CN115098463A

CN115098463A - Service message processing method, device, equipment, medium and product

Info

Publication number: CN115098463A
Application number: CN202210708874.7A
Authority: CN
Inventors: 王诗章; 杨玉新; 贝飞; 胡威予
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2022-06-22
Filing date: 2022-06-22
Publication date: 2022-09-23

Abstract

The application relates to a method, a device, equipment, a medium and a product for processing service messages, relating to the field of financial science and technology or other related fields, wherein the method comprises the following steps: if the target service message is monitored to be in an abnormal state in the target service operation link, acquiring an error report log of the target service message; acquiring a first state conversion scheme corresponding to the target service message from a preset state conversion strategy library according to an error report log of the target service message; and based on the first state conversion scheme, converting the processing state of the target service message in the target service operation link. The target service operation link is any service operation link of the target service message in the service processing link. The method and the device monitor the processing state of the service message in real time, and match the first state conversion scheme from the state conversion strategy library so as to automatically process the target service message in the abnormal state, ensure the accuracy of processing the service message and improve the processing efficiency of the abnormal service event.

Description

Service message processing method, device, equipment, medium and product

Technical Field

The present application relates to the field of financial technology or other related fields, and in particular, to a method, an apparatus, a device, a medium, and a product for processing a service packet.

Background

The business systems in the bank are numerous, and a plurality of business systems are mutually associated, so that the derived transaction scene is also complex. In a complex business scenario, a transmission packet of a transaction may have various states due to the reliability of a business system, the stability of a network, the availability of a called transaction service, and the like. The intermediate message means that the transaction is not closed-loop completed, and needs to be processed to complete the conversion from the intermediate state to the final state.

However, the complexity of the service scenario results in the diversity of message states, and each transaction message may have three states of successful processing, failed processing and unknown processing after passing through one processing unit, and the transaction message may generate various message states after the multiple processing units sequentially circulate. The more processing units that are circulated, the more complex the message state. Moreover, the number of transactions in a banking system is huge, the service call volume behind the system can reach millions, and a large amount of transaction data must be guaranteed to be processed timely and efficiently. Therefore, it is necessary to process the transaction service as soon as possible so that the transaction message is in the final state.

Therefore, how to quickly respond to various transaction events in a complex service scene, complete the conversion of the transaction message from the intermediate state to the final state, and improve the transaction service processing efficiency becomes a technical problem to be solved urgently.

Disclosure of Invention

Therefore, it is necessary to provide a service packet processing method, an apparatus, a device, a medium, and a product, which can quickly respond to various transaction events in a complex service scenario and complete the transition of a transaction packet from an intermediate state to a final state, in view of the above technical problems.

In a first aspect, the present application provides a method for processing a service packet, where the method includes:

if the target service message is monitored to be in an abnormal state in a target service operation link, acquiring an error log of the target service message; the target service operation link is any service operation link of the target service message in the service processing link;

acquiring a first state conversion scheme corresponding to the target service message from a preset state conversion strategy library according to an error report log of the target service message;

and based on the first state conversion scheme, converting the processing state of the target service message in the target service operation link.

In one embodiment, acquiring an error log of a target service packet includes:

acquiring service calling information of a target service message in a service processing link;

acquiring message information of a target service message in a target service operation link according to the service calling information;

and acquiring an error report log of the target service message according to the message information.

In one embodiment, the obtaining message information of the target service message according to the service invocation information includes:

determining an abnormal message retrieval statement according to the service calling information;

and acquiring the message information of the target service message from the service log database according to the abnormal message retrieval statement.

In one embodiment, acquiring an error log of a target service packet according to packet information includes:

screening key log information from the message information according to the abnormal identifier;

and cleaning the key log information according to a preset log cleaning strategy to obtain an error report log of the target service message.

In one embodiment, acquiring a first state transition scheme corresponding to a target service packet from a preset state transition policy library according to an error log of the target service packet includes:

analyzing and processing an error report log of the target service message, and acquiring abnormal error report information of the target service message; the abnormal error reporting information comprises an error reporting scene, an error reporting object and an error reporting reason;

and determining a first state conversion scheme corresponding to the target service message in a state conversion strategy library according to the abnormal error reporting information of the target service message.

In one embodiment, analyzing and processing an error report log of a target service packet to obtain abnormal error report information of the target service packet includes:

vectorizing an error log of the target service message to obtain a plurality of real word vectors corresponding to the target service message;

and inputting a plurality of real word vectors corresponding to the target service message into a preset abnormal prediction model, and determining abnormal error reporting information of the target service message through the abnormal prediction model.

In one embodiment, based on the first state transition scheme, the method for performing transition processing on the processing state of the target service packet in the target service operation link includes:

resetting the processing state of the target service message in the target service operation link to a to-be-processed state;

carrying out re-operation processing on the target service message in a target service operation link according to the first state conversion scheme;

and according to the result of the re-operation processing, converting the processing state of the target service message in the target service operation link.

In one embodiment, the converting the processing state of the target service packet in the target service operation link according to the result of the re-operation processing includes:

if the re-operation processing result is successful, the processing state of the target service message in the target service operation link is converted so as to update the processing state of the target service message in the target service operation link to be a successful state;

if the re-operation processing result is processing failure, writing the target service message into a preset abnormal event registration table; the abnormal event registration table is used for indicating a user to carry out conversion processing on the processing state of the target service message in the target service operation link.

In one embodiment, the method further comprises:

acquiring a second state conversion scheme which is manually input according to a target service operation link in which a target service message in an abnormal event registry is positioned;

and according to the second state conversion scheme, converting the processing state of the target transaction processing message in the service operation link so as to update the processing state of the target service message in the target service operation link to a successful state.

In one embodiment, the method further comprises:

and adding the abnormal error reporting information of the target service message in the abnormal event registration table and the second state conversion scheme into the state conversion strategy library.

In one embodiment, if the target service operation link is a resource deduction service link, the re-operation processing is resource deduction operation; if the target service operation link is a report service link, then the re-operation processing is report operation; if the target service operation link is a resource adding service link, performing re-operation processing as resource adding operation; and if the target service operation link is an abnormal resource cancellation service link, the re-operation is processed into abnormal resource cancellation operation.

