CN109271367A - Distributed file system multinode snapshot rollback method and system - Google Patents
Distributed file system multinode snapshot rollback method and system Download PDFInfo
- Publication number
- CN109271367A CN109271367A CN201811496247.1A CN201811496247A CN109271367A CN 109271367 A CN109271367 A CN 109271367A CN 201811496247 A CN201811496247 A CN 201811496247A CN 109271367 A CN109271367 A CN 109271367A
- Authority
- CN
- China
- Prior art keywords
- rollback task
- snapshot
- rollback
- information
- server
- 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.)
- Pending
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a kind of distributed file system multinode snapshot rollback method and systems.Wherein, method includes that master metadata server receives respectively after the snapshot rollback task processing information that meta data server is sent, and after executing corresponding operating to global rollback task queue according to snapshot rollback task processing information, is updated to object storage server；Then to the information for updating rollback task queue is respectively sent from meta data server, more new information respectively is read from meta data server according to from object storage server, and be updated to the local rollback task queue in respective caching.The application is managed collectively the update of snapshot rollback task all in distributed file system using master metadata server, it is newest for realizing the rollback task of the rollback task queue in all meta data server local caches, and it is consistent, improves manageability, maintainability of the storage system to snapshot rollback task.
Description
Technical field
The present embodiments relate to distributed file system technology fields, more more particularly to a kind of distributed file system
Node snapshot rollback method and system.
Background technique
With cloud computing, the fast development of big data, explosive growth is presented in data, merely by increase hard disk number come
The mode for extending the memory capacity of computer file system, can no longer meet the data that information explosion epoch index increases again
Storage demand, distributed file system are applied and are given birth to.
Distributed file system (Distributed File System) is the file based on customer end/server mode
The physical memory resources of system administration are connected by computer network with node, and the storage and management of data can be effectively solved
Problem.It will be fixed on some file system in some place, expands to any number of places/multiple file system, numerous sections
Point one Filesystem Network of composition.Each node can be distributed in different places, carry out the communication between node by network
It is transmitted with data.User when using distributed file system, without be concerned about data be stored on which node or be from
Which node is from acquisition, it is only necessary to managed as using local file system and storage file system in data.
As storage NAS products application is more more and more universal, file system snapshot is also increasingly used widely.File system
Immediately copy contains file system and generates in snapshot snapshot, that is, file system of uniting one for a complete available copy
All information at moment.Distributed file system includes the metadata cluster being made of multiple meta data servers, is deployed in more
On a different node.When multiple client accesses different Metadata Services, while the snapshot in one file system of initiation
When rolling back action, or when a rollback does not complete also, another Metadata Service receives other rollback tasks, is easy
It will appear the inconsistent phenomenon of each node snapshot rollback.
Summary of the invention
The embodiment of the present disclosure provides a kind of distributed file system multinode snapshot rollback method and system, realizes more
The consistency of node snapshot rollback.
In order to solve the above technical problems, the embodiment of the present invention the following technical schemes are provided:
On the one hand the embodiment of the present invention provides a kind of distributed file system multinode snapshot rollback method, distributed text
Part system includes master metadata server and multiple from meta data server, is based on master metadata server, comprising:
Receive the snapshot rollback task processing information respectively sent from meta data server；
Corresponding operating is carried out to the global rollback task queue constructed in advance according to the snapshot rollback task processing information,
And the global rollback task queue in upgating object storage server；
To respectively from meta data server send update rollback task queue information so that respectively from meta data server according to
Information is read from the object storage server to be updated the local rollback task queue in respective caching.
Optionally, the snapshot rollback task processing information is the snapshot rollback task requests of client, described according to institute
State snapshot rollback task processing information includes: to the global rollback task queue progress corresponding operating constructed in advance
The snapshot rollback task requests are added to the global rollback task queue.
Optionally, the snapshot rollback task processing information is snapshot rollback task completed information, described according to described fast
Carrying out corresponding operating to the global rollback task queue constructed in advance according to rollback task processing information includes:
By the corresponding snapshot rollback task of the snapshot rollback task completed information, from the global rollback task queue
It deletes.
