Fault positioning method and device, equipment and medium

Abstract

The invention relates to the technical field of artificial intelligence and the field of financial science and technology, and provides a fault positioning method and device, equipment and a medium, and the method comprises the steps: obtaining fault description information of a fault event generated in a business system; obtaining a workflow corresponding to the fault description information through a workflow agent; executing the workflow through a workflow engine to obtain fault positioning reference information; according to the fault positioning reference information and the fault description information, fault positioning is carried out through a fault positioning model, the fault reason of the fault event is obtained, and the fault positioning model is obtained by carrying out fine tuning training on a large model. According to the invention, the full-automatic fault positioning of the fault event in the service system can be realized, and the fault positioning efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence technology and financial technology, and in particular to a fault location method, a fault location device, a computer device, and a computer-readable storage medium. Background Art

[0002] In the digital age, microservices architecture has become widespread. An increasing number of business systems, such as insurance, banking, and e-commerce, are adopting it. The core of microservices architecture is the decomposition of large business systems into multiple small, autonomous services. For example, an e-commerce system can be divided into user services, product services, order services, payment services, and logistics services. Each service focuses on a single business function, such as a product service responsible for managing and displaying product information.

[0003] As business systems become increasingly complex, inter-service call relationships become intricate. A single service failure can quickly propagate along the call chain, causing widespread service disruption. This makes quickly locating the cause of a failure crucial. However, related technologies often rely on manual troubleshooting by operations personnel, limiting fault location efficiency to the capabilities of individual operators. Summary of the Invention

[0004] The embodiments of the present invention provide a fault location method, a fault location apparatus, a computer device, and a computer-readable storage medium, which can realize fully automated fault location of fault events occurring in a business system and improve fault location efficiency.

[0005] In a first aspect, a fault location method is provided, comprising: Obtain fault description information of fault events that occur in the business system; Obtaining the workflow corresponding to the fault description information through the workflow agent; Execute the workflow through the workflow engine to obtain fault location reference information; According to the fault location reference information and fault description information, the fault location model is used to locate the fault and obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning the large model.

[0006] In a second aspect, a fault location device is provided, comprising: Description information acquisition module, used to obtain fault description information of fault events occurring in the business system; A workflow acquisition module is used to acquire the workflow corresponding to the fault description information through a workflow agent; A reference information acquisition module is used to execute the workflow through the workflow engine to obtain fault location reference information; The fault location module is used to locate the fault based on the fault location reference information and fault description information through the fault location model to obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning the large model.

[0007] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned fault location method when executing the computer program.

[0008] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned fault location method are implemented.

[0009] In the scheme implemented by the above-mentioned fault location method, device, equipment and medium, fault description information of a fault event occurring in a business system is obtained; a workflow corresponding to the fault description information is obtained through a workflow agent; the workflow is executed by a workflow engine to obtain fault location reference information; based on the fault location reference information and the fault description information, fault location is performed through a fault location model to obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning and training a large model. In this way, when a fault event occurs in a business system, the fault description information and the fault location reference information can be automatically obtained, and the fault location model can complete the fault location of the fault event and obtain the fault cause of the fault event. This achieves fully automated fault location of fault events occurring in the business system, which improves the efficiency of fault location compared to manual fault location methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0011] Figure 1 This is a schematic diagram of an application environment of a fault location method according to an embodiment of the present invention; Figure 2 is a flow chart of a fault location method according to an embodiment of the present invention; Figure 3 This is an example diagram of a workflow for obtaining fault description information through a workflow agent in one embodiment of the present invention; Figure 4 This is a flow chart of optimizing the accuracy of locating the cause of a fault by updating the workflow in one embodiment of the present invention; Figure 5is a structural diagram of a fault location device in one embodiment of the present invention; Figure 6 is a structural diagram of a computer device in one embodiment of the present invention; Figure 7 FIG. 2 is another structural diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0012] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0013] The fault location method provided by the embodiment of the present invention can be applied as follows: Figure 1 In the application environment shown, the client communicates with the server through the network. The server can obtain the fault description information of the fault event occurring in the business system through the client; obtain the workflow corresponding to the fault description information through the workflow agent; and execute the workflow through the workflow engine to obtain fault location reference information; finally, based on the fault location reference information and the fault description information, the fault location model is used to locate the fault and obtain the fault cause corresponding to the fault description information, wherein the fault location model is obtained by fine-tuning the large model; and the fault cause obtained by fault location is returned to the client so that the client can display the fault cause to the user. The client may include but is not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server can be implemented as an independent server or a server cluster consisting of multiple servers. The present invention is described in detail below through specific embodiments.