In one embodiment, the method further comprises:

if the processing state of the resource deduction operation is deduction success, a report sending operation is carried out on the target service message;

if the processing state of the sending operation is the sending failure, executing resource adding operation on the target service message;

and if the processing state of the resource deduction operation is that the deduction is unknown, executing abnormal resource cancellation operation on the target service message.

In a second aspect, the present application further provides a service packet processing apparatus, where the apparatus includes:

the log acquisition module is used for acquiring an error report log of the target service message if the target service message is monitored to be in an abnormal state in a target service operation link; the target service operation link is any service operation link of the target service message in the transaction link;

the first scheme acquisition module is used for acquiring a first state conversion scheme corresponding to the target service message from a preset state conversion strategy library according to the error report log of the target service message;

and the first state conversion module is used for converting the processing state of the target service message in the target service operation link based on the first state conversion scheme.

In a third aspect, the present application further provides a computer device, where the computer device includes a memory and a processor, where the memory stores a computer program, and the processor implements the steps of any one of the method embodiments in the first aspect when executing the computer program.

In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of any of the method embodiments of the first aspect.

In a fifth aspect, the present application also provides a computer program product comprising a computer program that, when executed by a processor, performs the steps of any of the method embodiments of the first aspect.

According to the service message processing method, the device, the equipment, the medium and the product, if the computer equipment monitors that the target service message is in an abnormal state in the target service operation link, an error report log of the target service message is obtained; acquiring a first state conversion scheme corresponding to the target service message from a preset state conversion strategy library according to an error report log of the target service message; and based on the first state conversion scheme, converting the processing state of the target service message in the target service operation link. The target service operation link is any service operation link of the target service message in the service processing link. That is, according to the application, a computer device acquires a target service message in an abnormal state in any service operation link on a service processing link, and further obtains an optimal first state conversion scheme from a preset state conversion strategy library according to an error log of the target service message, so that the processing state of the target service message is converted in the abnormal state service operation link according to the first state conversion scheme, and a service is closed and the target service message is in a final state. Therefore, through the real-time monitoring of the abnormal state and the preset state conversion strategy library, the computer equipment can automatically process the target service message in the abnormal state through the matched state conversion scheme, the conversion of the target service message from the intermediate state to the final state is completed, the accuracy of the processing of the service message is ensured, and meanwhile, the processing efficiency of the abnormal service event can be improved.

Drawings

Fig. 1 is a schematic flow chart of a service packet processing method in an embodiment;

fig. 2 is a schematic diagram illustrating a service message transmission flow in an embodiment;

FIG. 3 is a flowchart illustrating the process of obtaining an error log according to one embodiment;

FIG. 4 is a schematic diagram illustrating a first example of a state transition scenario acquisition process;

fig. 5 is a schematic flow chart illustrating a process of performing state transition processing on a target service packet in an embodiment;

FIG. 6 is a diagram illustrating a message state transition flow of a resource deduction service link in an embodiment;

fig. 7 is a schematic diagram illustrating a message state transition flow of an abnormal resource revocation service link in an embodiment;

FIG. 8 is a diagram illustrating a message state transition flow of an abnormal resource revocation service link in another embodiment;

FIG. 9 is a diagram illustrating a message state transition flow of a messaging service link in an embodiment;

FIG. 10 is a diagram illustrating a message state transition flow of a resource addition service link in an embodiment;

fig. 11 is a schematic flow chart of a service packet processing method in another embodiment;

fig. 12 is a block diagram of a service packet processing apparatus in an embodiment;

FIG. 13 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The banking field has a plurality of business systems, and a plurality of business systems are related to each other, so that the derived business scene is relatively complex. For example, a clearing domain service has the characteristics of multiple associated applications and long transaction links, a service system experienced by a typical clearing service can be extracted into a link structure of 'channel application-product application-clearing application', various applications are mutually invoked according to an associated relationship, a plurality of service groups with a loose coupling relationship are deployed in the applications according to an architecture design, and each service group provides different types of background distributed services.

For example, a closed loop of a transaction service is processed by a "application-group-service" model that flows between multiple service systems. Taking a personal cross-bank remittance transaction scenario as an example, the transaction link can be described as "personal internet bank-personal remittance-personal settlement-cross-bank routing-cross-bank clearing-size/internet bank interconnection-pedestrian front" according to the application dimension. When a transaction link is executed, the former application calls a specific distributed service of the latter application according to a transaction scene and functional requirements, the latter application receives a call request and then carries out response processing in real time, and a processing result is returned to the former application, so that closed loop of one business in the transaction scene is realized.

However, in a complex service scenario, due to the reliability of the system, the stability of the network, and the availability of the service, the service packet sent in the service processing link may have a variety of states, including an intermediate state and a final state. The intermediate state service message means that the service does not close the loop and needs to be further processed to complete the conversion of the service message from the intermediate state to the final state.

However, the service packet generally has the following two characteristics:

(1) diversity of message states.

The complexity of a service scene causes the diversity of message states, a service message of a service may have three states of successful processing, failed processing and unknown processing every time the service message passes through a service operation link, and the service message flows in sequence in a plurality of service operation links to generate a plurality of message states. That is, the more processing units that are circulated, the more complicated the message state.

(2) And the timeliness of message processing.

The transaction data in the business scene of the banking system is huge, for example, the daily average transaction amount in the business scene of a personal client is hundreds of thousands, the service call amount behind the business scene can reach millions, and a large amount of transaction data needs to be processed timely and efficiently.

Based on the two reasons, the application provides a service message processing method, a service message processing device and computer equipment aiming at service messages with various states, so as to monitor the service messages, convert intermediate-state service messages under different scenes, quickly respond to various abnormal service events and improve the service processing efficiency.

The service message processing method provided by the application can be applied to a service message processing device, and the service message processing device can be realized to be part or all of computer equipment in a software, hardware or software and hardware combined mode.

Wherein the computer device may be a server or a terminal. The server in the embodiment of the present application may be, but is not limited to, at least one independent server, distributed server, cloud server, and server cluster; the terminal in the embodiment of the application can be intelligent equipment such as a smart phone, a personal computer, a notebook computer and portable wearable equipment. In the following method embodiments, the execution subject is a computer device as an example.