Optionally, it is described to respectively from meta data server send update rollback task queue information after, further includes:
Completion is updated when receiving all local caches fed back from meta data server in the distributed file system
After information, to the information for respectively sending update completion from meta data server.
The embodiment of the invention also provides a kind of distributed file system multinode snapshot rollback method, distributed field systems
System includes master metadata server and multiple from meta data server, based on from meta data server, comprising:
Snapshot rollback task processing information is sent to the master metadata server；
When the information for receiving the update rollback task queue that the master metadata server is sent, from object storage service
Device reads more new information；
Local rollback task queue is updated according to the more new information；
Wherein, the object storage server stores the master metadata server and is handled according to the snapshot rollback task
Global rollback task queue after information update.
Optionally, described to include: to master metadata server transmission snapshot rollback task processing information
When the snapshot rollback task requests for receiving client, snapshot rollback is constructed according to the snapshot rollback task requests
Task queue update message；
The snapshot rollback task queue update message is sent to the master metadata server.
Optionally, described to include: to master metadata server transmission snapshot rollback task processing information
Judge whether to complete the first snapshot rollback task；
If so, sending the information that the first snapshot rollback task is completed to the master metadata server.
Optionally, further includes:
After restarting successfully, is read from the object storage server and do not complete rollback task；
Each unfinished rollback task is updated in the local rollback task queue in local cache.
Optionally, after described, the more new information according to is updated local rollback task queue, further includes:
After the update is complete, local cache is sent to the master metadata server update the information completed.
On the other hand the embodiment of the present invention provides a kind of distributed file system multinode snapshot rollback system, comprising:
Including master metadata server and multiple from meta data server:
The master metadata server is used to receive the snapshot rollback task processing information respectively sent from meta data server；
Corresponding operating, and update pair are carried out to the global rollback task queue constructed in advance according to the snapshot rollback task processing information
As the global rollback task queue in storage server；To the letter for respectively sending update rollback task queue from meta data server
Breath, so as to respectively read information to the local rollback in respective caching from meta data server according to from the object storage server
Task queue is updated；
It is respectively used to send snapshot rollback task processing information to the master metadata server from meta data server；When connecing
The information for receiving the update rollback task queue that the master metadata server is sent, reads from object storage server and updates letter
Breath；Local rollback task queue is updated according to the more new information.
The advantages of technical solution provided by the present application, is, is managed collectively distributed field system using master metadata server
The update of all snapshot rollback tasks in system, master metadata server update after the completion of, notice respectively from meta data server into
Row is corresponding to update operation, so that the rollback task in the rollback task queue in all meta data server local caches is
It is newest, and be consistent, so that the snapshot that all meta data servers can receive that client is sent in storage system returns
Rolling task, and all rollback tasks that processing client is initiated can be received simultaneously, to improve storage system to snapshot rollback
Manageability, the maintainability of task.
In addition, the embodiment of the present invention provides corresponding system also directed to distributed file system multinode snapshot rollback method
System, further such that the method has more practicability, the system has the advantages that corresponding.
It should be understood that the above general description and the following detailed description are merely exemplary, this can not be limited
It is open.
Detailed description of the invention
It, below will be to embodiment or correlation for the clearer technical solution for illustrating the embodiment of the present invention or the relevant technologies
Attached drawing needed in technical description is briefly described, it should be apparent that, the accompanying drawings in the following description is only this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of interaction flow of distributed file system multinode snapshot rollback method provided in an embodiment of the present invention
Schematic diagram；
Fig. 2 is the interactive stream of another distributed file system multinode snapshot rollback method provided in an embodiment of the present invention
Journey schematic diagram；
Fig. 3 is the interactive stream of another distributed file system multinode snapshot rollback method provided in an embodiment of the present invention
Journey schematic diagram；
Fig. 4 is that the embodiment of the present invention is returned according to the distributed file system multinode snapshot shown in the first exemplary embodiment
Roll the structural schematic diagram of device；
Fig. 5 is that the embodiment of the present invention is returned according to the distributed file system multinode snapshot shown in the second exemplary embodiment
Roll the structural schematic diagram of device；
Fig. 6 is that the embodiment of the present invention is returned according to the distributed file system multinode snapshot shown in third exemplary embodiment
Roll the structural schematic diagram of device；
Fig. 7 is that the embodiment of the present invention is returned according to the distributed file system multinode snapshot shown in the 4th exemplary embodiment
Roll the structural schematic diagram of device.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.Obviously, described embodiments are only a part of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall within the protection scope of the present invention.
