WO2021082465A1

WO2021082465A1 - Method for ensuring data consistency and related device

Info

Publication number: WO2021082465A1
Application number: PCT/CN2020/096005
Authority: WO
Inventors: 孟俊才; 徐鹏
Original assignee: 华为技术有限公司
Priority date: 2019-10-31
Filing date: 2020-06-14
Publication date: 2021-05-06
Also published as: CN112749178A

Abstract

A method for ensuring data consistency and a related device. The method comprises: a first node receiving an upgrade message sent by a node management server, the node management server being used to manage a node cluster, and the node cluster comprising the first node; the first node updating tenure management data, wherein the tenure management data comprises a root metadata identifier and a tenure identifier, the root metadata identifier is used to determine root metadata, the root metadata is used to manage the metadata corresponding to the node cluster, and the tenure identifier is used to signify that the first node is upgraded to a master node of the node cluster; and the first node setting the data corresponding to the node cluster to a read-only mode, the data comprising root metadata. The described method can ensure data consistency and prevent data conflicts.

Description

A method and related equipment for ensuring data consistency

Technical field

The invention relates to the technical field of computer distributed storage systems, in particular to a method and related equipment for ensuring data consistency.

Background technique

At present, in the application of distributed database, in order to ensure the uniqueness of data writing, in the entire database cluster, it must be ensured that only one node can write data at the same time, that is, at the same time, it must be ensured that only exists in the database cluster A master node.

The Raft protocol is a distributed consensus protocol that adopts the plan of the elite to lead the overall situation. In the entire cluster, only the master node can process requests sent by clients, and other nodes must forward them to the master node for processing even if they receive requests. The Raft protocol strongly relies on the master node to ensure cluster data consistency.

In the cloud computing environment, distributed locks have a very wide range of usage scenarios. For example, in a distributed system, when different devices access a shared resource, the system often needs a distributed lock to support the mutual exclusion of access to the shared resource to ensure consistency, that is, only one node can hold the lock. The lock preemption can guarantee the uniqueness of the master node.

Then, whether it is the raft protocol or the distributed lock, there are certain flaws. To use the raft protocol to select the master node, you must ensure that the number of surviving nodes in the cluster is greater than half of the total number of nodes, otherwise the master node cannot be selected. Distributed locks are used to strictly ensure data consistency (that is, write operations do not conflict). Add a lock every time it is written, and database performance will suffer severe losses. Therefore, how to ensure data consistency without loss of database performance is an urgent problem to be solved.

Summary of the invention

The embodiment of the invention discloses a method and related equipment for ensuring data consistency, which can ensure data consistency and avoid data conflicts without loss of database performance.

In a first aspect, the present application provides a method for ensuring data consistency, including: a first node receives an upgrade message sent by a node management server, the node management server is used to manage a node cluster, and the node cluster includes the first node A node; the first node updates the tenure management data, the tenure management data includes root metadata identification and tenure identification, the root metadata identification is used to determine root metadata, the root metadata is used to manage the Metadata corresponding to the node cluster, the term identifier is used to indicate that the first node is upgraded to the master node of the node cluster; the first node sets the data corresponding to the node cluster to read-only mode, and The data includes the root metadata.

In the solution provided by this application, the first node upgrades to the master node after receiving the upgrade message sent by the node management server, updates the tenure management data, and sets the data corresponding to the node cluster to read-only mode to ensure that at the same time, Only one node can write data, thereby ensuring data consistency and avoiding data conflicts. In addition, since the entire process does not need to negotiate with other nodes, the performance of the system is guaranteed.

With reference to the first aspect, in a possible implementation of the first aspect, the first node updates the tenure identifier while reading the root metadata identifier.

In the embodiment of the present application, the first node guarantees that reading the root metadata identifier and updating the tenure identifier are atomic, which can prevent other nodes, such as the original master node, from concurrently modifying the root metadata identifier during this process, resulting in data conflicts. Lead to data inconsistencies.

With reference to the first aspect, in a possible implementation of the first aspect, the node cluster further includes a second node, and the second node is used to read and write data corresponding to the node cluster and update the root element Data identification; after the first node updates the tenure management data, the root metadata identification is locked, and the root metadata identification is prohibited from being updated by the second node.