Next, the technical solutions of the embodiments of the present application and how to solve the above technical problems will be specifically described by embodiments with reference to the accompanying drawings. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. It is to be understood that the embodiments described are only some of the embodiments of the present application and not all of them.

In an embodiment, as shown in fig. 1, a method for processing a service packet is provided, which includes the following steps:

step 110: if the target service message is monitored to be in an abnormal state in the target service operation link, acquiring an error report log of the target service message; the target service operation link is any service operation link of the target service message in the service processing link.

The target service packet is a service packet of a service processed on any one or more service processing links, that is, the number of the target service packets may be one or multiple, and this embodiment does not limit this. The abnormal state refers to that the recording state of the target service message in the service state table is unknown or failed after the target service operation link executes corresponding operation on the service.

Further, the service operation links involved in the traffic processing link may include: a resource deduction service link, a report service link, a resource addition service link and an abnormal resource cancellation service link.

For the service message of the dynamic account class, the reporting flow is as shown in fig. 2, and the complete reporting flow can be described as follows: after receiving a service message of the account-based service, the bank system firstly performs resource deduction processing, and realizes product side resource deduction operation of the service scene by calling a resource deduction service interface of the product application. If the deduction is unknown, switching to an abnormal resource cancellation link; if the resource deduction service link is operated normally and corresponding resources are successfully deducted, sending a report, wherein the clearing application is responsible for forwarding the service message, and if the report is successful, ending the processing flow; if the report fails, the resource addition processing needs to be performed on the deducted service message, the resource addition service on the application side is called, the resource addition service further calls the abnormal resource cancellation service applied on the product side, and the deduction operation performed on the deducted resource is cancelled.

For example, in the dynamic account type transaction service, the resource deduction service is a deduction service, the resource addition service is a reimbursement service, and the abnormal resource cancellation operation is an account cancellation service.

In this example, for an account-based transaction link, if the target service operation link is an account-based deduction service link, the abnormal state of the target service packet in this link includes that account-based deduction is unknown and account-based deduction is failed; if the target service operation link is a report service link, the abnormal state of the target service message in the link comprises report unknown and report failure; if the target service operation link is a service link for canceling the account doubt, the abnormal state of the target service message in the link comprises cancellation failure and unknown cancellation; if the target service operation link is an account checking service link, the abnormal state of the target service message in the link comprises account checking failure and account checking unknown.

It should be noted that, in this step 110, the computer device may monitor a service packet in any service in the banking system, and the monitoring mode may be real-time monitoring or periodic monitoring, which is not limited in this embodiment.

In addition, the error log of the target service packet may be original log information obtained from a service log database, or may be obtained by preprocessing the original log information, which is not limited in this embodiment.

As an example, the error log may include information of a target service operation link in which the target service packet is currently located, an accounting state of the target service packet, and an error reason.

Step 120: and acquiring a first state conversion scheme corresponding to the target service message from a preset state conversion strategy library according to the error log of the target service message.

The state transition strategy library is constructed in advance and used for storing some state transition schemes given by experts aiming at service messages in abnormal states. In other words, the state transition policy library stores abnormal error reporting information and state transition schemes of a plurality of service messages, and the corresponding state transition schemes are adopted to perform transition processing on the states of the service messages, so that the service is in a closed loop and a final state of the service messages.

In one possible implementation manner, the implementation process of step 120 may be: analyzing and processing the error report log of the target service message, and extracting abnormal error report information of the target service message from the error report log; and then matching a corresponding first state conversion scheme for the target service message from the state conversion strategy library according to the abnormal error reporting information of the target service message.

As an example, the abnormal error reporting information at least includes an error reporting scene, an error reporting object, and an error reporting reason, and of course, the abnormal error reporting information extracted from the error reporting log may further include other information for improving the matching accuracy of the state transition scheme, which is not limited by this embodiment.

Step 130: and based on the first state conversion scheme, converting the processing state of the target service message in the target service operation link.

In one possible implementation manner, the implementation procedure of step 130 may be: after the computer equipment acquires the first state conversion scheme corresponding to the target service message, the computer equipment runs the service message modification script in the first state conversion scheme, and performs re-operation processing on the target service operation link so as to perform conversion processing on the processing state of the target service message in the target service operation link.

If the processing is successful, changing the processing state of the target service operation link of the target service message into a successful state; and if the processing fails, the processing state of the target service message in the target service operation link is still in an abnormal state.

In the service message processing method, if the computer equipment monitors that the target service message is in an abnormal state in a target service operation link, an error report log of the target service message is obtained; acquiring a first state conversion scheme corresponding to the target service message from a preset state conversion strategy library according to an error report log of the target service message; and based on the first state conversion scheme, converting the processing state of the target service message in the target service operation link. The target service operation link is any service operation link of the target service message in the service processing link. That is, according to the application, a computer device acquires a target service message in an abnormal state in any service operation link on a service processing link, and further obtains an optimal first state conversion scheme from a preset state conversion strategy library according to an error log of the target service message, so that the processing state of the target service message is converted in the abnormal state service operation link according to the first state conversion scheme, and a service is closed and the target service message is in a final state. Therefore, through the real-time monitoring of the abnormal state and the preset state conversion strategy library, the computer equipment can automatically process the target service message in the abnormal state through the matched state conversion scheme, complete the conversion of the target service message from the intermediate state to the final state, and improve the processing efficiency of the abnormal service event while ensuring the accuracy of the service message processing.

Based on the above embodiment, when monitoring the processing state of the target service packet in each service operation link in the service processing link, the present application may monitor the sending state of the service packet from three aspects according to the sequence of the precedence dependence: and service calling information, message information and log information are used for acquiring the target service message of which the service operation link is in an abnormal state. Next, a process of obtaining an error log of a target service packet is explained through an embodiment corresponding to fig. 3.

In an embodiment, as shown in fig. 3, the implementation process of obtaining the error log of the target service packet in step 110 may include the following steps:

step 310: and acquiring service calling information of the target service message in the service processing link.