The description and claims of this application and term " first ", " second ", " third " " in above-mentioned attached drawing
Four " etc. be for distinguishing different objects, rather than for describing specific sequence.Furthermore term " includes " and " having " and
Their any deformations, it is intended that cover and non-exclusive include.Such as contain a series of steps or units process, method,
System, product or equipment are not limited to listed step or unit, but may include the step of not listing or unit.
After describing the technical solution of the embodiment of the present invention, the various non-limiting realities of detailed description below the application
Apply mode.
Referring first to Fig. 1, Fig. 1 is a kind of distributed file system multinode snapshot rollback side provided in an embodiment of the present invention
The interaction flow schematic diagram of method, the embodiment of the present invention may include the following contents:
S101: snapshot rollback task processing information is sent from meta data server to master metadata server.
Distributed file system includes metadata server cluster, and a meta data server conduct is chosen from the cluster
Master metadata server, other each meta data servers are used as from meta data server.Those skilled in the art can be according to reality
Situation carries out selection master metadata server, can also randomly choose any one, this does not influence the realization of the application.
Snapshot rollback task processing information can be the snapshot rollback task sent from the received client of meta data server
Request, can also be snapshot rollback task completed information.
From meta data server in the snapshot rollback task requests for receiving client, first by snapshot rollback task requests
It is configured to snapshot rollback task queue update message and is sent to master metadata server.
From meta data server after completing snapshot rollback task, the snapshot rollback task of completion is configured to snapshot rollback
Task queue update message is sent to master metadata server.
From meta data server when constructing snapshot rollback task queue update message, it can increase for different types different
Identification information so that master metadata server is according to the identification information identification types.
S102: master metadata server receives the snapshot rollback task processing information respectively sent from meta data server, root
Corresponding operating is carried out to the global rollback task queue constructed in advance according to snapshot rollback task processing information, and upgating object stores
Global rollback task queue in server.
Snapshot rollback task processing information includes increasing by two kinds of snapshot rollback task and snapshot rollback task completion, pivot number
It is executed according to server according to the type of the snapshot rollback task processing information received and increases and delete two kinds of operations.
Each meta data server corresponds to a snapshot rollback task queue, stores snapshot rollback task.Pivot number
Snapshot rollback task queue is corresponded to as global rollback task queue according to server, and is written into object storage server.Respectively
It is local rollback task queue from the corresponding snapshot rollback task queue of meta data server, is stored in its local cache.From
Meta data server reads snapshot rollback task from object storage server, and by the information update read to local cache
In rollback task queue in, be consistent itself and global rollback task queue.
It is the snapshot rollback task requests of client, pivot number from meta data server to master metadata server transmission
After receiving information according to server, rollback task queue is read through from object storage server first, to guarantee that object stores
Rollback task queue in server be it is newest, new snapshot rollback task requests are added to global rollback task queue
In, and object storage server is written, with the rollback task in upgating object storage server.
What is sent from meta data server is to complete snapshot rollback task (such as the first snapshot rollback task is completed), pivot
After data server receives information, rollback task queue is read through from object storage server first, to guarantee that object is deposited
Store up server in rollback task queue be it is newest, by the snapshot rollback task completed from global rollback task queue
It deletes, and the rollback task in upgating object storage server.
S103: master metadata server updates the information of rollback task queue to respectively sending from meta data server.
After master metadata server updates the rollback task queue for completing itself, and guarantee in object storage server
Rollback task also updates completion, then to the information for respectively broadcasting update rollback task queue from meta data server.
S104: receiving the information for the update rollback task queue that master metadata server is sent from meta data server,
More new information is read from object storage server.
After the broadcast message for respectively receiving master metadata server from meta data server, from object storage server
Read the information updated.
S105: local rollback task queue is updated from meta data server according to more new information.