In the embodiment of the present application, before the first node updates the tenure management data, the second node (for example, the original master node) can read and write the data corresponding to the node cluster, and can update the root metadata identification, but the tenure is updated on the first node After managing the data, the root metadata identifier will be locked. Although the second node can continue to read and write data, it is not allowed to modify the root metadata identifier. This ensures that in the subsequent process, only one node will be able to exist at any one time. Write data to ensure data consistency.

With reference to the first aspect, in a possible implementation of the first aspect, the data corresponding to the node cluster includes root metadata, metadata, and user data, the metadata is used to manage the user data, and the User data is data written to the node cluster; after the first node sets the data corresponding to the node cluster to read-only mode, sets the metadata to read-only mode, and finally sets the user data It is read-only mode.

In the embodiment of this application, the first node is set in layers, and the root metadata, metadata, and user data are set to read-only mode in turn to ensure the consistency of the data corresponding to the node cluster and ensure that the first node can accurately Find all user data written to the node cluster.

With reference to the first aspect, in a possible implementation of the first aspect, the first node updates the root metadata identifier and writes data to the node cluster.

In the embodiment of the present application, after the first node is upgraded to become the new master node, it can write user data to the node cluster, and can manage the written user data by updating the root metadata identifier.

In a second aspect, the present application provides a first node, including: a receiving module, configured to receive an upgrade message sent by a node management server, where the node management server is used to manage a node cluster, and the node cluster includes the first node. Node; update module, used to update tenure management data, the tenure management data includes root metadata identification and tenure identification, the root metadata identification is used to determine root metadata, the root metadata is used to manage the node Metadata corresponding to the cluster, the tenure identifier is used to indicate that the first node is upgraded to the master node of the node cluster; a processing module is used to set the data corresponding to the node cluster to a read-only mode, the data Including the root metadata.

With reference to the second aspect, in a possible implementation of the second aspect, the update module is further configured to read the root metadata identifier and update the tenure identifier at the same time.

With reference to the second aspect, in a possible implementation of the second aspect, the node cluster further includes a second node, and the second node is used to read and write data corresponding to the node cluster and update the root element Data identification; after the update module updates the tenure management data, the root metadata identification is prohibited from being updated by the second node.

With reference to the second aspect, in a possible implementation of the second aspect, the data corresponding to the node cluster includes root metadata, metadata, and user data, the metadata is used to manage the user data, and the User data is data written to the node cluster; the processing module is also used to set the root metadata to read-only mode, then set the metadata to read-only mode, and finally set the user data Set to read-only mode.

With reference to the second aspect, in a possible implementation of the second aspect, the processing module is further configured to update the root metadata identifier and write data to the node cluster.

In a third aspect, the present application provides a computing device, the computing device includes a processor and a memory, the memory is configured to store program code, and the processor is configured to call the program code in the memory to execute the above-mentioned first aspect And a method combining any one of the above-mentioned first aspects.

In a fourth aspect, the present application provides a computer-readable storage medium that stores a computer program. When the computer program is executed by a processor, it can implement the above-mentioned first aspect and in combination with the above-mentioned first aspect. The process of the method provided by any implementation method.

In a fifth aspect, the present application provides a computer program product, the computer program product includes instructions, when the computer program is executed by a computer, the computer can execute the first aspect and any one of the first aspects mentioned above. The flow of the method provided by the method.

Description of the drawings

In order to explain the technical solutions of the embodiments of the present invention more clearly, the following will briefly introduce the drawings used in the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present invention. Ordinary technicians can obtain other drawings based on these drawings without creative work.

FIG. 1 is a schematic diagram of a node state switching provided by an embodiment of the present application;

Figure 2 is a schematic diagram of an application scenario provided by an embodiment of the present application;

FIG. 3 is a schematic flowchart of a method for ensuring data consistency provided by an embodiment of the present application;

FIG. 4 is a schematic diagram of a data storage relationship provided by an embodiment of the present application;

FIG. 5 is a schematic structural diagram of a first node provided by an embodiment of the present application;

Fig. 6 is a schematic structural diagram of a computing device provided by an embodiment of the present application.

Detailed ways

The following describes the technical solutions in the embodiments of the present application clearly and completely with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

First of all, some terms and related technologies involved in this application will be explained in conjunction with the accompanying drawings, so as to facilitate the understanding of those skilled in the art.

