WO2020258665A1

WO2020258665A1 - Process guarantee method and device for distributed transactions

Info

Publication number: WO2020258665A1
Application number: PCT/CN2019/118181
Authority: WO
Inventors: 张兵; 夏勇
Original assignee: 深圳前海微众银行股份有限公司
Priority date: 2019-06-28
Filing date: 2019-11-13
Publication date: 2020-12-30
Also published as: CN110288255A

Abstract

A process guarantee method and device for distributed transactions, for use in solving the problem that a distributed transaction exception processing mechanism lacks of flexibility. The method comprises: receiving a processing request sent by a calling system (201); determining a main process and N sub-processes according to the processing request (202); for any one sub-process in the N sub-processes, sending the sub-process to a corresponding subsystem, so that the subsystem processes the sub-process (203); monitoring the processing processes of the N sub-processes, and when an exception occurs in the processing process of any sub-process, determining a corresponding exception processing strategy according to the exception type, the main process type, and the sub-process type (204); and according to the exception processing strategy, performing exception processing on the sub-processes, the exception processing comprising performing an asymmetric reverse operation on the successfully processed sub-processes, the asymmetric reverse operation being to perform reverse sub-processes, and the reverse sub-processes and the processed sub-processes being asymmetric (205).

Description

Process guarantee method and device for distributed transaction

Cross references to related applications

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on June 28, 2019, the application number is 201910580564.X, and the application name is "a method and device for process protection of distributed transactions", and the entire content of it is approved The reference is incorporated in this application.

Technical field

The present invention relates to the field of computer technology in financial technology (Fintech), and in particular to a process guarantee method and device for distributed transactions.

Background technique

With the continuous development of financial technology, especially Internet technology and finance, more and more technologies (such as distributed, blockchain, artificial intelligence, etc.) are applied in the financial field, but the financial industry has also proposed higher technology Requirements, such as the transaction guarantee process for distributed transactions.

Distributed transaction provides a mechanism to incorporate all operations involved in an activity into an indivisible execution unit. All operations that make up a transaction can only be submitted when all operations can be executed normally, as long as any operation fails. Will cause the rollback of the entire transaction.

The existing distributed transaction processing solution usually rolls back the entire pre-processing logic in a reverse order from the failure point of the sub-transaction, which cannot achieve the flexibility and difference of the correction logic.

Summary of the invention

The present application provides a process guarantee method and device for distributed transactions to solve the problem of lack of flexibility in a distributed transaction exception handling mechanism.

An embodiment of the present invention provides a process guarantee method for distributed transactions, including:

Receive processing requests sent by the calling system;

Determine the main process and N sub-processes according to the processing request, N≥1;

For any sub-process of the N sub-processes, send the sub-process to the corresponding subsystem, so that the subsystem can process the sub-process;

Monitor the processing process of N sub-processes, and when an exception occurs in the processing process of any sub-process, determine the corresponding exception processing strategy according to the exception type, the type of the main process, and the type of the sub-process;

Perform exception processing on a sub-process according to the exception processing strategy, the exception processing includes performing an asymmetric reverse operation on a sub-process that has been successfully processed, and the asymmetric reverse operation is executing a reverse sub-process, and The sub-process is asymmetric to the sub-process that has been processed.

In an optional embodiment, the N sub-processes include asynchronous sub-processes; for any of the N sub-processes, the sub-process is sent to the corresponding subsystem, so that all After the subsystem processes the sub-process, it also includes:

For the asynchronous sub-process, send a result query request to the asynchronous subsystem corresponding to the asynchronous sub-process at a set frequency until the processing result fed back by the asynchronous subsystem is received.

In an optional embodiment, the performing exception handling on the sub-process according to the exception handling strategy includes:

For the asynchronous sub-process, retry processing is performed.

In an optional embodiment, the monitoring the processing process of the N sub-processes includes:

Monitoring the processing time of the main process and each sub-process, and if the processing time of the main process and/or any of the sub-processes exceeds a corresponding time threshold, it is determined that a timeout exception occurs;

The performing exception handling on the sub-process according to the exception handling strategy includes:

If a timeout exception occurs in the asynchronous sub-process, retry the asynchronous sub-process;

Regarding the timeout exception in the main process and/or the synchronization sub-process, feedback the processing process timeout to the calling system and determine the processing status of the time-out sub-process;

If the processing status of the timeout sub-process is clear, perform an asymmetric reverse operation on the successfully processed sub-process;

If the processing status of the timeout sub-process is not clear, the main process is transferred to batch asynchronous processing.

In an optional embodiment, if the processing status of the timeout sub-process is not clear, transferring the main process to batch asynchronous processing includes:

After a predetermined period of time, send a result query request to the subsystem corresponding to the timeout sub-process;

If a successful processing result sent by the subsystem is received, perform an asymmetric reverse operation on the timeout subprocess and the subprocess before the timeout subprocess;

If the processing failure result sent by the subsystem is received, perform an asymmetric reverse operation on the sub-process before the time-out sub-process;

If an unclear feedback of the processing result sent by the subsystem is received, the step of sending a result query request to the subsystem corresponding to the timeout sub-process after the predetermined interval is executed.

In an optional embodiment, the receiving the processing request sent by the calling system includes:

Count the number of processing requests received;

If the number of processing requests is greater than the traffic warning value, an warning message is sent to the invoking system, and the time for feedback of response messages to the invoking system is controlled.

In an optional embodiment, for any one of the N sub-processes, the sub-process is sent to the corresponding subsystem, so that after the subsystem processes the sub-process, the include:

Receive response messages fed back by the subsystem;

Format the response message to obtain a response message;

Sending the response message to the calling system.

A process guarantee device for distributed transactions, including:

The transceiver unit is used to receive the processing request sent by the calling system;

The configuration unit is configured to determine the main process and N sub-processes according to the processing request, N≥1;

The transceiver unit is further configured to send the sub-process to the corresponding subsystem for any one of the N sub-processes, so that the subsystem can process the sub-process;

The monitoring unit is used to monitor the processing of N sub-processes. When an exception occurs in the processing of any sub-process, determine the corresponding exception processing strategy according to the type of the exception, the type of the main process, and the type of the sub-process ；

The exception processing unit is configured to perform exception processing on a sub-process according to the exception processing strategy, and the exception processing includes performing an asymmetric reverse operation on a sub-process that has been successfully processed, and the asymmetric reverse operation is executing a reverse sub-process Process, the reverse sub-process is asymmetrical with the processed sub-process.

In an optional embodiment, the N sub-processes include asynchronous sub-processes; the transceiver unit is further configured to:

In an optional embodiment, the exception handling unit is configured to:

For the asynchronous sub-process, retry processing is performed.

In an optional embodiment, the monitoring unit is configured to monitor the processing time of the main process and each sub-process, if the processing time of the main process and/or any sub-process exceeds the corresponding time Threshold, it is determined that a timeout exception occurs;

The exception processing unit is configured to perform a retry process on the asynchronous sub-process when a timeout exception occurs in the asynchronous sub-process; for the time-out exception in the main process and/or the synchronous sub-process, feedback processing to the calling system The process times out, and the processing status of the timeout sub-process is determined; if the processing status of the time-out sub-process is clear, an asymmetric reverse operation is performed on the successfully processed sub-process; if the processing status of the time-out sub-process is not clear, Then the main process is transferred to batch asynchronous processing.

In an optional embodiment, the exception handling unit is configured to:

In an optional embodiment, the transceiver unit is configured to:

Count the number of processing requests received;

In an optional embodiment, the transceiver unit is configured to:

Receive response messages fed back by the subsystem;

Format the response message to obtain a response message;

Sending the response message to the calling system.

The embodiment of the present invention also provides an electronic device, including:

At least one processor; and,

A memory communicatively connected with the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the method described above.

The embodiment of the present invention also provides a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium storing computer instructions, and the computer instructions are used to make the computer execute the method described above.

In the embodiment of the present invention, the processing request sent by the calling system is received, the processing request is parsed, and the corresponding main process and N sub-processes are matched. For N sub-processes, the sub-processes are sent to the corresponding subsystems in turn, so that the sub-processes are processed by the sub-processes, and the processing processes of the N sub-processes are monitored at the same time. When an exception occurs in the processing of any sub-process, according to the exception The type, main process type, and sub-process type determine the corresponding exception handling strategy. Exception handling is performed on the sub-process according to the exception handling strategy. The exception handling includes performing an asymmetric reverse operation on the sub-process that has been successfully processed. The asymmetric reverse operation here is to execute the reverse sub-process, and the reverse sub-process is the same as the reverse sub-process. The sub-processes that have been processed are asymmetric. In the embodiment of the present invention, the exception handling of the sub-process can be a symmetric reverse operation or an asymmetric reverse operation. The exception handling strategy can be flexibly set according to the actual situation and scenario, thereby flexibly solving the distributed transaction exception problem .

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings needed in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings may be obtained from these drawings without creative labor.

Figure 1 is a schematic structural diagram of a possible system architecture provided by an embodiment of the present invention;

2 is a schematic flowchart of a method for guaranteeing a distributed transaction process according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a symmetrical correction provided by a specific embodiment of the present invention;

FIG. 4 is a schematic diagram of an asymmetric correction provided by a specific embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a process guarantee apparatus for distributed transactions according to an embodiment of the present invention;

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

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. . Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

As shown in FIG. 1, a system architecture to which the embodiment of the present invention is applicable includes a calling system 101, a transaction processing server 102, and N subsystems 103. Among them, the calling system 101 can be a client installed on a terminal or an operating system installed on a server. The terminal can be an electronic device with wireless communication functions such as a mobile phone, a tablet computer, or a dedicated handheld device, or it can be It is a personal computer (PC), notebook computer, server and other wired access devices connected to the Internet. The terminal can be an independent device or a terminal cluster formed by multiple terminals. Preferably, the terminal can use cloud computing technology for information processing. The transaction processing server 102 may be a network device such as a computer, it may be an independent device, or a server cluster formed by multiple servers. The N subsystems 103 can be installed on N servers respectively, or on less than N servers, that is, there are two or more subsystems 103 installed on the same server. Preferably, the transaction processing server 102 and the N subsystems 103 can use cloud computing technology for information processing.

The calling system 101 can communicate with the transaction processing server 102 through the INTERNET network, or through mobile communication systems and servers such as the Global System for Mobile Communications (GSM), long term evolution (LTE) system, etc. 101 to communicate. The transaction processing server 102 can remotely call the subsystem 103 in RMB, or through other methods such as RPC, or through the Internet, Global System for Mobile Communications (GSM), or long term evolution (long term). A mobile communication system such as an evolution (LTE for short) system communicates with the subsystem 103.

For ease of understanding, the following defines and explains the terms that may be involved in the embodiments of the present invention.

Deposit combination transaction system: mainly used for the transaction access system for the linkage consumption of current + and deposit balance under the bank support consumption scenario. It can also be used for the linkage consumption of bank deposits and loans, as well as the linkage consumption of cooperative bank deposit accounts and bank deposit accounts .

Distributed sub-transactions: part of the transaction-type processing logic that constitutes a group of distributed transactions, which can return the processing exceptions to the pre-processing state or the equivalent state through correction or compensation.

Main transaction: Corresponding to a set of distributed business processing logic, which can be composed of one or more synchronous or asynchronous processing sub-transactions, and has key attributes such as the main transaction type, processing status, correction status, and processing time.

Sub-transaction: A business processing sub-unit split from a set of distributed business processing logic, corresponding to a distributed sub-transaction, which can be synchronous or asynchronous processing, with processing status, reversing status, and processing type And other business attributes.

Active reversal transaction: A type of main transaction, a compensation processing mechanism automatically generated by the processing model of this invention when the distributed transaction is abnormal, and the purpose is to make the original main transaction achieve the final business consistency state.

Process decision controller: The main function of the component of the present invention is to determine the form of the distributed transaction corresponding to this request based on the input information item through the pre-configured processing method of the business, such as the split and definition of sub-transactions, and the remote system interface Of ok. The system can change the form of business processing according to the preset configuration to meet the needs of business changes.

Anomaly detection: This model recognizes anomaly when the distributed transaction is abnormal.

Exception handling decision maker: This model generates corresponding exception handling control parameters according to the type of the exception after the occurrence and detection of a distributed transaction exception.

Reverse operation: After an exception occurs in the data processing process, the compensation processing mechanism is executed in order to make the data stored in the database reach the final consistency state. In the specific business scenario of the embodiment of the present invention, the reverse operation includes a rollback operation and/or correction processing. The reverse operation also includes one or more reverse sub-processes, which are symmetrical or asymmetrical with the normal processing sub-processes. The reverse sub-process is a reversal sub-process in the specific business scenario of the embodiment of the present invention.

In the prior art, in order to implement distributed transactions, a group of distributed transactions is generally fixed first, denoted as T ₁ , T ₂ , T ₃ ... T _n , where T _i is an equal sub-transaction. Corresponding to each sub-transaction, corresponding compensation sub-transactions C ₁ , C ₂ , C ₃ … C _{j are set} , where 0<j<n. When executing, it is executed in the order of T ₁ , T ₂ , T ₃ …T _n . If a transaction occurs when performing sub abnormal T _j, in addition to the pair T _j transaction rollback is required, but also the sub-sub-transaction executed before the transaction T _j Reversal, the compensation is performed subtransactions provided, performed The order is C _j-1 , C _j-2 , ... C ₂ , C ₁ . In the prior art, the reconciliation of sub-transactions can only be executed according to a fixed process, and there is no flexibility.

In order to solve the above problems, based on the above architecture, an embodiment of the present invention provides a process guarantee method for distributed transactions. As shown in FIG. 2, the process guarantee method for distributed transactions provided by the embodiment of the present invention includes the following steps:

Step 201: Receive a processing request sent by the calling system.

In the specific implementation process, the system provides a message service interface that can be called externally. The interface can be designed as multiple. Each interface defines the message structure and field meaning and other configuration information. Specifically, the corresponding interface request message field and The definition of the business meaning of the response message fields, including the field type, length, whether it is required, description of use, etc. Therefore, the device in the embodiment of the present invention can receive the processing request sent by the external calling system through the interface, and feedback the response message to the external calling system through the interface.

Step 202: Determine the main process and N sub-processes according to the processing request, N≥1.

In the specific implementation process, after a new processing request is entered, the request message field and other information of the processing request are analyzed and combined with the configuration service fields controlled by the system to determine whether the processing request is a legal request. The configuration field can be the calling system ID, business request channel number, etc.

After a legal processing request is received, the corresponding main process and sub-process instances will be matched after parsing and processing according to the business information items carried in the processing request. In the embodiment of the present invention, the type definition of the main process, the sub-process, and the process matching factory class are designed.

For the main process, the main business information items include: requested channel, business scenario ID, partner ID, summary code, account type of the borrower and lender, etc., which can be expanded vertically or horizontally according to actual needs, and finally matched to obtain the main process type.

Sub-processes are sub-processes that are determined based on business information items of more dimensions after the main process type is determined. For example, the customer Zhang San paid by scanning the code and bought a cup of coffee for 30 yuan. After analysis, the matching result of the main process is the consumption process, not a refund or other. Furthermore, the sub-process will determine whether it is necessary to link the bank wealth management account and the bank card according to whether Zhang San has a bank wealth management account. Therefore, the sub-processes under the same main process may be different. If a bank wealth management account is not opened, then the sub-process may be a single-card consumption by a bank card; if a bank wealth management account is opened, then the sub-process is a combined consumption of a bank wealth management account and a bank card. The matching of sub-process types is also set through a matrix type, which can be expanded horizontally and vertically according to different business needs.

Step 203: For any sub-process in the N sub-processes, send the sub-process to the corresponding subsystem, so that the subsystem can process the sub-process.

In the specific implementation process, a processing request will determine a main process and multiple sub-processes linked to it. This is a two-tier transaction model. The main process is mainly used to track and control the progress and status of the overall distributed transaction processing, the correction status, and abnormal information. The sub-process will correspond to a specific interaction with the remote system, and also set the status of the sub-process, the sub-process's correction status, exception information, remote call type and other information. In the embodiment of the present invention, according to the processing steps of the sub-process and the information of the corresponding sub-process, the message of the sub-process can be sent to the corresponding subsystem for remote transaction processing through the remote calling tool.

Multiple sub-processes can be processed synchronously or asynchronously. Take the example of Zhang San buying coffee. Assuming that Zhang San’s bank card balance is 10 yuan, and the share of wealth management products is 100 yuan, then the main process generated for this consumption request to buy coffee is the consumption process, and the sub-processes will be contain:

1. Go to the bank's wealth management system to freeze the share of wealth management products at 20 yuan;

2. Go to the deposit core system for group one-to-one transfer transactions of 30 yuan, of which 10 yuan is withdrawn from the bank card, and 20 yuan is withdrawn from the cleared funds in transit;

3. Go to the bank's wealth management system to process the share redemption of wealth management products, and complete the settlement process of the in-transit liquidation funds.

Among the above three sub-processes, sub-processes 1 and 2 are synchronous processes, that is, they need to be executed immediately and wait for the result to be returned. Sub-process 3 is an asynchronous processing process. You can wait for the previous synchronous sub-processes 1 and 2 to execute successfully and return, and then choose an opportunity to process, and different processes can be performed according to the type of sub-process. In this way, the timeliness of the synchronization operation can be made higher, and the overall throughput of the system can be improved.

Step 204: Monitor the processing processes of the N sub-processes. When an exception occurs in the processing process of any sub-process, determine the corresponding exception processing strategy according to the type of the exception, the type of the main process, and the type of the sub-process.

Specifically, the reverse operation process in the embodiment of the present invention is also processed in the manner of the main-sub-process model. At the same time, the sub-processes of the reverse operation process and the sub-processes of the original process can have a one-to-one correspondence, which we call symmetric reverse operations, or the sub-processes of the reverse operation process and the sub-processes of the original process do not correspond one-to-one. However, the effect of the execution of the reverse sub-process is the same as the effect of all the atomic processes that have not been executed. We call this asymmetric reverse operation. The following example illustrates the symmetrical reverse operation and the asymmetrical reverse operation with the correction process.

Symmetrical punching:

If Zhang San buys coffee, his bank balance is 30 yuan, enough to buy a cup of coffee, then the actual main process is the consumption process, and there is only one sub-process, namely the "core one-to-one transfer", and there is no need to link bank financial accounts. All consumption amounts are withdrawn from the bank card. Assuming that the deposit core withdrawing money from the bank card is abnormal, but Zhang San’s bank card has transferred the amount of 30 yuan at this time, then the corresponding reconciliation sub-process needs to be set up, and the reconciliation sub-process is also one, that is, "core pair "One charge", you only need to go to the core of the deposit to charge the original consumer transaction. In this case, the original process and the correction process are completely symmetrical, as shown in Figure 3.

Asymmetrical punching:

If Zhang San’s bank card balance is 10 yuan, which is less than 30, and the bank’s wealth management products have 100 yuan, then the actual main process is the consumption process, and there are three sub-processes, namely "freezing financial products" and "core one-to-one transfer "And "Financial management unfreeze and redeem." However, after the original process is completed, an external reversal transaction is initiated. If it is a symmetric reversal, three reversal sub-processes need to be set up for the three sub-processes, namely "defrosting of financial products", "core one-to-one reversal" and "financial products "Buy" means to return 10 yuan to the bank card and re-purchase 20 yuan of financial products. Instead of asymmetric reconciliation, you can set up a reconciliation sub-process, such as "core one-to-one transfer", and directly refund 30 yuan to the bank card, then there is only one corresponding reconciliation sub-process, and there are three atomic processes, so it is right and wrong Symmetrical, as shown in Figure 4. For Zhang San at this time, his assets have not been reduced, that is, the effect after the correction is equivalent to the effect of the atomic process is not executed, and the operation process is greatly simplified compared with the symmetrical correction. It should be noted that the asymmetric correction is not only because the number of offset sub-processes is different from the number of atomic processes, but also the number of offset sub-processes is the same as the number of atomic processes, but there is no one-to-one correspondence. For example, in the above example, It can be divided into three transactions, and each transaction returns 10 yuan to the bank card. This situation is also asymmetrical.

Step 205: Perform exception processing on the sub-process according to the exception processing strategy. The exception handling includes performing an asymmetric reverse operation on a sub-process that has been successfully processed. The asymmetric reverse operation is performing a reverse operation sub-process, and the reverse operation sub-process is different from the processed sub-process. symmetry.

The reverse operation here can be a rollback operation on the sub-process, a reversal operation on the sub-process, or a reversal operation on the sub-process.

Further, in order to control message flow and reduce system pressure, in the embodiment of the present invention, said receiving the processing request sent by the calling system includes:

Count the number of processing requests received;

In the specific implementation process, the number of processing requests received is monitored 24/7. When a new processing request is received, the counter variable is increased by 1. At the same time, different traffic counters can be set up according to different types of interfaces to target different Configure the high-flow warning value and maximum value for the request type or total system flow. When the number of a certain type of request or the total number of processing requests triggers the warning value, a message warning can be provided; when the maximum value is reached, the limit response message is immediately returned to the calling system to achieve the purpose of system protection.

In the prior art, the subsystem provides a message service interface for asynchronous processing, but generally distributed transactions require the subsystem to synchronously feedback the message processing result. In order to solve this problem, in the embodiment of the present invention, the N sub-processes include asynchronous sub-processes; for any one of the N sub-processes, the sub-process is sent to the corresponding subsystem, So that after the subsystem processes the sub-process, it also includes:

The embodiment of the present invention sets an implementation mechanism of asynchronous to synchronous conversion, which better solves the problem that the asynchronous interface of the subsystem cannot meet the requirement of synchronously returning processing results. Specifically, after the asynchronous sub-process is sent to the corresponding asynchronous subsystem, the remote result query interface of the asynchronous subsystem is called according to the serial number of the main process to obtain the processing result of the asynchronous sub-process. If the timeout time of the asynchronous sub-process is not exceeded, wait for a certain time interval and retry the call of the remote result query interface until the specific return result is obtained. In addition, if the result is not obtained by the set timeout time of the asynchronous sub-process, it is directly considered that the asynchronous sub-process has timed out abnormally, and the exception processing logic will be entered.

Further, the performing exception handling on the sub-process according to the exception handling strategy includes:

For the asynchronous sub-process, retry processing is performed.

Specifically, since the execution of the asynchronous sub-process does not need to return the processing result synchronously, in response to the abnormal situation of the asynchronous sub-process, the asynchronous sub-process can be retried.

During the execution of the main process or sub-process, if it is clear that business or technical anomalies occur, for example, abnormal accounting status or database abnormality occurred during transfer, in the embodiment of the present invention, the sub-process can immediately execute the correction processing , It can also be the processing of transferring asynchronous tasks, which can be specifically configured as required, and the configuration table itself can also be horizontally or vertically expanded to meet the needs of the business.

The other type of exception is the timeout exception. The monitoring of the processing process of the N sub-processes includes:

In the specific implementation process, timeout exceptions are divided into the following different situations. In the embodiment of the present invention, the main flow timeout time and the sub flow timeout time are introduced. From the time when a processing request is received, to the current processing time, the main process timeout time is a system threshold set for the above time period, used to determine whether the execution of a main process has a timeout abnormal situation. If the main process exceeds the set main process timeout time, the calling system returns an exception status code of the main process timeout, which indicates that the process is in an intermediate state of processing timeout. The intermediate state of the processing timeout is regarded as a processing failure to the calling system, and exception handling is required. Specifically, it is necessary to obtain the processing status of the timeout sub-process. If the time-out sub-process has been processed successfully, the time-out sub-process and the sub-processes that have been successfully processed before the time-out sub-process are subjected to correction processing, here the correction processing It can be either symmetrical or asymmetrical, which ultimately brings the data back to the equivalent state at the beginning of the business. If the timeout sub-process is not processed successfully, there is no need to process the time-out sub-process, and only the sub-processes that have been successfully processed before the time-out sub-process need to be corrected. If the processing status of the timeout sub-process is not clear, the main process should be converted to batch asynchronous processing. For example, the main process that needs batch asynchronous processing should be gathered together, and the processing status can be re-queried at a fixed time every day, and then according to The query results are processed later. Turning the main process with unclear processing status to batch asynchronous processing for continuous subsequent processing operations can ensure that the final results of abnormal transactions meet the consistency of business requirements.

Batch asynchronous processing is realized through qrtz timing tasks. For different types of batch asynchronous processing main processes, different asynchronous processing strategies can be configured, such as task scheduling processing frequency, scheduling time range, scheduling processing thread resources Input etc.

Similar to the timeout status of the main process, the subprocesses are also set with different system thresholds for the timeout status. The processing time of a sub-process is the time from the moment a sub-process is sent to the corresponding subsystem to the current processing time interval. Sub-process timeout means that the time-consuming time recorded by the timer of the sub-process exceeds the system threshold of the timeout period set by the system for the current type of sub-process. Regardless of the timeout of the main process or the timeout of the sub-process that requires synchronous processing, it will affect the processing result of the request. The system will intercept these timeout exceptions and interrupt the execution of the process, and return the process status to timeout.

In addition, for an asynchronous sub-process with a timeout exception, since the timeout of the asynchronous sub-process does not affect the processing result of the current request, the asynchronous sub-process can be retryed. The retry processing here can be found that the asynchronous sub-process appears When the timeout is abnormal, retry directly, or perform batch asynchronous processing of the asynchronous sub-process, for example, perform the retry processing of the asynchronous sub-process within a fixed time period every day. Of course, you can also perform offset processing on asynchronous sub-processes according to the configuration.

Further, the embodiment of the present invention provides the invoking system with consistent feedback results. After sending the sub-process to the corresponding subsystem for any one of the N sub-processes, so that the subsystem processes the sub-process, the method further includes:

Receive response messages fed back by the subsystem;

Format the response message to obtain a response message;

Sending the response message to the calling system.

For example, a calling system sends a transaction processing request to the transaction processing server, and the transaction processing server times out and returns a timeout status. The calling system may query the result of the transaction in the transaction processing server according to the specific exception code that has timed out, and it needs to process the returned response message according to the specific transaction status information.

In addition, the packaging of the feedback response messages can be done as much as possible. When there are system changes, especially when new system semantics appear, after message formatting, consistent and stable output results are provided to the outside world. For example, the external feedback of the timeout exception code is 000960, because the service needs access to the new downstream subsystem, but the timeout exception code of the newly accessed subsystem is a new value such as XXXTIMEOUT, then the response message format To change it, XXXTIMEOUT needs to be converted and packaged into the timeout exception code 000960 agreed with the calling system, so that the calling system can also recognize it.

In order to understand the present invention more clearly, the above process will be described in detail with specific embodiments below. In specific embodiments of the present invention, the specific scenario is that users use mobile phones to make quick payments. The system architecture includes client, transaction processing system, and deposit core system. And the financial management system, where the deposit core system and the financial management system are the corresponding subsystems of the sub-transactions.

The steps of the specific embodiment are as follows, including:

Step S501: The client sends a quick consumption payment request to the transaction processing system.

Step S502: The transaction processing system parses the quick consumption payment request, and sends a balance query request to the deposit core system.

Step S503: The transaction processing system determines that the balance in the user's bank card is insufficient according to the feedback result of the deposit core system, and the client is set to allow linked consumption of wealth management products, and then sends a balance inquiry request to the wealth management system.

Step S504: The transaction processing system determines the sub-processes as "frozen wealth management products", "core one-to-one transfer" and "financial management unfreeze and redeem" according to the feedback result of the wealth management system.

Step S505: The transaction processing system sends a request to freeze the wealth management product to the wealth management system. If the financial management system is executed successfully, step 506 is executed. If the execution of the financial management system fails, since it has no effect on the result of the distributed transaction, it does not need to be processed, and the failure result is directly reported to the client. If the execution of the wealth management system expires, the processing result of the request to freeze the wealth management product needs to be verified. If it fails, there is no need to deal with it. If it succeeds, it needs to perform the correction processing, that is, execute the unfreeze of the wealth management product.

Step S506: The transaction processing system sends a deduction transfer request to the deposit core system. If the deposit core system is executed successfully, the successful deduction result is fed back to the client, and step 507 is executed. If the implementation of the deposit core system fails, the successful "frozen wealth management products" need to be corrected, that is, the wealth management products are unfrozen. If the deposit core system executes overtime, the processing result of the deduction transfer request needs to be verified. If it fails, the financial product thawing is executed; if it succeeds, the asynchronous redemption is executed and the status is updated. Here, between the sub-process "asynchronous redemption and update status" and the sub-process "freezing financial products" and "core one-to-one transfer" is an asymmetric correction.

Step S507: The transaction processing system sends to the wealth management system a request for wealth management thawing and wealth management redemption, where the requests for wealth management thawing and wealth management redemption are asynchronous requests. If the financial management system is executed successfully, the transaction processing system will feedback the successful redemption result to the client. If the financial management system fails to execute, it will switch to batch asynchronous processing and perform retry processing. If the financial management system executes overtime, the processing status needs to be verified and the processing can be manually intervened.

The embodiment of the present invention also provides a process guarantee device for distributed transactions, as shown in FIG. 5, including:

The transceiver unit 601 is configured to receive a processing request sent by the calling system;

The configuration unit 602 is configured to determine the main process and N sub-processes according to the processing request, where N≥1;

The transceiver unit 601 is further configured to send the sub-process to the corresponding subsystem for any one of the N sub-processes, so that the subsystem can process the sub-process;

The monitoring unit 603 is used to monitor the processing of N sub-processes. When an abnormality occurs in the processing of any sub-process, determine the corresponding exception processing according to the type of the exception, the type of the main process, and the type of the sub-process Strategy

The exception processing unit 604 is configured to perform exception processing on the sub-processes according to the exception processing strategy. The exception processing includes performing an asymmetric reverse operation on the sub-processes that have been successfully processed, and the asymmetric reverse operation is performing reverse operations. A sub-process, the reverse sub-process is asymmetric with the processed sub-process.

Optionally, the N sub-processes include asynchronous sub-processes; the transceiver unit 601 is further configured to:

Optionally, the exception handling unit 604 is configured to:

For the asynchronous sub-process, retry processing is performed.

Optionally, the monitoring unit 603 is configured to monitor the processing time of the main process and each sub-process, and if the processing time of the main process and/or any of the sub-processes exceeds the corresponding time threshold, determine A timeout exception occurred;

The exception processing unit 604 is configured to perform retry processing on the asynchronous sub-process when a timeout exception occurs in the asynchronous sub-process; and feedback to the calling system when a time-out exception occurs in the main process and/or the synchronous sub-process The processing process is timed out, and the processing status of the timeout sub-process is determined; if the processing status of the time-out sub-process is clear, an asymmetric reverse operation is performed on the successfully processed sub-process; if the processing status of the time-out sub-process is not clear , The main process is transferred to batch asynchronous processing.

Optionally, the exception handling unit 604 is configured to:

Optionally, the transceiver unit 601 is configured to:

Count the number of processing requests received;

Optionally, the transceiver unit 601 is configured to:

Receive response messages fed back by the subsystem;

Format the response message to obtain a response message;

Sending the response message to the calling system.

Based on the same principle, the present invention also provides an electronic device, as shown in FIG. 6, including:

It includes a processor 701, a memory 702, a transceiver 703, and a bus interface 704, wherein the processor 701, the memory 702 and the transceiver 703 are connected through the bus interface 704;

The transceiver 703 is configured to receive a processing request sent by the calling system under the control of the processor 701;

The processor 701 is configured to read a program in the memory 702 and execute the following method: determine the main process and N sub-processes according to the processing request, N≥1;

The transceiver 703 is further configured to, under the control of the processor 701, for any one of the N sub-processes, send the sub-process to the corresponding subsystem, so that the subsystem can process the sub-process ；

The processor 701 is configured to read the program in the memory 702 and execute the following method: monitor the processing process of N sub-processes, and when an exception occurs in the processing process of any sub-process, according to the abnormal type and the type of the main process And the type of the sub-process determines the corresponding exception handling strategy; the exception handling is performed on the sub-process according to the exception handling strategy, and the exception handling includes performing an asymmetric reverse operation on the successfully processed sub-process, and the asymmetric The reverse operation is to execute a reverse sub-process, and the reverse sub-process is asymmetrical to the processed sub-process;

The memory 702 is configured to store one or more executable programs, and can store data used by the processor when performing operations;

The 704 bus interface provides interfaces, and the processor is responsible for managing the bus architecture and general processing.

Wherein, in FIG. 6, the bus architecture may include any number of interconnected buses and bridges. Specifically, one or more processors represented by the processor 701 and various circuits of the memory represented by the memory 702 are linked together. The bus architecture can also link various other circuits such as peripherals, voltage regulators, power management circuits, etc., which are all known in the art, and therefore, will not be further described herein.

The present invention is described with reference to flowcharts and/or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and/or block in the flowchart and/or block diagram, and the combination of processes and/or blocks in the flowchart and/or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing equipment to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing equipment are generated It is a device that realizes the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing equipment to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction device. The device implements the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

These computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operation steps are executed on the computer or other programmable equipment to produce computer-implemented processing, so as to execute on the computer or other programmable equipment. The instructions provide steps for implementing functions specified in a flow or multiple flows in the flowchart and/or a block or multiple blocks in the block diagram.

Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. In this way, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

A process guarantee method for distributed transactions is characterized in that it includes:

Receive processing requests sent by the calling system;

Determine the main process and N sub-processes according to the processing request, N≥1;

For any sub-process of the N sub-processes, send the sub-process to the corresponding subsystem, so that the subsystem can process the sub-process;

Monitor the processing process of N sub-processes, and when an exception occurs in the processing process of any sub-process, determine the corresponding exception processing strategy according to the exception type, the type of the main process, and the type of the sub-process;

Perform exception processing on a sub-process according to the exception processing strategy, the exception processing includes performing an asymmetric reverse operation on a sub-process that has been successfully processed, and the asymmetric reverse operation is executing a reverse sub-process, and The sub-process is asymmetric to the sub-process that has been processed.
The method according to claim 1, wherein the N sub-processes include asynchronous sub-processes; for any one of the N sub-processes, the sub-process is sent to the corresponding subsystem , So that after the subsystem processes the sub-process, it also includes:

If the sub-process is an asynchronous sub-process, a result query request is sent to the asynchronous subsystem corresponding to the asynchronous sub-process at a set frequency until the processing result fed back by the asynchronous sub-process is received.
The method according to claim 2, wherein the performing exception handling on the sub-process according to the exception handling strategy comprises:

For the asynchronous sub-process, retry processing is performed.
The method of claim 1, wherein the monitoring the processing of the N sub-processes comprises:

Monitoring the processing time of the main process and each sub-process, and if the processing time of the main process and/or any of the sub-processes exceeds a corresponding time threshold, it is determined that a timeout exception occurs;

The performing exception handling on the sub-process according to the exception handling strategy includes:

If a timeout exception occurs in the asynchronous sub-process, retry the asynchronous sub-process;

Regarding the timeout exception in the main process and/or the synchronization sub-process, feedback the processing process timeout to the calling system and determine the processing status of the time-out sub-process;

If the processing status of the timeout sub-process is clear, perform an asymmetric reverse operation on the successfully processed sub-process;

If the processing status of the timeout sub-process is not clear, the main process is transferred to batch asynchronous processing.
The method of claim 4, wherein if the processing status of the timeout sub-process is not clear, transferring the main process to batch asynchronous processing includes:

After a predetermined period of time, send a result query request to the subsystem corresponding to the timeout sub-process;

If a successful processing result sent by the subsystem is received, perform an asymmetric reverse operation on the timeout subprocess and the subprocess before the timeout subprocess;

If the processing failure result sent by the subsystem is received, perform an asymmetric reverse operation on the sub-process before the time-out sub-process;

If an unclear feedback of the processing result sent by the subsystem is received, the step of sending a result query request to the subsystem corresponding to the timeout sub-process after the predetermined interval is executed.
The method according to claim 1, wherein the receiving a processing request sent by the calling system comprises:

Count the number of processing requests received;

If the number of processing requests is greater than the traffic warning value, an warning message is sent to the invoking system, and the time for feedback of response messages to the invoking system is controlled.
The method according to any one of claims 1 to 6, wherein for any one of the N sub-processes, the sub-process is sent to a corresponding subsystem, so that the sub-process is After the system processes the sub-process, it also includes:

Receive response messages fed back by the subsystem;

Format the response message to obtain a response message;

Sending the response message to the calling system.
A process guarantee device for distributed transactions is characterized in that it includes:

The transceiver unit is used to receive the processing request sent by the calling system;

The configuration unit is configured to determine the main process and N sub-processes according to the processing request, N≥1;

The transceiver unit is further configured to send the sub-process to the corresponding subsystem for any one of the N sub-processes, so that the subsystem can process the sub-process;

The monitoring unit is used to monitor the processing of N sub-processes. When an exception occurs in the processing of any sub-process, determine the corresponding exception processing strategy according to the type of the exception, the type of the main process, and the type of the sub-process ；

The exception processing unit is configured to perform exception processing on a sub-process according to the exception processing strategy, and the exception processing includes performing an asymmetric reverse operation on a sub-process that has been successfully processed, and the asymmetric reverse operation is executing a reverse sub-process Process, the reverse sub-process is asymmetrical to the processed sub-process.
The apparatus according to claim 8, wherein the N sub-processes include asynchronous sub-processes; the transceiver unit is further configured to:

For the asynchronous sub-process, send a result query request to the asynchronous subsystem corresponding to the asynchronous sub-process at a set frequency until the processing result fed back by the asynchronous subsystem is received.
The device according to claim 9, wherein the abnormal processing unit is configured to

For the asynchronous sub-process, retry processing is performed.
The device according to claim 8, wherein the monitoring unit is configured to monitor the processing time of the main process and each sub-process, if the processing time of the main process and/or any sub-process If the corresponding time threshold is exceeded, it is determined that a timeout exception occurs;

The exception processing unit is configured to perform a retry process on the asynchronous sub-process when a timeout exception occurs in the asynchronous sub-process; for the time-out exception in the main process and/or the synchronous sub-process, feedback processing to the calling system The process times out, and the processing status of the timeout sub-process is determined; if the processing status of the time-out sub-process is clear, an asymmetric reverse operation is performed on the successfully processed sub-process; if the processing status of the time-out sub-process is not clear, Then the main process is transferred to batch asynchronous processing.
The device according to claim 11, wherein the exception processing unit is configured to:

After a predetermined period of time, send a result query request to the subsystem corresponding to the timeout sub-process;

If a successful processing result sent by the subsystem is received, perform an asymmetric reverse operation on the timeout subprocess and the subprocess before the timeout subprocess;

If the processing failure result sent by the subsystem is received, perform an asymmetric reverse operation on the sub-process before the time-out sub-process;

If an unclear feedback of the processing result sent by the subsystem is received, the step of sending a result query request to the subsystem corresponding to the timeout sub-process after the predetermined interval is executed.
The device according to claim 8, wherein the transceiver unit is configured to:

Count the number of processing requests received;

If the number of processing requests is greater than the traffic warning value, an warning message is sent to the invoking system, and the time for feedback of response messages to the invoking system is controlled.
The device according to any one of claims 8 to 13, wherein the transceiver unit is configured to:

Receive response messages fed back by the subsystem;

Format the response message to obtain a response message;

Sending the response message to the calling system.
An electronic device, characterized by comprising: a processor, a memory, a transceiver, and a bus interface, wherein the processor, the memory and the transceiver are connected through the bus interface;

The transceiver is configured to receive a processing request sent by the calling system under the control of the processor;

The processor is configured to read the program in the memory and execute the following method: determine the main process and N sub-processes according to the processing request, N≥1;

The transceiver is configured to send the sub-process to the corresponding subsystem for any one of the N sub-processes under the control of the processor, so that the subsystem performs the sub-process deal with;

The processor is configured to read the program in the memory, and execute the following method: monitor the processing process of N sub-processes, and when an exception occurs in the processing process of any sub-process, according to the abnormal type and the main process The type of the sub-process and the type of the sub-process determine the corresponding exception handling strategy; perform exception handling on the sub-process according to the exception handling strategy, and the exception handling includes performing an asymmetric reverse operation on the successfully processed sub-process, the An asymmetric reverse operation is to execute a reverse sub-process, and the reverse sub-process is asymmetric to the processed sub-process;

The memory is used to store one or more executable programs, and can store data used by the processor when performing operations;

The bus interface is used to provide an interface.
The electronic device according to claim 15, wherein the N sub-processes include asynchronous sub-processes; the transceiver is further used for:

If the sub-process is an asynchronous sub-process, the result query request is sent to the asynchronous subsystem corresponding to the asynchronous sub-process at a set frequency until the processing result fed back by the asynchronous sub-process is received.
The electronic device of claim 16, wherein the processor is further configured to:

For the asynchronous sub-process, retry processing is performed.
The electronic device according to claim 15, wherein the processor is specifically configured to:

Monitoring the processing time of the main process and each sub-process, and if the processing time of the main process and/or any of the sub-processes exceeds a corresponding time threshold, it is determined that a timeout exception occurs;

If a timeout exception occurs in the asynchronous sub-process, retry the asynchronous sub-process;

Regarding the timeout exception in the main process and/or the synchronization sub-process, feedback the processing process timeout to the calling system and determine the processing status of the time-out sub-process;

If the processing status of the timeout sub-process is clear, perform an asymmetric reverse operation on the successfully processed sub-process;

If the processing status of the timeout sub-process is not clear, the main process is transferred to batch asynchronous processing.
The electronic device according to claim 18, wherein the transceiver is further configured to send a result query request to the subsystem corresponding to the timeout sub-process after a predetermined period of time has passed;

The processor is further configured to perform an asymmetric reverse operation on the timeout subprocess and the subprocess before the timeout subprocess if the processing success result sent by the subsystem is received; if the subprocess is received If the processing result sent by the subsystem fails, an asymmetric reverse operation is performed on the sub-process before the timeout sub-process; if an unclear feedback of the processing result sent by the subsystem is received, it will execute after the predetermined interval , The step of sending a result query request to the subsystem corresponding to the timeout sub-process.
The electronic device according to claim 15, wherein the transceiver is further used for:

Count the number of processing requests received;

If the number of processing requests is greater than the traffic warning value, an warning message is sent to the invoking system, and the time for feedback of response messages to the invoking system is controlled.
The electronic device according to any one of claims 15 to 20, wherein the transceiver is further used for:

Receive response messages fed back by the subsystem;

Format the response message to obtain a response message;

Sending the response message to the calling system.
A non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to make the computer execute the method of any one of claims 1-7 .