In technical solution provided in an embodiment of the present invention, distributed field system is managed collectively using master metadata server
The update of all snapshot rollback tasks in system, master metadata server update after the completion of, notice respectively from meta data server into
Row is corresponding to update operation, so that the rollback task in the rollback task queue in all meta data server local caches is
It is newest, and be consistent, so that the snapshot that all meta data servers can receive that client is sent in storage system returns
Rolling task, and all rollback tasks that processing client is initiated can be received simultaneously, to improve storage system to snapshot rollback
Manageability, the maintainability of task.
Optionally, respectively from meta data server to the snapshot rollback task of the local rollback task queue in local cache more
After the completion of new, it can will update the information completed and be sent to master metadata server.Master metadata server owns in receiving
After the local cache fed back from meta data server updates completion information, to the letter respectively completed from meta data server transmission update
Breath.
Follow-up advantageously ensures that all meta data server local caches respectively from the update progress of meta data server in time
In rollback task queue in rollback task be newest, and be consistent.
For any one of distributed file system meta data server, when some meta data server causes extremely
The meta data server is restarted, and after the meta data server is restarted successfully, can directly be read from object storage server not complete
At rollback task, then each unfinished rollback task is updated in the local rollback task queue in local cache, thus real
Now the rollback task in the rollback task queue in all meta data server local caches is newest, and is consistent.
Present invention also provides one embodiment, refer to Fig. 2, and Fig. 2 is another distribution provided in an embodiment of the present invention
The interaction flow schematic diagram of formula file system multinode snapshot rollback method, the embodiment of the present invention for example can be applied to distributed text
Part system, it may include the following contents:
S201: the snapshot rollback task requests of client are received from meta data server, are asked according to snapshot rollback task
Seek construction snapshot rollback task queue update message.
S202: master metadata server is sent to from meta data server by snapshot rollback task queue update message.
S203: master metadata server receives snapshot rollback task queue update message, by snapshot rollback task requests
It is added to global rollback task queue, and is written into object storage server.
S204: master metadata server updates the information of rollback task queue to respectively sending from meta data server.
S205: from meta data server according to from object storage server according to reading information to the local in respective caching
Rollback task queue is updated.
S206: after the update is complete from meta data server, local cache is sent to master metadata server and has been updated
At information.
S207: master metadata server all sheets fed back from meta data server in receiving distributed file system
After ground buffer update completes information, to the information for respectively sending update completion from meta data server.
The realization process of method or step same as the previously described embodiments of the embodiment of the present invention, sees above-described embodiment
Description, herein, just repeats no more.
From the foregoing, it will be observed that the embodiment of the present invention realizes the consistency of multinode snapshot rollback.
Present invention also provides another embodiments, refer to Fig. 3, and Fig. 3 is another kind provided in an embodiment of the present invention
The interaction flow schematic diagram of distributed file system multinode snapshot rollback method, the embodiment of the present invention for example can be applied to be distributed
Formula file system, it may include the following contents:
S301: completing snapshot rollback task from meta data server, and the snapshot rollback task of completion is configured to snapshot and is returned
Roll task queue update message.
S302: master metadata server is sent to from meta data server by snapshot rollback task queue update message.
S303: master metadata server receives snapshot rollback task queue update message, from global rollback task queue
It is middle to delete completed snapshot rollback task, and upgating object storage server.
S304: master metadata server updates the information of rollback task queue to respectively sending from meta data server.
S305: information is read to the local rollback in respective caching from meta data server according to from object storage server
Task queue is updated.
S306: after the update is complete from meta data server, local cache is sent to master metadata server and has been updated
At information.
S307: master metadata server all sheets fed back from meta data server in receiving distributed file system
After ground buffer update completes information, to the information for respectively sending update completion from meta data server.
The realization process of method or step same as the previously described embodiments of the embodiment of the present invention, sees above-described embodiment
Description, herein, just repeats no more.
From the foregoing, it will be observed that the embodiment of the present invention realizes the consistency of multinode snapshot rollback.
The embodiment of the present invention provides corresponding realize also directed to distributed file system multinode snapshot rollback method and fills
It sets, further such that the method has more practicability.Below to distributed file system more piece provided in an embodiment of the present invention
Point snapshot rollback device be introduced, distributed file system multinode snapshot rollback device described below with it is above-described
Distributed file system multinode snapshot rollback method can correspond to each other reference.