Cloud database is a stable, reliable, and elastically scalable online database service. This database is deployed in a virtual computing environment and managed through a unified management system, which can effectively reduce maintenance costs. The computing and storage of the cloud database are separated. The computing layer database does not store user data, but is only responsible for computing. The storage layer uses a new storage system as a shared storage system.

Atomic write means that all write operations in an inseparable transaction must end or roll back together, and are indivisible. Traditional storage systems manage data in units of blocks (for example, a block is 512 bytes). In the absence of atomic writes, a certain write will partially succeed and partially fail; or in a concurrent scenario, two threads The written data overwrites each other. For example, one thread needs to write 123 and one thread needs to write 456. Concurrency may cause the write to 126. The function of atomic write is to ensure that either all writes succeed or all fail at one time, so the above situation can be avoided.

Append only storage is a file system customized for new storage hardware such as solid state drive (SSD). This file system provides append, delete, seal, etc. Basic operation, modification operation is not supported. The seal operation is an operation peculiar to the append only storage system, that is, a file is set to a read-only state and cannot be added.

In a distributed storage system, data is divided into user data and metadata. User data refers to the data that users really need to read and write. It is stored in fixed-size folders. Each folder corresponds to a unique identification (identification, ID). ). Metadata is the system data used to describe and manage the characteristics of a file, such as access permissions, file owner, file data block distribution information, etc. In a cluster file system, the distribution information includes the location of the file on the disk and the location of the disk. The location in the cluster. Users who need to manipulate a file must first get its metadata before they can locate the location of the file and get the content or related attributes of the file.

Generally, in a distributed scenario, in order to ensure the uniqueness of data writing, the master node can be selected by using a distributed consistency protocol, and then the master node will write the data, so as to ensure that only one node can write at any time Import data to ensure data consistency.

Refer to Figure 1. As shown in Figure 1, each node has three states, namely the standby state, the election state, and the master node state. All nodes are in the standby state at startup. If the message (such as heartbeat information) from the master node is not received within a preset time (for example, within 2 mental states), a state switch will occur, and the standby state will be switched to election status. When a node is in an election state, it will vote for itself first, and then pull votes from other nodes, that is, request other nodes to vote for itself so that it becomes the master node. If a node gets more than half of the total number of nodes in the cluster, then that node will become the new primary node, and other nodes will be switched to the standby state. The node in the master node state is the master node of the entire cluster, and all operations such as adding, modifying, and deleting system data can only be completed through the master node.

It can be understood that under the algorithm logic of this distributed consistency protocol, it can be guaranteed that there is only one master node at any time, that is, data consistency can be guaranteed. However, when the surviving nodes in the cluster are less than or equal to half of the total number of nodes, it can no longer be Election of a new master node.

In addition, in a distributed scenario, the uniqueness of the host can also be guaranteed through lock preemption. For example, when using a zookeeper cluster for distributed lock management, each node applies for a distributed lock from zookeeper. Zookeerer can authorize the use of locks to only one node based on first-come, first-served or weight distribution, and the node that finally obtains the lock can write data.

It should be understood that the distributed lock requires a long time and occupies a large bandwidth. In actual use, it is often necessary to negotiate between multiple nodes, and the performance of the entire system will suffer severe losses.

In order to solve the above-mentioned problems, this application provides a method and related equipment for ensuring data consistency, which can ensure data consistency and avoid data conflicts without losing database performance. Among them, the node may be a container, a virtual machine, a physical machine, etc., which is not limited in this application.

Refer to Fig. 2, which shows a possible application scenario of an embodiment of the present application. In this application scenario, the distributed storage system 200 includes a node management server 210, a node cluster 220, and a storage device 230. The node cluster 220 includes a master node 221, a backup node 222, a backup node 223, and a backup node 224. It should be understood that the node cluster 220 may also include more or less nodes, and the description is here by taking 4 nodes as an example. The storage device 230 includes a tenure management data storage unit 2310 and other data storage units 2320. The tenure management data storage unit 2310 is used to store the tenure identifier and root metadata identifier of the master node; the other data storage unit 2320 is used to store root metadata and metadata. Data and user data. The node management server 210 is used to monitor the nodes in the node cluster 220. When it is monitored that the master node 221 is abnormal, a backup node is selected, for example, the backup node 222 is selected to be upgraded to a new master node. The master node 221 works in a readable mode and a write mode, that is, the master node 221 can read data in the storage device 230 and can also write data to the storage device 230. The standby node 222, the standby node 223, and the standby node 224 work in The readable mode, that is, only the data in the storage device 230 can be read, and the data cannot be written. In particular, after the standby node 222 is upgraded to the new primary node, the term identifier of the primary node in the tenure management data storage unit 2310 can be updated, and data can be written to other data storage units 2320. The original primary node 221 is in the standby node. After updating the tenure identifier in 222, it can be determined that a new master node currently exists, and the original master node 221 can no longer write data, and can commit suicide or switch the working mode to read-only mode.

It is easy to understand that by setting the node management server 210 to manage the node cluster 220, a new master node can be directly selected when the master node 221 fails, without the need to use a distributed lock to determine the new master node, which can improve system performance . In addition, the storage device 230 stores the tenure identifier, which can ensure that the original master node can recognize that a new master node is generated. The original master node can avoid writing data with the new master node at the same time by suicide or switching working modes, so as to avoid causing data. Conflict to ensure data consistency.

With reference to the application scenario shown in FIG. 2, the method for ensuring data consistency provided by the embodiment of the present application will be described below in conjunction with FIG. 3. As shown in FIG. 3, the method includes but is not limited to the following steps:

S301: The first node receives the upgrade message sent by the node management server.

Specifically, the first node may specifically be a virtual machine or a container, etc., running in a physical machine. Multiple nodes form a cluster. For example, the node cluster 220 mentioned above. Under normal working conditions, there is only one primary node in the cluster, and other nodes are standby nodes. The primary node can write data, and other standby nodes cannot write data to ensure data. consistency. A cluster composed of multiple nodes can be deployed in a cloud environment, specifically one or more computing devices in the cloud environment (such as a central server); it can also be deployed in an edge environment, specifically one or more in the edge environment On the computing device (edge computing device), the edge computing device may be a server. Among them, the cloud environment refers to the central computing equipment cluster owned by the cloud service provider and used to provide computing, storage, and communication resources; the edge environment refers to the geographically far away from the central cloud environment, which is used to provide computing and storage. , The edge computing equipment cluster of communication resources.

Further, the first node may be any backup node in the node cluster, such as the above-mentioned backup node 222. During the operation of the first node, if it receives an upgrade message sent by the node management server, it indicates that the current master node may exist If it fails or is abnormal, the first node needs to be upgraded to the new master node.

S302: The first node updates the tenure management data.

Specifically, the first node needs to update the tenure management data after determining to upgrade to become the new master node. The tenure management data includes a root metadata identifier and a tenure identifier. The root metadata identifier is used to determine root metadata. That is, the root metadata identifier can be used to determine the specific location where the data is stored in the storage device. The tenure identifier is used It characterizes that the first node is upgraded to become a new master node, that is, the term identifier will change with the change of the master node. Whenever the node management server determines a new master node, the determined new master node The tenure indicator will be updated. Exemplarily, before the node management server determines the new master node, the term of office is identified as 5, that is, the cluster has produced 5 master nodes. When the node management server determines the new master node, the new master node will take the term of office The identifier is updated to 6, indicating that the new master node is the sixth master node generated by the cluster.

It should be noted that in a distributed storage system, the data written by users is eventually written into files of a fixed size, and each file is assigned a unique identifier. With the increase of written data, more files are needed. In order to manage these files, some specific data need to be used. These specific data are called metadata, and the metadata includes the identification of these files. Each metadata also has a unique identifier. Similarly, in order to facilitate the management of metadata, a root file needs to be used. The root file is also called root metadata. The root metadata includes all metadata identifiers.

Exemplarily, refer to Figure 4. As shown in Figure 4, all user data is stored in different files, such as file 1.1, file 1.2, etc. One metadata manages multiple files, for example, metadata 1 manages file 1.1 , File 1.2, …file 1.N, metadata 2 manages file 2.1, file 2.2, …file 2.N, all metadata is managed by root metadata, there is only one root metadata, and root metadata corresponds to one The identifier, the identifier and the tenure identifier belong to tenure management data and are stored together in the tenure management data unit.

It can be seen that by updating the tenure identifier of the first node, the original master node can be identified and a new master node can be determined to avoid writing data again, data conflicts can be avoided, and data consistency can be ensured.

In a possible implementation manner, the first node updates the tenure identifier while reading the root metadata identifier.

Specifically, the first node must ensure that reading root metadata and updating the tenure identifier occur at the same time, that is, the operation is atomic, otherwise the operation is abandoned and the execution continues again.

It is easy to understand that the original master node may have failed due to network fluctuations and other reasons, and the node management server has re-elected the first node as the new master node. However, the original master node may return to normal after a while, but The original master node cannot perceive that a new master node has appeared. At this time, there may be concurrency, that is, the original master node may continue to write data and update the root metadata identifier. If the first node does not read the root metadata and update the tenure identifier at the same time, there is a period of time between these two operations. For example, the first node reads the root metadata identifier first, and then updates the tenure identifier, which may result in data Inconsistent.

Exemplarily, when the first node reads the root metadata identifier, the original master node needs to modify the root metadata identifier, then the root metadata identifier read by the first node and the root metadata actually stored in the storage device Data identification will be inconsistent, and the first node needs to rely on root metadata identification to read or write data, which will eventually lead to data loss or data inconsistency. If the first node reads the root metadata identifier and modifies the tenure identifier at the same time, since the root metadata identifier is bound to the tenure identifier, the original master node needs to determine the tenure identifier and write it by itself when modifying the root metadata identifier. The entry term is the same, otherwise the modification will not succeed. Therefore, in this case, the original master node will not be able to successfully modify the root metadata identifier. At this time, the original master node can determine that there is a new master node. Then the original master node will stop modifying the root metadata identifier to avoid data conflicts and ensure data consistency.

S303: The first node sets the root metadata to a read-only mode.

Specifically, after the first node updates the tenure management data, the root metadata is set to a read-only mode, that is, the root metadata will not be allowed to be modified.

It should be understood that when the original master node does not perceive that there is a new master node, it still writes data to the storage device. As shown in Figure 4 above, the written data will be stored in a fixed-size file. If the file storage space can no longer support continued storage, a new file will be created for storage. At this time, the metadata needs to be modified, and relevant information such as the identification of the newly created file should be added to the metadata. Similarly, when the metadata The storage space of is also full. At this time, a new metadata needs to be created, and relevant information such as the identification of the newly created metadata needs to be added to the root metadata, and the first node has set the root metadata to read-only Therefore, the original master node will not be able to successfully modify the relevant information in the root metadata. At this time, the original master node will be able to determine that there is a new master node, and the original master node will abandon this operation and stop Write data to the storage device, use suicide and other methods to avoid data conflicts to ensure data consistency.

S304: The first node sets all metadata to a read-only mode.

Specifically, after setting the root metadata to the read-only mode, the first node sets all the metadata to the read-only mode, that is, all the metadata is not allowed to be modified.

It is easy to understand that when the original master node does not perceive the new node to write data, and the written data causes metadata to be modified, because the first node has set all metadata to read-only mode, the original The master node of will fail to modify. At this time, it can be confirmed that there is a new master node, then the original master node will abandon this operation, stop writing data to the storage device, and use suicide methods to avoid data conflicts to ensure data consistency.

S305: The first node sets all user data to a read-only mode.

Specifically, after setting all metadata to read-only mode, the first node sets all files storing user data to read-only mode, that is, all files are no longer allowed to write data.

From the description in S303 and S304 above, if the original master node needs to write data, it needs to write the data to the corresponding file, and the first node sets all files to read-only mode, resulting in the original master node. The node cannot write data to the storage device, that is, the write fails. At this time, the original master node can confirm that there is a new master node, and the original master node will abandon this operation and stop writing data to the storage device. Use suicide and other methods to avoid data conflicts to ensure data consistency.

It can be understood that the first node is set in a hierarchical manner, and the root metadata, metadata, and user data are set to read-only mode in turn, which can avoid data loss, ensure data consistency, and ensure that all files that have been written to the storage device can be accurately found. data.

S306: The first node updates the root metadata identifier and writes data to the storage device.

Specifically, the first node starts to perform the function of the master node after setting all files storing user data to the read-only mode. If the first node needs to write user data, because the first node has set all files to read-only mode, the first node needs to rebuild a file to store the written data. Because the file is newly created, more metadata is needed , And then need to update the root metadata and root metadata identification, and other nodes can only access the data in the storage device, and cannot write data to ensure data consistency.

It should be understood that steps S301 to S306 involved in the foregoing method embodiments are only schematic descriptions and summaries, and should not constitute specific limitations. The involved steps can be added, reduced, or combined as needed.

The foregoing describes the methods of the embodiments of the present application in detail. In order to facilitate better implementation of the above solutions in the embodiments of the present application, correspondingly, the following provides related equipment for cooperating with the implementation of the foregoing solutions.

Referring to FIG. 5, FIG. 5 is a schematic structural diagram of a first node provided by an embodiment of the present application. The first node may be the first node in the method embodiment described in FIG. 3, and may execute the method and steps in the method embodiment described in FIG. 3 where the first node is the execution subject. As shown in FIG. 5, the first node 500 includes a receiving module 510, an updating module 520, and a processing module 530. among them,

The receiving module 510 is configured to receive an upgrade message sent by a node management server, where the node management server is used to manage a node cluster, and the node cluster includes the first node;

The update module 520 is configured to update tenure management data, the tenure management data includes root metadata identification and tenure identification, the root metadata identification is used to determine root metadata, and the root metadata is used to manage the node cluster Corresponding metadata, where the tenure identifier is used to characterize that the first node is upgraded to the master node of the node cluster;

The processing module 530 is configured to set the data corresponding to the node cluster to a read-only mode, and the data includes the root metadata.

As an embodiment, the update module 520 is further configured to read the root metadata identifier and update the tenure identifier at the same time.

As an embodiment, the node cluster further includes a second node, the second node is used to read and write data corresponding to the node cluster and update the root metadata identifier; the update module 520 updates the tenure management After the data, the root metadata identifier is prohibited from being updated by the second node.

As an embodiment, the data corresponding to the node cluster includes root metadata, metadata, and user data, the metadata is used to manage the user data, and the user data is data written to the node cluster; The processing module 530 is further configured to set the metadata to the read-only mode after setting the root metadata to the read-only mode, and finally set the user data to the read-only mode.

As an embodiment, the processing module 530 is further configured to update the root metadata identifier and write data to the node cluster.

It can be understood that the receiving module 510 in the embodiment of the present application may be implemented by a transceiver or transceiver-related circuit components, and the update module 520 and the processing module 530 may be implemented by a processor or processor-related circuit components.

It should be noted that the above-mentioned structure of the first node is only used as an example, and should not constitute a specific limitation, and each module in the first node may be added, reduced, or combined as needed. In addition, the operation and/or function of each module in the first node is to realize the corresponding process of the method described in FIG. 3 above, and is not repeated here for brevity.

Refer to FIG. 6, which is a schematic structural diagram of a computing device provided by an embodiment of the present application. As shown in FIG. 6, the computing device 600 includes a processor 610, a communication interface 620, and a memory 630, and the processor 610, the communication interface 620, and the memory 630 are connected to each other through an internal bus 640. It should be understood that the computing device 600 may be a computing device in cloud computing or a computing device in an edge environment.

The processor 610 may be composed of one or more general-purpose processors, such as a central processing unit (CPU), or a combination of a CPU and a hardware chip. The aforementioned hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The above-mentioned PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a general array logic (generic array logic, GAL), or any combination thereof.

The bus 640 may be a peripheral component interconnect standard (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 640 can be divided into an address bus, a data bus, a control bus, and the like. For ease of presentation, only one thick line is used in FIG. 6, but it does not mean that there is only one bus or one type of bus.

The memory 630 may include a volatile memory (volatile memory), such as a random access memory (random access memory, RAM); the memory 630 may also include a non-volatile memory (non-volatile memory), such as a read-only memory (read-only memory). Only memory, ROM, flash memory, hard disk drive (HDD), or solid-state drive (SSD); the memory 630 may also include a combination of the above types. The memory 730 may be used to store programs and data, so that the processor 610 can call the program codes stored in the memory 630 to implement the aforementioned method for ensuring data consistency. The program code may be used to implement the functional module of the first node shown in FIG. 5, or used to implement the method steps in the method embodiment shown in FIG. 3 with the first node as the execution subject.

The present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it can implement any part of the method described in the above method embodiments. Or all steps.

The embodiment of the present invention also provides a computer program, which includes instructions, when the computer program is executed by a computer, the computer can execute part or all of the steps of any method for ensuring data consistency.

In the above-mentioned embodiments, the description of each embodiment has its own focus. For a part that is not described in detail in an embodiment, reference may be made to related descriptions of other embodiments.

It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that this application is not limited by the described sequence of actions. Because according to this application, some steps may be performed in other order or at the same time. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

In the several embodiments provided in this application, it should be understood that the disclosed device may be implemented in other ways. For example, the device embodiments described above are only illustrative, for example, the division of the above-mentioned units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or integrated. To another system, or some features can be ignored, or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical or other forms.

The units described above as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

In addition, the functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.

Claims

A method for ensuring data consistency, characterized in that the method includes:

The first node receives an upgrade message sent by a node management server, where the node management server is used to manage a node cluster, and the node cluster includes the first node;

The first node updates the tenure management data, the tenure management data includes a root metadata identifier and a tenure identifier, the root metadata identifier is used to determine root metadata, and the root metadata is used to manage the node cluster correspondence Metadata of, the tenure identifier is used to characterize the upgrade of the first node to the master node of the node cluster;

The first node sets the data corresponding to the node cluster to a read-only mode, and the data includes the root metadata.
The method according to claim 1, wherein the first node to update the tenure management data comprises:

The first node updates the tenure identifier while reading the root metadata identifier.
The method according to claim 1 or 2, wherein the node cluster further comprises a second node, and the second node is used to read and write data corresponding to the node cluster and update the root metadata identifier; After the first node updates the tenure management data:

The root metadata identifier is locked, and the root metadata identifier is prohibited from being updated by the second node.
The method according to any one of claims 1-3, wherein the data corresponding to the node cluster includes root metadata, metadata, and user data, and the metadata is used to manage the user data, and the User data is data written to the node cluster; the first node setting the data corresponding to the node cluster in a read-only mode includes:

After the first node sets the root metadata to the read-only mode, sets the metadata to the read-only mode, and finally sets the user data to the read-only mode.
The method according to claims 1-4, wherein after the first node sets the data corresponding to the node cluster to read-only mode, the method further comprises:

The first node updates the root metadata identifier and writes user data to the node cluster.
A first node, characterized in that it comprises:

A receiving module, configured to receive an upgrade message sent by a node management server, where the node management server is used to manage a node cluster, and the node cluster includes the first node;

The update module is used to update the tenure management data, the tenure management data includes root metadata identification and tenure identification, the root metadata identification is used to determine root metadata, and the root metadata is used to manage the corresponding node cluster Metadata of, the tenure identifier is used to characterize the upgrade of the first node to the master node of the node cluster;

The processing module is configured to set the data corresponding to the node cluster to a read-only mode, and the data includes the root metadata.
7. The first node according to claim 6, wherein the update module is further configured to read the root metadata identifier and update the tenure identifier at the same time.
The first node according to claim 6 or 7, wherein the node cluster further comprises a second node, and the second node is used to read and write data corresponding to the node cluster and update the root metadata Identification; after the update module updates the tenure management data, the root metadata identification is prohibited from being updated by the second node.
8. The first node according to any one of claims 6-8, wherein the data corresponding to the node cluster includes root metadata, metadata, and user data, and the metadata is used to manage the user data, The user data is data written to the node cluster;

The processing module is further configured to set the metadata to the read-only mode after setting the root metadata to the read-only mode, and finally set the user data to the read-only mode.
The first node according to any one of claims 6-9, wherein:

The processing module is also used to update the root metadata identifier and write data to the node cluster.
A computing device, wherein the computing device includes at least one storage unit and at least one processor, the at least one storage unit is configured to store at least one instruction, and the at least one processor executes the at least one instruction, Implement the method according to any one of claims 1-5.
A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1-5 is realized.
A computer program product, characterized in that the computer program product includes instructions, when the computer program is executed by a computer, the method according to any one of claims 1-5 is realized.