In this step, for a specific service scenario, an interface of a Host Monitoring Protocol (HMP) of the holographic Monitoring system is called in a hypertext Transfer Protocol (HTTP) manner, so as to obtain a specific service calling condition.

The holographic monitoring system may provide service invocation information such as invocation amount change, success/failure times, average/maximum time consumption, and the like of each application service on one service processing link according to link ID (i.e., Trace ID) information of the service processing link.

Further, whether all the business processing links are communicated can be detected according to the service calling information, and if the interruption exists, the abnormal service operation link of the business processing link can be positioned at which distributed service position of which application occurs.

Step 320: and acquiring message information of the target service message in a target service operation link according to the service calling information.

In one possible implementation manner, the implementation procedure of step 320 may be: determining an abnormal message retrieval statement according to the service calling information; and acquiring the message information of the target service message from the service log database according to the abnormal message retrieval statement.

The service log database can be a MySQL database, and the abnormal message retrieval statement is a monitoring SQL statement and is used for acquiring message information from the service log database. Further, the acquired message information includes field information such as a message state, an error code, an error reason, and the like of the target service message.

As an example, first, according to the service invocation information, a specific application system and a corresponding service group of the transaction scenario interrupt are located, and by using key information (e.g., a service event number) in the service as an index condition, a monitoring SQL statement is generated. Then, according to the mapping relation configuration table of the application group-database, the field information such as the service message state, error reporting code, error reason and the like registered by the service log table and/or the abnormal registration table and the like is inquired in the specific service log database. And finally, extracting message information of which the service message state is 'failure to send' or 'unknown to send', and obtaining the message information of the target service message.

Step 330: and acquiring an error report log of the target service message according to the message information.

In one possible implementation manner, the implementation procedure of step 330 may be: screening key log information from the message information according to the abnormal identifier; and cleaning the key log information according to a preset log cleaning strategy to obtain an error log of the target service message.

Specifically, based on message information with a message state of "failure to send" or "unknown to send", such as key information related to transaction event numbers and the like, as a screening condition, an interface of a Distributed Service Framework (DSF) log center is called, and log information of the Service is searched and located from a Docker container of a corresponding application Service group. Detailed ERROR reporting information after service calling failure or unknown is recorded in the normalized log, and log paragraphs with the ERROR mark are collected in real time to obtain key log information.

Further, according to the alarm level policy and the user-defined policy of the user (for example, special attention needs to be paid to some programs and some tables), the collected key log information is cleaned, invalid log information is cleaned, and an error log of the target service message is obtained.

In this embodiment, the message information of the target service message in the target service operation link is obtained through the service call information of the target service message in the service processing link, and then the error log of the target service message is obtained according to the message information. Therefore, the processing state of each service operation link on the message transmission link is obtained, the abnormal state of the message is positioned, and the comprehensive and accurate error log is obtained.

In an embodiment, as shown in fig. 4, the process of obtaining the first state transition scheme corresponding to the target service packet from the preset state transition policy library in step 120 includes the following steps:

step 410: and analyzing and processing the error report log of the target service message to acquire abnormal error report information of the target service message.

In one possible implementation manner, the implementation process of step 410 may be: vectorizing an error log of the target service message to obtain a plurality of real word vectors corresponding to the target service message; and inputting a plurality of real word vectors corresponding to the target service message into a preset abnormal prediction model, and determining abnormal error reporting information of the target service message through the abnormal prediction model.

The process of vectorizing the error report of the target service message to obtain the real word vector may include the following steps:

s1: and constructing an error-reporting dictionary, wherein the error-reporting dictionary calls error-reporting vocabularies corresponding to various error-reporting factors such as service, return information, error prompt information and the like.

Taking a sample error reporting log (2022-06-0110: 11:37.575INFO businessLogger- > calling RPC service IbpsCredMessageSendservice. invoke success, and returning information { "apptat" { "transosk": 1, "return _ msg": field length error pobnknnam, not more than 60, value: "," return _ code ":5399} }) as an example, because the occurrence of each word in the error reporting log is independent, the words occurring in the error reporting log can be converted into a bag word matrix.

Specifically, first, the category C1, C2, … …, Cm (such as Remote Procedure Call (RPC) Call, return information, error, return _ code) of each error object in the error log of the statistical sample is counted in the proportion P (C) to which the training sample belongs _i ) Since the normalization of the sample log ensures that the prior probability of the class can be P (C) _i )＝S _i Calculation of/S, where S _i Is of the class C _i And S is the total number of training samples. Then, according to bayes' theorem, for sample log X to be classified, only P (X | C) is needed since P (X) is constant for all classes _i )P(C _i ) And (4) the maximum is obtained. Furthermore, because the attribute values of the split sample log are mutually condition-independent, that is, there is no dependency relationship between the attributes, it is very suitable to make a naive assumption that the class conditions are independent, and then the original P (X | C) with large cost is caused by more attributes _i ) That is, joint probability statistics can be employed, and finally X will be assigned to the maximum P (X | C) _i )P(C _i ) Class C of _i In (1).

S2: based on an error-reporting dictionary, single-text word frequency (TF) -Inverse text frequency Index (IDF) is adopted to perform clustering analysis on error-reporting logs.

Because the bag-of-words matrix is expressed in a 0 and 1 mode, the importance of different words in the error log can not be accurately expressed by using the bag-of-words matrix, and therefore in the practical application process, the bag-of-words matrix is replaced by the word frequency of the words in the words.

Further, an IDF technology is adopted to remove invalid high-frequency words and stop words in the error log. For example, the date, time, punctuation marks and the like are found in the training abnormal prediction model, which is not useful for classifying the error-reporting logs, but when the word frequency is high, the classification efficiency is affected, and the word frequency should be eliminated.

S3: and determining a plurality of real word vectors corresponding to the target service message according to the clustering analysis result of the error report log.

In a possible implementation mode, splitting each word according to a unified rule for each input row of error-reporting logs, reserving information such as word length, number position, letter combination and the like, automatically carrying out weight adjustment on the split words in the process, and finally obtaining real word vectors with the same size as error-reporting words in an error-reporting dictionary, wherein the real word vectors are composed of TF-IDF values, and the size of each dimension in the real word vectors represents the contribution degree of each real word to the classification of the row of logs.

Further, the real word vector after being converted into calculable numerical data, and the formula of the prior correlation calculation is changed into weighted sum from simple sum of word frequency, namely TF ₁ *IDF ₁ +TF ₂ *IDF ₂ +...+TF _n *IDF _n 。

In addition, after the above steps S1-S3 are performed, after the correlation is calculated for the words in the error report log, a plurality of real word vectors corresponding to the target service packet may be input into the abnormal prediction model, and classified and identified by the abnormal prediction model, so as to determine the abnormal error report information of the target service packet.

Alternatively, the anomaly prediction model may be a naive bayes model with improved laplace smoothing.

The abnormal error reporting information comprises an error reporting scene, an error reporting object and an error reporting reason. The error reporting object can be RPC call, return information, error, return _ code, etc.; further, an error reporting scene is determined according to the error reporting code of the identified error reporting object, for example, service call timeout, table update failure, no configuration information read, parameter setting overrun and the like.

As an example, after analyzing and processing an error log (2022-06-0110: 11:37.575INFO businessLogger- > calling RPC service IbpsCredMessageSendservice. invoke success and returning information { "appstat": { "transosk": 1, "return _ msg": field length error pobnkanam, not more than 60, and having a value of "return _ code":5399 }), determining that an error object may be a service name IbpsCredMessageSendService, an error scene is a credit issuing service, and the error reason is a field length error.

Step 420: and determining a first state conversion scheme corresponding to the target service message in a state conversion strategy library according to the abnormal error reporting information of the target service message.

The state transition strategy library stores abnormal error reporting information and state transition schemes of a plurality of service messages. By adopting the corresponding state conversion scheme, the state of the service message can be converted, so that the service processing is closed and the service message is in a final state.

In one possible implementation manner, the implementation procedure of step 420 may be: according to the abnormal error reporting information of the target service message, traversing the abnormal error reporting information corresponding to each state conversion scheme in the state conversion strategy library, and determining the state conversion scheme corresponding to the abnormal error reporting information with the highest matching degree in the state conversion strategy library as a first state conversion scheme corresponding to the target service message.

In this embodiment, the error report log of the target service packet is analyzed and processed to determine abnormal error report information of the target service packet, and then the first state transition scheme corresponding to the target service packet is determined from the state transition policy library according to the abnormal error report information.

In an embodiment, as shown in fig. 5, the implementation process of performing conversion processing on the processing state of the target service packet in the target service operation link based on the first state conversion scheme in the step 130 includes the following steps:

step 510: and resetting the processing state of the target service message in the target service operation link to be a to-be-processed state.

The target service operation link comprises a resource deduction service link, a report service link, a resource addition service link and an abnormal resource cancellation service link.

It should be noted that, when the target service packet includes multiple service packets, after the processing state of the target service packet is set to the to-be-processed state, multiple background batch processing tasks may be introduced, the to-be-processed records are periodically scanned by means of the batch processing tasks, corresponding batch processing operations are executed, and the automatic processing of the intermediate state packet is completed in quasi-real time according to a certain frequency.

As an example, referring to table 1 below, the processing state of a transmission packet of a service includes a processing state, a processing sub-state, a batch processing task state, and a record state.

TABLE 1

Step 520: and carrying out re-operation processing on the target service message in a target service operation link according to the first state conversion scheme.

It should be noted that, if the target service operation link is a resource deduction service link, the re-operation processing is a resource deduction operation; if the target service operation link is a report service link, the re-operation processing is report operation; if the target service operation link is a resource adding service link, performing re-operation processing to be resource adding operation; and if the target service operation link is an abnormal resource cancellation service link, the re-operation is processed into abnormal resource cancellation operation.

As an example, for dynamic account transaction, resource deduction is account deduction, resource addition is account rush-out, abnormal resource revocation is suspected account revocation, and a reporting operation is performed when account deduction is successful.

In the step, in a target service operation link, a first state conversion scheme is adopted to perform re-operation processing on the target service message in the abnormal state. Namely, the operation corresponding to the target service operation link is executed again to change the abnormal state of the target service message in the link.

Step 530: and according to the result of the re-operation processing, converting the processing state of the target service message in the target service operation link.

In one possible implementation manner, the implementation procedure of step 530 may be: if the re-operation processing result is successful, converting the processing state of the target service message in the target service operation link so as to update the processing state of the target service message in the target service operation link into a successful state; and if the result of the heavy operation processing is processing failure, writing the target service message into a preset abnormal event registry.

The abnormal event registry is used for indicating a user to perform conversion processing on the processing state of the target service message in the target service operation link.

Further, under the condition of successful processing, the target service message enters the next service operation link from the current target service operation link until the service completes the closed loop and the target service message is in the final state.

That is, after the conversion processing, if the processing state of the resource deduction operation is successful, the sending operation is performed on the target service message; if the processing state of the sending operation is the sending failure, executing resource adding operation on the target service message; and if the processing state of the resource deduction operation is that the deduction is unknown, executing abnormal resource cancellation operation on the target service message.

In this embodiment, according to the first state transition scheme, when the target service operation link performs transition processing on the processing state of the target service packet, the processing state of the target service operation link is reset to the to-be-processed state. Then, all the target service messages in the to-be-processed state are subjected to re-operation processing in a batch processing task mode, so that the processing state of the target service messages is converted through the re-operation processing result, and the processing efficiency of the target service messages is improved.

Based on the above table 1 and the embodiments of the methods, and with reference to fig. 6 to 10, the resource deduction service link, the abnormal resource cancellation service link, the report service link, and the resource addition service link are taken as target service operation links for example, and the processing state resetting, the re-operation processing, and the conversion processing process of the processing state of the target service packet are explained.

In an embodiment, as shown in fig. 6, the process of performing conversion processing on the processing state of the target service packet in the resource deduction service link is as follows: firstly, the processing state of the target service message in the abnormal state is reset, and the record state of the target service message is updated to be 'to be deducted-3', the processing state is 'accepting-2', the processing sub-state is 'resource deduction-1' and the batch processing task state is 'no retry-4' in a service state table in a database. And then, calling the resource deduction service through HTTP, initiating a resource deduction operation, and receiving a result returned by the resource deduction service in real time.

If the resource deduction service is successfully called and the resource deduction result is failure, the record state of the target service message is updated to be deduction failure-1 in the service state table, and the service processing flow is ended. If the resource deduction service is successfully called and the resource deduction result is successful, updating the record state of the target service message to be deduction success-0 in the service state table, and executing a sending operation on the target service message.

If the calling of the resource deduction service fails and the resource deduction result is unknown, the target service message is registered in the retry list, the batch processing task state of the target service message is updated to be 'to retry-0' in the service state list, and the record state is 'deduction unknown-2'.

Further, the background batch processing task traverses and circularly scans the service message record with the batch processing state of 'waiting for retry-0' in the retry table according to a certain frequency, triggers a multiple-time re-operation processing mechanism, and repeatedly calls the resource deduction service.

For example, if the service invocation interval is 5 minutes, the threshold of the retry time is 6 times, and the result of the resource deduction operation executed by the target service message is still unknown after 6 retries are reached, the batch processing task state of updating the target service message in the service state table is "retry failure-2", and the record state is "unknown-2" for deduction.

After the processing, for the service message with the processing sub-state of resource deduction-1 and the batch processing task state of retry failure-2 in the service state table, and the status of deduction unknown-2 is recorded, the abnormal resource cancellation service is called to perform abnormal resource processing.

In one embodiment, the abnormal resource revocation service may include an abnormal resource query sub-service and a resource deduction revocation sub-service.

As shown in fig. 7, the implementation process of calling the abnormal resource query sub-service and querying the service packet recorded in the service state table in the state of "deduction unknown-2" is as follows: firstly, the processing state of the target service message in the abnormal state is reset, the record state of the target service message is updated to be 'deduction unknown-2', the processing state is 'accepting in-2', the processing sub-state is 'resource deduction-1', and the batch processing task state is 'retry failure-2' in a service state table in a database. Then, the background batch processing task circularly traverses and scans the service state table, captures the transaction records of the state into a to-be-processed result set, and initiates abnormal resource query when every N records form a group.

Furthermore, a request message in a JSON format is spliced according to information in a background database, abnormal resource query sub-service processing in the liquidation application is requested to be called, and different operations are carried out according to different query result conditions.

If the inquiry result is deduction success, updating the record state of the target service message in the service state table to be 'account deduction success-0', and executing automatic sending operation; if the query result is deduction failure, updating the record state of the target service message in the service state table to be deduction failure-1, finishing the target service acceptance and failing the acceptance; if the query result is unknown and the query result is still unknown after the number of query re-operations reaches the threshold, updating the record state 'to be cancelled-11' of the target service message in the service state table.

Further, for the service message recorded in the service state table in the state of "to-be-revoked-11", the resource deduction revocation operation is synchronously initiated, and the processing is performed by the resource deduction revocation sub-service.

As shown in fig. 8, the implementation process of invoking the resource deduction revocation sub-service and performing resource deduction revocation on the service packet recorded in the service state table as "to-be-revoked-11" includes: firstly, the processing state of the target service message in the abnormal state is reset, the record state of the target service message is updated in a service state table in a database to be ' to-be-cancelled ' -11 ', the processing state is ' receiving-2 ', the processing sub-state is ' abnormal resource cancellation-4 ', and the batch processing task state is ' no retry-4 '.

It should be noted that the resource deduction revocation sub-service also relies on the background batch processing task to scan the service state table at regular time, group the service message whose recorded state is to be revoked, then call the resource deduction revocation sub-service, and initiate the resource deduction revocation operation. Further, the state of the service message is updated in the service state table according to the revocation result returned by the resource deduction revocation sub-service.

If the revocation is successful, the record state of the target service message is updated to be 'revocation success-8' in the service state table, and the target service acceptance is finished; if the revocation fails, updating the record state of the target service message in the service state table to be 'revocation failure-9', and failing to accept the target service; if the revocation is unknown and the number of times of the revocation and the re-operation reaches the threshold value, the result of the abnormal resource deduction operation is still unknown, the record state of the target service message is updated to 'revocation unknown-11' in the service state table, the target service acceptance fails, the target service message is recorded in the abnormal event registration table, and manual processing is switched.

In an embodiment, as shown in fig. 9, the process of performing conversion processing on the processing state of the target service packet in the messaging service link is as follows: firstly, the processing state of the target service message in the abnormal state is reset, the record state of the target service message is updated in a service state table in a database to be 'deduction success-0', the processing state is 'acceptance-2', the processing sub-state is 'transmission-3', and the batch processing task state is 'retry-free-4'.

And calling a report service of the clearing application to report the service message which is in the processing sub-state of report-3 and the recording state of account deduction success-0.

If the message is successfully sent, updating the processing state of the target service message to be 'acceptance success-0' in the service state table; if the sending fails, updating the record state of the target service message to be 'to be added-7' in the service state table, and calling the resource adding service for further processing; if the report is unknown, the batch processing task state of the target service message is updated to be-0 to be retried in the service state table, and the target service message is registered in the retry table.

Further, the background batch processing task regularly scans the report sending records to be retried in the retry table, and calls a report sending service in a grouping mode to carry out report sending and re-processing operation.

And in the sending service link, after the sending re-operation processing is carried out on the target service message, the sending result is inquired. If the query result is that the message is successfully sent, updating the processing state of the target service message to be 'acceptance success-0' in the service state table; if the query result is that the report fails, updating the record state of the target service message to be 'to be added-7' in the service state table, and calling the resource addition service for further processing; if the query result is that the transmission is unknown, the processing task state of the target service message is updated to be 'to-be-retried-0' in the service state table, and background retries are initiated.

Similarly, if the report re-sending operation times reach the threshold value, and the report query result is still unknown, the processing state of the target service message, namely 'acceptance failure-1' and 'retry failure-2' is updated in the service state table, the target service message is recorded in the abnormal event registration table, and manual processing is switched.

In an embodiment, as shown in fig. 10, the process of performing conversion processing on the processing state of the target service packet in the resource addition service link includes: firstly, the processing state of the target service message in the abnormal state is reset, the record state of the target service message is updated to be 'to be added-7', the processing state is 'receiving-2', the processing sub-state is 'sending-3' and the batch processing task state is 'no retry-4' in a service state table in a database.

And aiming at the service message with the record state of 'to be added-7', background batch processing operation scans the message data in the state at regular time, and calls a resource adding service to initiate resource adding operation.

If the resource adding result is successful, updating the record state of the target service message in a service state table to be 'adding success-4'; if the resource adding result is failure, updating the record state of the target service message in the service state table to be 'adding failure-5', recording the target service message in an abnormal event registration table, and waiting for manual intervention processing; if the resource adding result is unknown, the recording state of the target service message is updated to be 'adding unknown-6' in the service state table.

Further, for the service message with the record state of "add unknown-6", an abnormal resource query sub-service can be called to perform query of deduction of abnormal resources.

The number of times of the re-operation and the interval of the abnormal resource query may be preset, for example, the number of times of the re-operation is 6, and each time is 5 minutes.

Further, if the abnormal resource query result is that the addition is successful, updating the record state of the target service message in the service state table to be 'successful-4' of addition; if the abnormal resource query result is that the addition fails, the record state of the target service message is updated to be 'addition failure-5' in the service state table, the target service acceptance fails, the target service message is recorded in the abnormal event registration table, and manual processing is switched.

Similarly, if the number of times of querying the abnormal resource reaches the threshold value, and the query result is still unknown, the record state of the target service message is updated in the service state table, namely 'add unknown-6', the target service message is recorded in the abnormal event registration table, and manual processing is switched.

In an embodiment, as shown in fig. 11, for an abnormal event registry in any of the above method embodiments, the present application further provides another service packet processing method, including the following steps:

step 1110: and acquiring a second state conversion scheme manually input according to a target service operation link where the target service message in the abnormal event registry is located.

The abnormal event registry is used for recording abnormal service messages which cannot be automatically processed and abnormal error reporting information of each abnormal service message, wherein the abnormal error reporting information comprises an error reporting scene, an error reporting object and an error reporting reason.

In one possible implementation manner, the implementation procedure of step 1110 may be: aiming at the target service message in the abnormal event registration table, the staff in the bank system analyzes and processes the abnormal message information of the target service message and determines a second state conversion scheme corresponding to the target service message in a target service operation link.

Step 1120: and according to the second state conversion scheme, converting the processing state of the target transaction processing message in the service operation link so as to update the processing state of the target service message in the target service operation link to a successful state.

In the step, according to a second state conversion scheme determined manually, the processing state of the target transaction processing message in the service operation link is converted, so that the service is in a closed loop and the service message is in a final state.

Optionally, in the step 1120, the method may further include: and adding the abnormal error reporting information of the target service message in the abnormal event registration table and the second state conversion scheme into the state conversion strategy library.

In this embodiment, for a target service operation link in which a target service packet in the abnormal event registry is located, a second state conversion scheme that is manually input is used to perform conversion processing on the processing state of the target transaction processing packet in the service operation link, so as to update the processing state of the target service packet in the target service operation link to a successful state, thereby effectively processing the target service packet, and enabling the target service packet to be in a final state.

It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.

Based on the same inventive concept, the embodiment of the present application further provides a service packet processing apparatus for implementing the service packet processing method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so specific limitations in one or more embodiments of the service packet processing device provided below may refer to the above limitations on the service packet processing method, and details are not described here.

In one embodiment, as shown in fig. 12, a service packet processing apparatus is provided, where the apparatus 1200 includes: a log obtaining module 1210, a scheme obtaining module 1220, and a state transition module 1230, wherein:

the log obtaining module 1210 is configured to obtain an error log of the target service packet if the target service packet is monitored to be in an abnormal state in the target service operation link; the target service operation link is any service operation link of the target service message in the transaction link;

a first scheme obtaining module 1220, configured to obtain, according to the error report log of the target service packet, a first state transition scheme corresponding to the target service packet from a preset state transition policy library;

the first state transition module 1230 is configured to perform transition processing on a processing state of the target service packet in the target service operation link based on the first state transition scheme.

In one embodiment, the log obtaining module 1210 includes:

the service acquisition unit is used for acquiring service calling information of the target service message in the service processing link;

the message obtaining unit is used for obtaining message information of the target service message in a target service operation link according to the service calling information;

and the log acquiring unit is used for acquiring an error report log of the target service message according to the message information.

In one embodiment, the packet obtaining unit includes:

the retrieval subunit is used for determining an abnormal message retrieval statement according to the service calling information;

and the obtaining subunit is used for obtaining the message information of the target service message from the service log database according to the abnormal message retrieval statement.

In one embodiment, the log obtaining unit includes:

the information screening subunit is used for screening the key log information from the message information according to the abnormal identifier;

and the cleaning subunit is used for cleaning the key log information according to a preset log cleaning strategy to obtain an error log of the target service message.

In one embodiment, the scheme acquiring module 1220 includes:

the log analysis unit is used for analyzing and processing the error report log of the target service message and acquiring abnormal error report information of the target service message; the abnormal error reporting information comprises an error reporting scene, an error reporting object and an error reporting reason;

and the scheme matching unit is used for determining a first state conversion scheme corresponding to the target service message in the state conversion strategy library according to the abnormal error reporting information of the target service message.

In one embodiment, the log analysis unit includes:

the vectorization processing subunit is configured to perform vectorization processing on the error report log of the target service packet to obtain a plurality of real word vectors corresponding to the target service packet;

and the information determining subunit is used for inputting the plurality of real word vectors corresponding to the target service message into a preset abnormal prediction model, and determining abnormal error reporting information of the target service message through the abnormal prediction model.

In one embodiment, the state transition module 1230 includes:

the state resetting unit is used for resetting the processing state of the target service message in the target service operation link to a to-be-processed state;

the re-operation unit is used for performing re-operation processing on the target service message in a target service operation link according to the first state conversion scheme;

and the state conversion processing unit is used for converting the processing state of the target service message in the target service operation link according to the heavy operation processing result.

In one embodiment, the state transition processing unit includes:

the first conversion subunit is configured to, if the re-operation processing result is that the processing is successful, perform conversion processing on the processing state of the target service packet in the target service operation link to update the processing state of the target service packet in the target service operation link to a successful state;

the second conversion subunit is used for writing the target service message into a preset abnormal event registry if the heavy operation processing result is processing failure; the abnormal event registration table is used for indicating a user to carry out conversion processing on the processing state of the target service message in the target service operation link.

In one embodiment, the apparatus 1200 further comprises:

the second scheme acquisition module is used for acquiring a second state conversion scheme which is manually input according to a target service operation link where a target service message in the abnormal event registry is located;

and the second state conversion module is used for converting the processing state of the target transaction processing message in the service operation link according to a second state conversion scheme so as to update the processing state of the target service message in the target service operation link to a successful state.

In one embodiment, the apparatus 1200 further comprises:

and the adding module is used for adding the abnormal error reporting information of the target service message in the abnormal event registration table and the second state conversion scheme into the state conversion strategy library.

In one embodiment, the apparatus further comprises: if the processing state of the resource deduction operation is deduction success, a report sending operation is carried out on the target service message; if the processing state of the sending operation is the sending failure, executing resource adding operation on the target service message; and if the processing state of the resource deduction operation is that the deduction is unknown, executing abnormal resource cancellation operation on the target service message.

All or part of each module in the service message processing device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 13. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a method of service message processing. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on a shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the architecture shown in fig. 13 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

if the target service message is monitored to be in an abnormal state in a target service operation link, acquiring an error log of the target service message; the target service operation link is any service operation link of the target service message in the transaction link;

The implementation principle and technical effect of the computer device provided by the above embodiment are similar to those of the above method embodiment, and are not described herein again.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

if the target service message is monitored to be in an abnormal state in the target service operation link, acquiring an error report log of the target service message; the target service operation link is any service operation link of the target service message in the transaction link;

The implementation principle and technical effect of the computer-readable storage medium provided by the above embodiments are similar to those of the above method embodiments, and are not described herein again.

In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, performs the steps of:

The foregoing embodiments provide a computer program product, which has similar implementation principles and technical effects to those of the foregoing method embodiments, and will not be described herein again.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by hardware instructions of a computer program, which may be stored in a non-volatile computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), for example.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent application shall be subject to the appended claims.

Claims

1. A method for processing service messages is characterized in that the method comprises the following steps:

if the target service message is monitored to be in an abnormal state in a target service operation link, acquiring an error report log of the target service message; the target service operation link is any service operation link of the target service message in a service processing link;

acquiring a first state conversion scheme corresponding to the target service message from a preset state conversion strategy library according to the error log of the target service message;

2. The method according to claim 1, wherein the obtaining the error log of the target service packet comprises:

acquiring service calling information of the target service message in the service processing link;

acquiring message information of the target service message in the target service operation link according to the service calling information;

3. The method according to claim 2, wherein the obtaining the message information of the target service message according to the service invocation information includes:

and acquiring the message information of the target service message from a service log database according to the abnormal message retrieval statement.

4. The method according to claim 2, wherein the obtaining an error log of the target service packet according to the packet information comprises:

and cleaning the key log information according to a preset log cleaning strategy to obtain an error log of the target service message.

5. The method according to any one of claims 1 to 4, wherein the obtaining, according to the error log of the target service packet, the first state transition scheme corresponding to the target service packet from a preset state transition policy library includes:

analyzing and processing the error report log of the target service message, and acquiring abnormal error report information of the target service message; the abnormal error reporting information comprises an error reporting scene, an error reporting object and an error reporting reason;

and determining a first state conversion scheme corresponding to the target service message in the state conversion strategy library according to the abnormal error reporting information of the target service message.

6. The method according to claim 5, wherein the analyzing the error report log of the target service packet to obtain the abnormal error report information of the target service packet includes:

vectorizing the error log of the target service message to obtain a plurality of real word vectors corresponding to the target service message;

7. The method according to any one of claims 1 to 4, wherein the performing, based on the first state transition scheme, a transition process on the processing state of the target service packet in the target service operation link includes:

performing re-operation processing on the target service message in the target service operation link according to the first state conversion scheme;

8. The method according to claim 7, wherein the performing, according to the result of the re-operation processing, a conversion processing on the processing state of the target service packet in the target service operation link includes:

if the result of the heavy operation processing is successful, performing conversion processing on the processing state of the target service message in the target service operation link to update the processing state of the target service message in the target service operation link to be a successful state;

if the result of the heavy operation processing is processing failure, writing the target service message into a preset abnormal event registration table; and the abnormal event registry is used for indicating a user to carry out conversion processing on the processing state of the target service message in the target service operation link.

9. The method of claim 8, further comprising:

acquiring a second state conversion scheme manually input according to a target service operation link where a target service message in the abnormal event registry is located;

10. The method of claim 9, further comprising:

11. The method of claim 7, wherein if the target service operation link is a resource deduction service link, the re-operation process is a resource deduction operation; if the target service operation link is a report service link, the re-operation processing is report operation; if the target service operation link is a resource adding service link, the re-operation processing is resource adding operation; and if the target service operation link is an abnormal resource cancellation service link, processing the heavy operation as an abnormal resource cancellation operation.

12. The method of claim 11, further comprising:

if the resource deduction operation is successfully deducted, executing a message sending operation on the target service message;

if the processing state of the sending operation is a sending failure, executing a resource adding operation on the target service message;

and if the processing state of the resource deduction operation is deduction unknown, executing abnormal resource cancellation operation on the target service message.

13. A service packet processing apparatus, wherein the apparatus comprises:

the log acquisition module is used for acquiring an error report log of the target service message if the target service message is monitored to be in an abnormal state in a target service operation link; the target service operation link is any service operation link of the target service message in a transaction link;

and the first state conversion module is used for performing conversion processing on the processing state of the target service message in the target service operation link based on the first state conversion scheme.

14. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 12.

15. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 12.

16. A computer program product comprising a computer program, characterized in that the computer program realizes the steps of the method of any one of claims 1 to 12 when executed by a processor.