Fig. 4 is referred to, Fig. 4 is distributed file system multinode snapshot rollback device provided in an embodiment of the present invention one
Structure chart under kind specific embodiment, is based on master metadata server, the device can include:
Information receiving module 401, for receiving the snapshot rollback task processing information respectively sent from meta data server.
Snapshot rollback task update module 402, the global rollback according to snapshot rollback task processing information to constructing in advance
Task queue carries out corresponding operating, and the global rollback task queue in upgating object storage server.
Broadcast update information module 403, for respectively from meta data server send update rollback task queue information,
So as to respectively read information to the local rollback task team in respective caching from meta data server according to from object storage server
Column are updated.
In one embodiment, the snapshot rollback task update module 402 can receive to work as from Metadata Service
The snapshot rollback task requests for the client that device is sent, snapshot rollback task requests are added to the mould of global rollback task queue
Block.
In another embodiment, the snapshot rollback task update module 402 can take to work as to receive from metadata
It is engaged in the snapshot rollback task completed information that device sends, by the corresponding snapshot rollback task of snapshot rollback task completed information, from complete
The module deleted in office's rollback task queue.
In addition, referring to Fig. 5, described device for example further includes that broadcast updates completion information module 404, for that ought receive
After all local caches fed back from meta data server update completion information in distributed file system, to respectively from metadata clothes
Business device sends the information for updating and completing.
Fig. 6 is referred to, Fig. 6 is that distributed file system multinode snapshot rollback device provided in an embodiment of the present invention is another
Structure chart under kind specific embodiment is based on from meta data server, the device can include:
Information sending module 601, for sending snapshot rollback task processing information to master metadata server.
Information reading module 602, for the letter when the update rollback task queue for receiving master metadata server transmission
Breath reads more new information from object storage server；Object storage server stores master metadata server according to snapshot rollback
The updated global rollback task queue of task processing information.
Queue update module 603, for being updated according to more new information to local rollback task queue.
Optionally, the information sending module 601 can be to work as the snapshot rollback task requests for receiving client, according to fast
Snapshot rollback task queue update message is constructed according to rollback task requests；Snapshot rollback task queue update message is sent to master
The module of meta data server.
Optionally, the information sending module 601 can also be after the completion of detecting the first snapshot rollback task, to pivot
Data server sends the module for the information that the first snapshot rollback task is completed.
In another embodiment, referring to Fig. 7, described device for example may include restarting information read module 604,
For being read from object storage server and not completing rollback task after restarting successfully；Each unfinished rollback task is updated to
In local rollback task queue in local cache.
Described device for example may also include information feedback module 605, be used for after the update is complete, to main metadata service
Device sends local cache and updates the information completed.
The function of each functional module of distributed file system multinode snapshot rollback device can described in the embodiment of the present invention
According to the method specific implementation in above method embodiment, specific implementation process is referred to the correlation of above method embodiment
Description, details are not described herein again.
From the foregoing, it will be observed that the embodiment of the present invention using master metadata server unified management distributed file system in it is all
The update of snapshot rollback task, the rollback task for realizing the rollback task queue in all meta data server local caches are equal
It to be newest, and is consistent, improves manageability, maintainability of the storage system to snapshot rollback task.
The embodiment of the invention also provides a kind of distributed file system multinode snapshot rollback systems, it may include pivot number
According to server and multiple from meta data server:
Master metadata server is used to receive the snapshot rollback task processing information respectively sent from meta data server；According to
Snapshot rollback task processing information carries out corresponding operating, and upgating object storage clothes to the global rollback task queue constructed in advance
The global rollback task queue being engaged in device；To the information for updating rollback task queue is respectively sent from meta data server, so that respectively
Information is read to the local rollback task queue progress in respective caching from meta data server according to from object storage server
It updates；
The snapshot rollback task processing information for respectively being used to send to master metadata server from meta data server；Work as reception
The information of the update rollback task queue sent to master metadata server, reads more new information from object storage server；Root
Local rollback task queue is updated according to more new information.
The function of each functional module of distributed file system multinode snapshot rollback system can described in the embodiment of the present invention
According to the method specific implementation in above method embodiment, specific implementation process is referred to the correlation of above method embodiment
Description, details are not described herein again.
From the foregoing, it will be observed that the embodiment of the present invention using master metadata server unified management distributed file system in it is all
The update of snapshot rollback task, the rollback task for realizing the rollback task queue in all meta data server local caches are equal
It to be newest, and is consistent, improves manageability, maintainability of the storage system to snapshot rollback task.
The embodiment of the invention also provides a kind of distributed file system multinode snapshot rollback equipment, specifically can include:
Memory, for storing computer program；
Processor realizes distributed file system more piece described in any one embodiment as above for executing computer program
The step of point snapshot rollback method.
The function of each functional module of distributed file system multinode snapshot rollback equipment can described in the embodiment of the present invention
According to the method specific implementation in above method embodiment, specific implementation process is referred to the correlation of above method embodiment
Description, details are not described herein again.
From the foregoing, it will be observed that the embodiment of the present invention using master metadata server unified management distributed file system in it is all
The update of snapshot rollback task, the rollback task for realizing the rollback task queue in all meta data server local caches are equal
It to be newest, and is consistent, improves manageability, maintainability of the storage system to snapshot rollback task.
The embodiment of the invention also provides a kind of computer readable storage mediums, are stored with distributed file system multinode
Snapshot rollback program, as above any one implementation when the distributed file system multinode snapshot rollback program is executed by processor
The step of example distributed file system multinode snapshot rollback method.
The function of each functional module of computer readable storage medium described in the embodiment of the present invention can be according to above method reality
The method specific implementation in example is applied, specific implementation process is referred to the associated description of above method embodiment, herein no longer
It repeats.
From the foregoing, it will be observed that the embodiment of the present invention using master metadata server unified management distributed file system in it is all
The update of snapshot rollback task, the rollback task for realizing the rollback task queue in all meta data server local caches are equal
It to be newest, and is consistent, improves manageability, maintainability of the storage system to snapshot rollback task.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
For setting, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part
Explanation.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond the scope of this invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
A kind of distributed file system multinode snapshot rollback method provided by the present invention and system are carried out above
It is discussed in detail.Used herein a specific example illustrates the principle and implementation of the invention, above embodiments
Illustrate to be merely used to help understand method and its core concept of the invention.It should be pointed out that for the common skill of the art
, without departing from the principle of the present invention, can be with several improvements and modifications are made to the present invention for art personnel, these change
It is also fallen within the protection scope of the claims of the present invention into modification.
Claims (10)
1. a kind of distributed file system multinode snapshot rollback method, which is characterized in that distributed file system includes pivot
Data server and multiple from meta data server, is based on master metadata server, comprising:
Receive the snapshot rollback task processing information respectively sent from meta data server；
Corresponding operating is carried out to the global rollback task queue constructed in advance according to the snapshot rollback task processing information, and more
Global rollback task queue in new object storage server；
To the information for updating rollback task queue is respectively sent from meta data server, so that respectively from meta data server according to from institute
Object storage server reading information is stated to be updated the local rollback task queue in respective caching.
2. distributed file system multinode snapshot rollback method according to claim 1, which is characterized in that the snapshot
Rollback task processing information is the snapshot rollback task requests of client, described according to the snapshot rollback task processing information pair
The global rollback task queue constructed in advance carries out corresponding operating
The snapshot rollback task requests are added to the global rollback task queue.
3. distributed file system multinode snapshot rollback method according to claim 1, which is characterized in that the snapshot
Rollback task processing information be snapshot rollback task completed information, it is described according to the snapshot rollback task processing information to preparatory
The global rollback task queue of building carries out corresponding operating
By the corresponding snapshot rollback task of the snapshot rollback task completed information, deleted from the global rollback task queue
It removes.
4. distributed file system multinode snapshot rollback method according to claim 1 to 3, feature exist
In, it is described to respectively from meta data server send update rollback task queue information after, further includes:
Completion information is updated when receiving all local caches fed back from meta data server in the distributed file system
Afterwards, to the information for respectively sending update completion from meta data server.
5. a kind of distributed file system multinode snapshot rollback method, which is characterized in that distributed file system includes pivot
Data server and multiple from meta data server, based on from meta data server, comprising:
Snapshot rollback task processing information is sent to the master metadata server；
When the information for receiving the update rollback task queue that the master metadata server is sent, read from object storage server
Take more new information；
Local rollback task queue is updated according to the more new information；
Wherein, the object storage server stores the master metadata server according to the snapshot rollback task processing information
Updated overall situation rollback task queue.
6. distributed file system multinode snapshot rollback method according to claim 5, which is characterized in that described to institute
Stating master metadata server transmission snapshot rollback task processing information includes:
When the snapshot rollback task requests for receiving client, snapshot rollback task is constructed according to the snapshot rollback task requests
Queue update message；
The snapshot rollback task queue update message is sent to the master metadata server.
7. distributed file system multinode snapshot rollback method according to claim 5, which is characterized in that described to institute
Stating master metadata server transmission snapshot rollback task processing information includes:
Judge whether to complete the first snapshot rollback task；
If so, sending the information that the first snapshot rollback task is completed to the master metadata server.
8. being existed according to distributed file system multinode snapshot rollback method, feature described in claim 5-7 any one
In, further includes:
After restarting successfully, is read from the object storage server and do not complete rollback task；
Each unfinished rollback task is updated in the local rollback task queue in local cache.
9. distributed file system multinode snapshot rollback method according to claim 8, which is characterized in that at described
After being updated according to the more new information to local rollback task queue, further includes:
After the update is complete, local cache is sent to the master metadata server update the information completed.
10. a kind of distributed file system multinode snapshot rollback system, which is characterized in that including master metadata server and more
It is a from meta data server:
The master metadata server is used to receive the snapshot rollback task processing information respectively sent from meta data server；According to
The snapshot rollback task processing information carries out corresponding operating to the global rollback task queue constructed in advance, and upgating object is deposited
Store up the global rollback task queue in server；To respectively from meta data server send update rollback task queue information, with
Make respectively to read information to the local rollback task in respective caching from meta data server according to from the object storage server
Queue is updated；
It is respectively used to send snapshot rollback task processing information to the master metadata server from meta data server；When receiving
The information for the update rollback task queue that the master metadata server is sent, reads more new information from object storage server；
Local rollback task queue is updated according to the more new information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811496247.1A CN109271367A (en) | 2018-12-07 | 2018-12-07 | Distributed file system multinode snapshot rollback method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811496247.1A CN109271367A (en) | 2018-12-07 | 2018-12-07 | Distributed file system multinode snapshot rollback method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109271367A true CN109271367A (en) | 2019-01-25 |
Family
ID=65187438
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811496247.1A Pending CN109271367A (en) | 2018-12-07 | 2018-12-07 | Distributed file system multinode snapshot rollback method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109271367A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111600949A (en) * | 2020-05-14 | 2020-08-28 | 上海鸿翼软件技术股份有限公司 | Data transmission method, device, equipment and computer readable storage medium |
| CN113051055A (en) * | 2021-03-24 | 2021-06-29 | 北京沃东天骏信息技术有限公司 | Task processing method and device |
| CN115905114A (en) * | 2023-03-09 | 2023-04-04 | 浪潮电子信息产业股份有限公司 | Batch updating method and system of metadata, electronic equipment and readable storage medium |
| CN117009294A (en) * | 2023-08-21 | 2023-11-07 | 北京志凌海纳科技有限公司 | Method for implementing consistency snapshot in distributed file system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7827350B1 (en) * | 2007-04-27 | 2010-11-02 | Netapp, Inc. | Method and system for promoting a snapshot in a distributed file system |
| US20140214773A1 (en) * | 2013-01-30 | 2014-07-31 | Hewlett-Packard Development Company, L.P. | Reconstructing a state of a file system using a preserved snapshot |
| CN104933133A (en) * | 2015-06-12 | 2015-09-23 | 中国科学院计算技术研究所 | Metadata snap storage and access method in distributed file system |
| CN106055431A (en) * | 2016-05-30 | 2016-10-26 | 河海大学 | Data backup and recovery method and data backup and recovery device based on snapshot online rollback |
| CN108255638A (en) * | 2017-06-29 | 2018-07-06 | 新华三技术有限公司 | A kind of snapshot rollback method and device |
-
2018
- 2018-12-07 CN CN201811496247.1A patent/CN109271367A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7827350B1 (en) * | 2007-04-27 | 2010-11-02 | Netapp, Inc. | Method and system for promoting a snapshot in a distributed file system |
| US20140214773A1 (en) * | 2013-01-30 | 2014-07-31 | Hewlett-Packard Development Company, L.P. | Reconstructing a state of a file system using a preserved snapshot |
| CN104933133A (en) * | 2015-06-12 | 2015-09-23 | 中国科学院计算技术研究所 | Metadata snap storage and access method in distributed file system |
| CN106055431A (en) * | 2016-05-30 | 2016-10-26 | 河海大学 | Data backup and recovery method and data backup and recovery device based on snapshot online rollback |
| CN108255638A (en) * | 2017-06-29 | 2018-07-06 | 新华三技术有限公司 | A kind of snapshot rollback method and device |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111600949A (en) * | 2020-05-14 | 2020-08-28 | 上海鸿翼软件技术股份有限公司 | Data transmission method, device, equipment and computer readable storage medium |
| CN111600949B (en) * | 2020-05-14 | 2024-03-15 | 上海鸿翼软件技术股份有限公司 | Data transmission method, device, equipment and computer readable storage medium |
| CN113051055A (en) * | 2021-03-24 | 2021-06-29 | 北京沃东天骏信息技术有限公司 | Task processing method and device |
| CN115905114A (en) * | 2023-03-09 | 2023-04-04 | 浪潮电子信息产业股份有限公司 | Batch updating method and system of metadata, electronic equipment and readable storage medium |
| CN115905114B (en) * | 2023-03-09 | 2023-05-30 | 浪潮电子信息产业股份有限公司 | Batch updating method and system of metadata, electronic equipment and readable storage medium |
| CN117009294A (en) * | 2023-08-21 | 2023-11-07 | 北京志凌海纳科技有限公司 | Method for implementing consistency snapshot in distributed file system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109271367A (en) | Distributed file system multinode snapshot rollback method and system | |
| CN106662983B (en) | The methods, devices and systems of data reconstruction in distributed memory system | |
| CN103294675B (en) | Data-updating method and device in a kind of distributed memory system | |
| KR102025934B1 (en) | Partitioning and moving ranges in a distributed system | |
| CN102831156B (en) | Distributed transaction processing method on cloud computing platform | |
| CN109558215A (en) | Backup method, restoration methods, device and the backup server cluster of virtual machine | |
| CN105025084B (en) | A kind of cloud storage system based on sync agent and mixing storage | |
| US7693882B2 (en) | Replicating data across the nodes in a cluster environment | |
| CN108140009A (en) | B-tree key assignments manager of the distributed freedom formula based on RDMA | |
| JP2021513715A (en) | Distributed database cluster system and data synchronization method | |
| US10324799B2 (en) | Enhanced application write performance | |
| CN103186554B (en) | Distributed data mirror method and storage back end | |
| CN105338078B (en) | Date storage method and device for storage system | |
| CN105897946A (en) | Obtaining method and system of access address | |
| CN104572845B (en) | Document distribution method, device, equipment and system | |
| CN102594849A (en) | Data backup and recovery method and device, virtual machine snapshot deleting and rollback method and device | |
| CN108076090A (en) | Data processing method and storage management system | |
| CN103559319B (en) | The cache synchronization method and equipment of distributed cluster file system | |
| CN105138691B (en) | Analyze the method and system of subscriber traffic | |
| CN109144785A (en) | Method and apparatus for Backup Data | |
| CN109684273A (en) | A kind of snapshot management method, apparatus, equipment and readable storage medium storing program for executing | |
| CN109299122A (en) | A kind of method of data synchronization, equipment and computer can storage mediums | |
| CN109491832A (en) | Data disaster tolerance method and website | |
| CN104662522B (en) | The system and method backed up using the full virtual machine of storage system functionality | |
| JP6405255B2 (en) | COMMUNICATION SYSTEM, QUEUE MANAGEMENT SERVER, AND COMMUNICATION METHOD |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190125 |