DEVICE, COMMUNICATION METHOD, AND COMMUNICATION PROGRAM | |
| CN112748856A (en) | Method of managing disc, electronic device, and computer program product | |
| CN106470153B (en) | method and device for preventing temporary loop | |
| CN108459927A (en) | A kind of data back up method, device and server | |
| CN107291575B (en) | Processing method and equipment for data center fault | |
| JP5159777B2 (en) | Apparatus and method for managing and storing information and metadata | |
| CN104750869A (en) | File management method and device | |
| CN108133034A (en) | Shared storage access method and relevant apparatus | |
| US9973421B2 (en) | Method and device for connection management in a telecommunication network | |
| US20190065065A1 (en) | Data protection method and storage server | |
| CN106657227B (en) | The method and device of resource is subscribed in a kind of change | |
| CN109144766A (en) | A kind of storage of data, reconstructing method and device and electronic equipment | |
| CN108733753B (en) | File reading method and application entity | |
| US20090024880A1 (en) | System and method for triggering control over abnormal program termination |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |