Business system fault detection method, device, electronic equipment and storage medium

Abstract

The embodiments of the present application provide a method, device, electronic device and storage medium for detecting faults in a business system, by obtaining the operation and maintenance log of the first business system, the operation and maintenance log of the first business system is used to record the system error of the first business system; parsing the operation and maintenance log of the first business system to obtain a keyword identifier, the keyword identifier is used to characterize the error category corresponding to the system error; based on a preset prediction model and the keyword identifier, a second business system is determined, wherein the prediction model is used to predict the impact of system errors of different error categories on associated business systems, and the second business system is an associated business system affected by the system error of the first business system. Through the operation and maintenance log of the first business system, the error category of the system error of the first business system can be determined, and then according to the pre-trained prediction model, a prediction is made for the error category, thereby realizing early detection of potential faults and anomalies in the associated business system.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, electronic device and storage medium for detecting faults in a business system. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the application that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.

[0003] At present, the business processing system of banks is very large, and the various business systems are interconnected. When an abnormality or failure occurs in one of the business systems, it is likely to cause abnormalities in the related business systems.

[0004] In the prior art, a bank's business system is usually equipped with a fault monitoring system or module. When a fault is detected in the business system, an alarm will be sent to relevant maintenance personnel to remind them to handle the fault of the business system currently reporting the fault.

[0005] However, the existing fault monitoring system can only detect system faults and issue alarms after an abnormality occurs in the business system, but cannot detect potential faults in related business systems that are affected by the faulty business system but have not yet shown abnormalities. It can only issue an alarm after an abnormality occurs in the related business system, resulting in poor real-time fault detection and untimely alarms, affecting the overall stability of the banking business system. Summary of the invention

[0006] The present application provides a business system fault detection method, device, electronic device and storage medium to solve the problem that the existing fault monitoring system cannot timely detect potential faults of the associated business system.

[0007] According to the first aspect of an embodiment of the present application, the present application provides a business system fault detection method, the method comprising: obtaining an operation and maintenance log of a first business system, the operation and maintenance log of the first business system being used to record system errors of the first business system; parsing the operation and maintenance log of the first business system to obtain a keyword identifier, the keyword identifier being used to characterize an error category corresponding to the system error; determining a second business system based on a preset prediction model and the keyword identifier, wherein the prediction model is used to predict the impact of system errors of different error categories on associated business systems, the second business system being an associated business system affected by the system errors of the first business system.

[0008] In one possible implementation, based on a preset prediction model and the keyword identifier, determining the second business system includes: obtaining system association information, wherein the system association information is used to characterize the association relationship between different associated business systems and the first business system; inputting the system association information and the keyword identifier into the prediction model to obtain a system identifier output by the prediction model, wherein the system identifier is used to indicate the second business system.

[0009] In a possible implementation, the prediction model includes a pre-trained attention mechanism bidirectional encoding BERT model and a pre-trained bidirectional long short-term memory network BiLSTM model; the system association information and the keyword identifier are input into the prediction model to obtain the system identifier output by the prediction model, including: processing the keyword identifier according to the pre-trained BERT model to generate a word vector for the keyword identifier; and inputting the word vector and the system association information into the pre-trained BiLSTM model to obtain the system identifier.

[0010] In a possible implementation, the pre-trained BiLSTM model includes an attention layer; the attention layer is used to weight the context information of the keyword identifier.

[0011] In one possible implementation, the second business system includes multiple associated business systems, and the method also includes: obtaining a system impact weight output by the prediction model, the system impact weight corresponding to the system identifier one-to-one, and the system impact weight characterizing the degree to which the associated business system is affected by the system error of the first business system; according to a preset weight threshold, sending an alarm message to a terminal device corresponding to the associated business system whose system impact weight is greater than the weight threshold.

[0012] In a possible implementation, parsing the operation and maintenance log of the first business system to obtain a keyword identifier includes: obtaining a preset word segmentation library; and segmenting abnormal information in the operation and maintenance log based on the word segmentation library to generate a keyword identifier.

[0013] In a possible implementation, the method further includes: determining a weight value corresponding to each of the keyword identifiers based on preset keyword weight information, wherein the keyword weight information is used to characterize a mapping relationship between different keyword identifiers and weight values, and the weight value is used to characterize the degree of influence of the system error corresponding to the error category on the associated business system; determining a second business system based on a preset prediction model and the keyword identifier, including: determining a second business system based on a preset prediction model and a keyword identifier whose weight value is greater than a preset weight.

[0014] In a possible implementation, obtaining the operation and maintenance log of the first business system includes: when a system error is detected in the first business system, determining the time information when the system error occurs; and obtaining the corresponding operation and maintenance log of the first business system according to the time information.

[0015] In a possible implementation, after determining the second business system, the method further includes: sending an alarm message to a terminal device corresponding to the second business system, the alarm message including an identifier of the first business system and a system error of the first business system.

[0016] According to a second aspect of an embodiment of the present application, the present application provides a business system fault detection device, including:

[0017] An acquisition module, used to acquire an operation and maintenance log of a first business system, where the operation and maintenance log of the first business system is used to record a system error of the first business system;

[0018] A parsing module, used to parse the operation and maintenance log of the first business system to obtain a keyword identifier, where the keyword identifier is used to represent an error category corresponding to the system error report;

[0019] A determination module is used to determine a second business system based on a preset prediction model and the keyword identifier, wherein the prediction model is used to predict the impact of system errors of different error categories on associated business systems, and the second business system is an associated business system affected by the system errors of the first business system.

[0020] In one possible implementation, the determination module is specifically used to: obtain system association information, where the system association information is used to characterize the association relationship between different associated business systems and the first business system; input the system association information and the keyword identifier into the prediction model to obtain the system identifier output by the prediction model, where the system identifier is used to indicate the second business system.

[0021] In a possible implementation, the prediction model includes a pre-trained attention mechanism bidirectional encoding BERT model and a pre-trained bidirectional long short-term memory network BiLSTM model; when the determination module inputs the system association information and the keyword identifier into the prediction model to obtain the system identifier output by the prediction model, it is specifically used to: process the keyword identifier according to the pre-trained BERT model to generate a word vector for the keyword identifier; input the word vector and the system association information into the pre-trained BiLSTM model to obtain the system identifier.

[0022] In a possible implementation, the pre-trained BiLSTM model includes an attention layer; the attention layer is used to weight the context information of the keyword identifier.

[0023] In a possible implementation, the second business system includes multiple associated business systems, and the determination module is further used to: obtain the system impact weight output by the prediction model, the system impact weight corresponds one-to-one to the system identifier, and the system impact weight represents the degree to which the associated business system is affected by the system error of the first business system; according to a preset weight threshold, send an alarm message to the terminal device corresponding to the associated business system whose system impact weight is greater than the weight threshold.

[0024] In a possible implementation, the parsing module is specifically used to: obtain a preset word segmentation library; and based on the word segmentation library, segment the abnormal information in the operation and maintenance log to generate a keyword identifier.

[0025] In a possible implementation, the acquisition module is further used to: determine the weight value corresponding to each of the keyword identifiers based on preset keyword weight information, wherein the keyword weight information is used to characterize the mapping relationship between different keyword identifiers and weight values, and the weight value is used to characterize the degree of influence of the system error corresponding to the error category on the associated business system; the determination module is specifically used to: determine the second business system based on a preset prediction model and a keyword identifier whose weight value is greater than the preset weight.

[0026] In a possible implementation, the acquisition module is specifically used to: when a system error is detected in the first business system, determine the time information when the system error occurs; and acquire the operation and maintenance log of the corresponding first business system according to the time information.

[0027] In a possible implementation, after determining the second business system, the determination module is further used to: send an alarm message to a terminal device corresponding to the second business system, the alarm message including an identifier of the first business system and a system error of the first business system.

[0028] According to a third aspect of the embodiments of the present application, the present application provides an electronic device, including: a memory, a processor, and a computer program;

[0029] Wherein, the computer program is stored in the memory and is configured to be executed by the processor as the business system fault detection method as described in any one of the first aspects of the embodiments of the present application.

[0030] According to the fourth aspect of the embodiments of the present application, the present application provides a computer-readable storage medium, which stores computer execution instructions. When the computer execution instructions are executed by a processor, they are used to implement the business system fault detection method as described in any one of the first aspects of the embodiments of the present application.

[0031] According to a fifth aspect of the embodiments of the present application, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and various possible business system fault detection methods of the first aspect.

[0032] The business system fault detection method, device, electronic device and storage medium provided by the present application obtain the operation and maintenance log of the first business system, which is used to record the system error of the first business system; parse the operation and maintenance log of the first business system to obtain a keyword identifier, which is used to characterize the error category corresponding to the system error; determine the second business system based on a preset prediction model and the keyword identifier, wherein the prediction model is used to predict the impact of system errors of different error categories on the associated business system, and the second business system is the associated business system affected by the system error of the first business system. Since the error category of the system error of the first business system can be determined through the operation and maintenance log of the first business system, and then the error category is predicted according to the pre-trained prediction model, the associated business system affected by the system error of the first business system is located, and the potential faults and anomalies in the associated business system are detected in advance, the detection and alarm real-time of the associated business system is improved, and the overall stability of the banking business system is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0034] Figure 1 An application scenario diagram of the business system fault detection method provided in an embodiment of the present application;

[0035] Figure 2 A flowchart of a method for detecting a business system failure provided by an embodiment of the present application;

[0036] Figure 3 A flowchart of a business system fault detection method provided by another embodiment of the present application;

[0037] Figure 4 A schematic diagram of system association information provided in an embodiment of the present application;

[0038] Figure 5for Figure 3 Flowchart of step S207 in the illustrated embodiment;

[0039] Figure 6 A schematic diagram of the structure of a business system fault detection device provided by an embodiment of the present application;

[0040] Figure 7 A schematic diagram of an electronic device provided for one embodiment of the present application.

[0041] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0042] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0043] The application scenarios of the embodiments of the present application are explained below:

[0044] Figure 1 An application scenario diagram of the business system fault detection method provided in the embodiment of the present application, refer to Figure 1 The business system fault detection method provided in this embodiment can be applied to the scenario of system detection and alarm of banking business system, and specifically, can be applied to business system server, such as Figure 1 As shown, business system server A communicates with business system server B, business system server C and other servers through a local area network or a wide area network. The first business system deployed on business system server A will affect the business processing of the second business system deployed on business system server B and the third business system deployed on business system server C. Therefore, the second business system deployed on business system server B and the third business system deployed on business system server C are associated business systems of the first business system deployed on business system server A. Of course, it can be understood that in the scenario of this embodiment, the first business system, the second business system and the third business system can also be deployed in the same server or server cluster, and this situation is not exemplified one by one here. Business system server A, business system server B and business system server C are respectively connected to the terminal device for communication, and provide business services based on the first business system, the second business system and the third business system to the terminal device respectively.

[0045] In the prior art, when a system error occurs in the first business system, a fault alarm unit built into the first business system can send fault alarm information to a terminal device communicating with the first business system to remind the operation and maintenance personnel to confirm the fault and handle it in a timely manner. However, since the bank's business processing system is very large and complex, and the various business systems are interconnected, when one of the business systems is abnormal and fails, it is likely to cause abnormalities in the associated business systems related to it, such as Figure 1 In the application scenario shown, when the first business system reports an error, the associated business systems (second business system, third business system) related to the first business system may cause an exception when using a certain business service provided by the first business system due to a fault in the first business system. However, in the prior art, a fault alarm can only be issued for the corresponding associated business system after the associated business system uses the faulty business service provided by the first business system and an exception occurs, but it is impossible to issue an early warning and inform the relevant operation and maintenance personnel to handle it (for example, cutting off the call to the business service in the first business system). This way of handling business failures after they occur leads to the problem of poor real-time fault detection and processing of the business system, which in turn causes greater security risks and stability problems for the banking business system.

[0046] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0047] Figure 2 A flowchart of a business system fault detection method provided in one embodiment of the present application is applied to a business system server or a terminal device communicating with a business system server. In this embodiment, the business system server is used as the execution subject for introduction. Figure 2 As shown, the business system fault detection method provided in this embodiment includes the following steps:

[0048] Step S101: Acquire an operation and maintenance log of a first business system. The operation and maintenance log of the first business system is used to record system errors of the first business system.

[0049] Exemplarily, during the operation of the business system, relevant operation and maintenance data will be generated and stored in the business system server or in an independent electronic device or storage medium that is communicatively connected to the business system server. Such operation and maintenance data is the operation and maintenance log. When a system error occurs in the first business system, the time when the system error occurs, the description of the exception, and other information will be recorded through the operation and maintenance log. Through the operation and maintenance log, the operation and maintenance personnel can diagnose and handle the operation and fault conditions of the first business system. The operation and maintenance log can be generated by the relevant log function in the system software of the business system. The specific technical solution for generating the log is a prior art and will not be described in detail here.

[0050] Further, a system error is an abnormal response of a business system or a business service in the business system due to input value errors, network failures, etc. when the business system is invoking related business services. The business system can monitor such abnormal responses generated by the system or service through system software, and then record them in the operation and maintenance log. Exemplarily, system errors include: null pointer exception, file call exception, connection exception, etc.

[0051] Step S102: parsing the operation and maintenance log of the first business system to obtain a keyword identifier, where the keyword identifier is used to characterize an error category corresponding to a system error.

[0052] Exemplarily, the business system server reads the operation and maintenance log, parses the operation and maintenance log, and realizes the content recognition of the operation and maintenance log. Specifically, the operation and maintenance log includes multiple character strings for describing the content of the operation and maintenance log. By parsing the character string in the operation and maintenance log and performing word segmentation processing, one or more keyword identifiers can be obtained, and the keyword identifier is used to characterize the error category corresponding to the system error. For example, after parsing and word segmentation processing of the character string in the operation and maintenance log, the obtained keyword identifier is "Connection timed out", that is, the error category characterizing the system error is a connection timeout exception; for another example, after parsing and word segmentation processing of the character string in the operation and maintenance log, the obtained keyword identifier is "java.io.FileNotFoundException", that is, the error category characterizing the system error is a file call exception. Thereby, the recognition and extraction of the content related to the system error in the operation and maintenance log is realized.

[0053] Step S103, determining the second business system based on a preset prediction model and keyword identification, wherein the prediction model is used to predict the impact of system errors of different error categories on associated business systems, and the second business system is an associated business system affected by the system error of the first business system.

[0054] Exemplarily, after obtaining the keyword identifier representing the error category, different error categories may cause potential anomalies in different associated business systems according to the association relationship between different business systems, that is, other associated business systems that have an association relationship with the first business system will have anomalies when calling the business service corresponding to the error category in the first business system. Exemplarily, the prediction model is a pre-trained model based on a neural network, which can be used to predict the impact of system errors of different error categories on associated business systems. The prediction model has been supervised and trained based on the label data composed of historical keyword identifiers and corresponding historical operation and maintenance logs representing system errors of associated business systems, and converged. The specific training process of the prediction model is not described here. According to the keyword identifier and the preset prediction model, the associated business system that may be affected by the system error of the first business system, that is, the second business system, can be predicted. Furthermore, after the system alarm occurs in the first business system, automatic fault detection and positioning of the associated business system is realized, and early warning of the fault of the associated business system is realized, reducing the risk of business system failure.

[0055] In this embodiment, the operation and maintenance log of the first business system is obtained, and the operation and maintenance log of the first business system is used to record the system error of the first business system; the operation and maintenance log of the first business system is parsed to obtain a keyword identifier, and the keyword identifier is used to characterize the error category corresponding to the system error; based on a preset prediction model and keyword identifier, the second business system is determined, wherein the prediction model is used to predict the impact of system errors of different error categories on the associated business system, and the second business system is the associated business system affected by the system error of the first business system. Since the error category of the system error of the first business system can be determined through the operation and maintenance log of the first business system, and then the error category is predicted according to the pre-trained prediction model, the associated business system affected by the system error of the first business system is located, and the potential faults and anomalies in the associated business system are detected in advance, the detection and alarm real-time of the associated business system is improved, and the overall stability of the banking business system is improved.

[0056] Figure 3 A flowchart of a business system fault detection method provided by another embodiment of the present application is as follows: Figure 3 As shown, the business system fault detection method provided by this embodiment is Figure 2 On the basis of the business system fault detection method provided in the illustrated embodiment, steps S101-S103 are further refined, and the business system fault detection method provided in this embodiment includes the following steps:

[0057] Step S201, when a system error is detected in the first business system, the time information of the occurrence of the system error is determined.

[0058] Step S202: Obtain the operation and maintenance log of the corresponding first business system according to the time information.

[0059] For example, due to the special nature of banking business, for security reasons, after some system errors occur, they need to be manually processed and confirmed first. Therefore, the processing of system errors by the business system server may be non-real-time. When the business system server detects that a system error occurs in the first business system, the time information of the system error occurrence is determined. Afterwards, when the business system server starts to process the system error, the business system server reads the corresponding operation and maintenance log according to the time information of the previously confirmed system error occurrence, thereby tracing the fault.

[0060] Step S203, obtaining a preset word segmentation library, and based on the word segmentation library, performing word segmentation on the abnormal information in the operation and maintenance log to generate a keyword identifier.

[0061] Exemplarily, the word segmentation library includes a plurality of preset word segmentation character strings, which are used to perform word segmentation processing on the abnormal information in the operation and maintenance log, that is, the long character string describing the abnormality. The word segmentation library can be set by the user according to specific needs, such as the system software and development language of the business system, and is used to identify the specific error category of the system error and generate a keyword identifier. Among them, the specific implementation method of using the word segmentation library for word segmentation, for example, can be through calling a word segmentation engine, such as the Lucene engine, referring to the word segmentation library to analyze the content in the operation and maintenance log and output the word segmentation result. This process is a prior art known to those skilled in the art and will not be repeated here.

[0062] Step S204, determining the weight value corresponding to each keyword identifier according to preset keyword weight information, wherein the keyword weight information is used to characterize the mapping relationship between different keyword identifiers and weight values, and the weight value is used to characterize the impact of the system error corresponding to the error category on the associated business system.

[0063] Step S205 : The weight values ​​corresponding to the keyword identifiers are used to determine a target keyword identifier, where the target keyword identifier is a keyword identifier having a weight value greater than a preset weight.

[0064] Exemplarily, the keyword identifiers generated after word segmentation of the operation and maintenance log represent different error categories. Due to special requirements for business security of the banking business system, different error categories have different degrees of impact on other related business systems. For example, if a network connection timeout occurs, all business services that require network connection will be affected. Therefore, the error category corresponding to the keyword identifier has a greater impact on the first business system. Furthermore, the error category corresponding to the keyword identifier also has a greater impact on the related business systems that have an association relationship with the first business system, and the weight value of the keyword identifier is also greater. On the contrary, for example, if minor problems such as font errors and font size errors occur, the error category corresponding to the keyword identifier has a smaller impact on the first business system. Furthermore, the error category corresponding to the keyword identifier also has a smaller impact on the related business systems that have an association relationship with the first business system, and the weight value of the keyword identifier is also smaller.

[0065] Through the keyword weight information, each keyword identifier is screened according to the weight value corresponding to each keyword identifier, and one or more more important keyword identifiers are determined as target keyword identifiers. In the subsequent processing process, the second business system with potential failure risk is predicted based on the target keyword identifier, which can improve the accuracy and effectiveness of determining the second business system.

[0066] Step S206: acquiring system association information, where the system association information is used to characterize association relationships between different associated business systems and the first business system.

[0067] Step S207: input the system association information and the target keyword identifier into the prediction model to obtain the system identifier output by the prediction model, where the system identifier is used to indicate the second business system.

[0068] Exemplarily, the system association information is information used to characterize the mutual influence relationship between the association relationships of different relationship business systems. The system association information can be implemented in the form of an association information table. Figure 4 A schematic diagram of system association information provided in an embodiment of the present application, wherein business system A, business system B, and business system C are three business systems. Figure 4As shown, after a system error occurs in business system A, in addition to affecting itself (indicated by 1, the same below), it will also affect business system B, but will not affect business system C (indicated by 0, the same below). After a system error occurs in business system B, in addition to affecting itself, it will also affect business system A and business system C; after a system error occurs in business system C, in addition to affecting itself, it will also affect business system B, but will not affect business system C. In other implementations, there may be a situation where the business system is affected in one direction. For example, after a system error occurs in business system B, in addition to affecting itself, it will also affect business system D, but after a system error occurs in business system D, it will only affect itself and will not affect business system D (this situation is not shown in the figure).

[0069] Furthermore, the system association information and the target keyword identifier are input into the prediction model. Through the prediction model, the system association information and the target keyword identifier are used as input quantities to predict the associated business system that may be affected by the system error of the first business system, that is, the second business system, and obtain the system identifier output by the prediction model to indicate the second business system.

[0070] Optionally, the prediction model includes a pre-trained attention mechanism bidirectional encoding (Bidirectional Encoder Representations from Transformers, BERT) model and a pre-trained bidirectional long short-term memory network (Bi-directional Long Short-Term Memory, BiLSTM) model, such as Figure 5 As shown, step S207 includes two specific implementation steps, step S2071 and step S2072:

[0071] Step S2071: Process the keyword identifier according to the pre-trained BERT model to generate a word vector for the keyword identifier.

[0072] Step S2072, input the word vector and system association information into the pre-trained BiLSTM model to obtain the system identification.

[0073] Exemplarily, the BERT model is an existing preprocessing model for language processing. BERT can be used as a word vector model. It converts the strings in the operation and maintenance log into word vectors that can be processed by formulas and models through pre-training, and then realizes the classification prediction of word vectors through classification mode. BERT is a language model built based on bidirectional Transformer. The model is large and has good accuracy after training. Among them, the specific model structure, principle and training method of the BERT model are all existing technologies known to technicians in this field, and will not be repeated here.

[0074] Furthermore, after generating the word vector, based on the pre-trained BiLSTM model, the word vector and system association information are used as input to classify and predict the associated business systems involved in the system association information, and the second business system, that is, the system identifier of the associated business system affected by the system error of the first business system, can be output.

[0075] Among them, in the above steps, the BiLSTM model is pre-trained with labeled data samples before use. The process of training the BiLSTM model is an existing technology known to those skilled in the art and will not be repeated here.

[0076] Exemplarily, the pre-trained BiLSTM model includes an attention layer; the attention layer is used to weight the context information of the keyword identifier.

[0077] Step S208 , obtaining the system impact weight output by the prediction model, wherein the system impact weight corresponds to the system identifier in a one-to-one manner, and the system impact weight represents the degree of impact of the associated business system on the system error of the first business system.

[0078] Step S209: according to a preset weight threshold, an alarm message is sent to a terminal device corresponding to an associated business system whose system impact weight is greater than the weight threshold.

[0079] Exemplarily, the output of the prediction model also includes a system impact weight corresponding to the system identifier. The system impact weight represents the degree of influence of the associated business system on the system error of the first business system. There are differences in the degree of association between different business systems. When a system error occurs in one of the business systems with a large degree of association, it will also cause a large impact on the business system with a large degree of association. Therefore, when the second business system output by the pre-trained prediction model includes multiple associated business systems, while outputting the system identifier corresponding to each associated business system, the system impact weight corresponding to each system identifier is output, and according to the preset weight threshold, the associated business system with a system impact weight greater than the weight threshold, that is, the associated business system on which the system error of the first business system has a large impact, is alarmed, so that the operation and maintenance personnel corresponding to the business system can handle it in time. Avoid affecting the business of the associated business system. Improve the effectiveness of system detection and avoid sending false alarms to the associated business system. Exemplarily, the alarm information includes the identifier of the first business system and the system error of the first business system.

[0080] Figure 6 A schematic diagram of the structure of a business system fault detection device provided by an embodiment of the present application is applied to a business system server, such as Figure 6 As shown, the business system fault detection device 3 provided in this embodiment includes:

[0081] An acquisition module 31 is used to acquire an operation and maintenance log of the first business system, where the operation and maintenance log of the first business system is used to record system errors of the first business system;

[0082] The parsing module 32 is used to parse the operation and maintenance log of the first business system to obtain a keyword identifier, where the keyword identifier is used to represent an error category corresponding to a system error report;

[0083] The determination module 33 is used to determine the second business system based on a preset prediction model and keyword identification, wherein the prediction model is used to predict the impact of system errors of different error categories on the associated business system, and the second business system is an associated business system affected by the system error of the first business system.

[0084] The acquisition module 31, the analysis module 32 and the determination module 33 are connected in sequence. The business system fault detection device 3 provided in this embodiment can be executed as follows: Figure 2-5 The technical solutions of any of the method embodiments shown have similar implementation principles and technical effects, which will not be described in detail here.

[0085] Figure 7 A schematic diagram of an electronic device provided by an embodiment of the present application, such as Figure 7 As shown, the electronic device 4 provided in this embodiment includes: a memory 41, a processor 42 and a computer program.

[0086] The computer program is stored in the memory 41 and is configured to be executed by the processor 42 to implement the present application. Figure 2-Figure 5 A business system fault detection method provided by any one of the corresponding embodiments.

[0087] The memory 41 and the processor 42 are connected via a bus 43 .

[0088] For related instructions, please refer to Figure 2-Figure 5 The relevant descriptions and effects corresponding to the steps in the corresponding embodiments can be understood, and no further elaboration is made here.

[0089] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the present application. Figure 2-Figure 5 A business system fault detection method provided by any one of the corresponding embodiments.

[0090] Among them, the computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0091] One embodiment of the present application provides a computer program product, including a computer program, which implements the present application when executed by a processor. Figure 2-Figure 5 A business system fault detection method provided by any one of the corresponding embodiments.

[0092] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0093] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the application disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0094] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for detecting a business system fault, characterized in that: The method comprises: Acquire an operation and maintenance log of the first business system, where the operation and maintenance log of the first business system is used to record system errors of the first business system; Parsing the operation and maintenance log of the first business system to obtain a keyword identifier, where the keyword identifier is used to represent an error category corresponding to the system error report; Based on a preset prediction model and the keyword identifier, determining a second business system, wherein the prediction model is used to predict the impact of system errors of different error categories on associated business systems, and the second business system is an associated business system affected by the system error of the first business system; The determining of the second business system based on the preset prediction model and the keyword identifier includes: Acquire system association information, where the system association information is used to characterize association relationships between different associated business systems and the first business system; Inputting the system association information and the keyword identifier into the prediction model to obtain a system identifier output by the prediction model, wherein the system identifier is used to indicate the second business system; The prediction model includes a pre-trained attention mechanism bidirectional encoding BERT model and a pre-trained bidirectional long short-term memory network BiLSTM model; the system association information and the keyword identifier are input into the prediction model to obtain the system identifier output by the prediction model, including: Processing the keyword identifier according to the pre-trained attention mechanism bidirectional encoding BERT model to generate a word vector for the keyword identifier; Input the word vector and system association information into a pre-trained bidirectional long short-term memory network BiLSTM model to obtain the system identifier; The second business system includes a plurality of associated business systems, and the method further includes: Obtaining a system impact weight output by the prediction model, wherein the system impact weight corresponds to the system identifier in a one-to-one manner, and the system impact weight represents the degree to which the associated business system is affected by the system error of the first business system; According to a preset weight threshold, an alarm message is sent to a terminal device corresponding to an associated business system whose system impact weight is greater than the weight threshold.

2. The method according to claim 1, characterized in that The pre-trained bidirectional long short-term memory network BiLSTM model includes an attention layer; the attention layer is used to weight the context information of the keyword identifier.

3. The method according to claim 1, characterized in that Parsing the operation and maintenance log of the first business system to obtain keyword identifiers includes: Get the preset word segmentation library; Based on the word segmentation library, the abnormal information in the operation and maintenance log is segmented to generate keyword identifiers.

4. The method according to claim 3, characterized in that The method further comprises: Determine the weight value corresponding to each keyword identifier according to preset keyword weight information, wherein the keyword weight information is used to characterize the mapping relationship between different keyword identifiers and weight values, and the weight value is used to characterize the degree of influence of the system error corresponding to the error category on the associated business system; Determining the second business system based on the preset prediction model and the keyword identifier includes: Based on a preset prediction model and a keyword identifier whose weight value is greater than a preset weight, a second business system is determined.

5. The method according to any one of claims 1 to 4, characterized in that: Obtain the operation and maintenance log of the first business system, including: When a system error is detected in the first business system, determining the time information of the occurrence of the system error; According to the time information, an operation and maintenance log of the corresponding first business system is obtained.

6. The method according to any one of claims 1 to 4, characterized in that: After determining the second business system, the method further includes: Sending an alarm message to a terminal device corresponding to the second business system, the alarm message including an identifier of the first business system and a system error of the first business system.

7. A business system fault detection device, characterized in that: include: An acquisition module, used to acquire an operation and maintenance log of a first business system, where the operation and maintenance log of the first business system is used to record a system error of the first business system; A parsing module, used to parse the operation and maintenance log of the first business system to obtain a keyword identifier, where the keyword identifier is used to represent an error category corresponding to the system error report; A determination module, configured to determine a second business system based on a preset prediction model and the keyword identifier, wherein the prediction model is used to predict the impact of system errors of different error categories on associated business systems, and the second business system is an associated business system affected by the system error of the first business system; The determination module is specifically used to obtain system association information, where the system association information is used to characterize the association relationship between different associated business systems and the first business system; input the system association information and the keyword identifier into the prediction model to obtain the system identifier output by the prediction model, where the system identifier is used to indicate the second business system; The prediction model includes a pre-trained attention mechanism bidirectional encoding BERT model and a pre-trained bidirectional long short-term memory network BiLSTM model; the determination module is further specifically used to: process the keyword identifier according to the pre-trained attention mechanism bidirectional encoding BERT model to generate a word vector of the keyword identifier; input the word vector and system association information into the pre-trained bidirectional long short-term memory network BiLSTM model to obtain the system identifier; The second business system includes a plurality of associated business systems, and a sending module is used to obtain a system impact weight output by the prediction model, wherein the system impact weight corresponds to the system identifier in a one-to-one manner, and the system impact weight represents the degree of influence of the associated business system on the system error of the first business system; According to a preset weight threshold, an alarm message is sent to a terminal device corresponding to an associated business system whose system impact weight is greater than the weight threshold.

8. An electronic device, characterized in that: include: Memory, processor and computer program; The computer program is stored in the memory and is configured to be executed by the processor to implement the business system fault detection method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the business system fault detection method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for detecting a fault in a business system according to any one of claims 1 to 6 is implemented.

Images