[0014] See also Figure 2 As shown, Figure 2 A schematic flow chart of a fault location method provided in an embodiment of the present invention includes the following steps: S110: Obtain fault description information of a fault event occurring in the business system.

[0015] Business systems include those that utilize a microservices architecture, such as insurance, banking, and e-commerce systems. For example, e-commerce systems can include user services, product services, order services, payment services, and logistics services. Each service focuses on a single business function, such as a product service responsible for managing and displaying product information.

[0016] A failure event refers to a failure condition during the operation of a business system where some or all of the functions of the services within the business system cannot be performed normally. These failure conditions will interfere with the normal functions of the business system and, in severe cases, cause business interruption.

[0017] The fault description information is used to describe the fault time point and fault phenomenon of the fault event. The information format of the fault description information is not specifically limited here. For example, the acquired fault description information may be: Product failure: Error Server response 502; Failure Time: 2025-03-13 10:00:00.

[0018] For another example, the acquired fault description information may be: Failure time: November 15, 2024, 14:32:05; Symptom: Users reported that when placing an order on the APP, there was no response after clicking the Submit Order button.

[0019] S120: Obtain the workflow corresponding to the fault description information through the workflow agent.

[0020] A workflow is a set of executable instructions used to complete a specific operation.

[0021] The workflow agent is configured to generate workflow requirements based on input, and output a workflow that meets the workflow generation requirements and can be executed by the workflow engine. The workflow agent can be obtained by fine-tuning a large language model. The specific large language model used and the specific fine-tuning training method are not specifically limited here.

[0022] In an embodiment of the present invention, after obtaining the fault description information of a fault event occurring in a business system, a workflow generation requirement corresponding to the fault event is generated based on the fault description information. The workflow generation requirement instructs the workflow agent to generate a workflow for the workflow engine to obtain the fault location reference information required by the fault location model to perform fault location on the fault event. The workflow generation requirement is accordingly input into the workflow agent, and a workflow that can be executed by the workflow engine is generated through the workflow agent, so that the workflow engine executes the workflow to obtain the fault location reference information required by the fault location model to perform fault location.

[0023] For example, you can construct a workflow generation requirement in the form of a prompt: "The following fault event {fill in the fault description information here} has occurred in the business system. Now you need to locate the cause of the fault event through the fault location model. Please generate a workflow that is compatible with the workflow engine {fill in the workflow engine description information here} so that the workflow engine can execute the generated workflow and obtain the following fault location reference information {fill in the fault location reference information description here} required for fault location. " The workflow engine description information is used to describe the type of workflow engine, and the fault location reference information description information is used to describe the type of information.

[0024] S130: Execute the workflow through the workflow engine to obtain fault location reference information.

[0025] As described above, in the embodiment of the present invention, a workflow corresponding to the fault description information is obtained through the workflow agent. When the workflow is executed by the workflow engine, the fault location reference information required by the fault location model for fault location is obtained.

[0026] It should be noted that the embodiments of the present invention do not impose any specific restrictions on the workflow engine to be used. For example, Conductor can be used as the workflow engine to execute workflows to obtain fault location reference information. Conductor is an open source distributed workflow engine that can help users build, manage and execute complex workflows. It uses the JSON format to define workflows. For example, the following takes the example of a fault description information describing a fault event of a product service in an e-commerce system. Assuming that the required fault location reference information includes the service call relationship chain of the product service, the JSON format workflow generated by the workflow agent can be: "name": "get_faulty_product_service_call_chain", "description": "Get the service call relationship chain of the product service that failed in the e-commerce system", "version": 1, "tasks": [ { "name":"detect_product_service_fault", "taskReferenceName": "detect_fault_task", "type": "HTTP", "uri": "http: / / ecommerce - system / monitoring / product - service / fault", "httpMethod": "GET", "inputParameters": {}, "retryCount": 3, "retryLogic": "FIXED", "retryDelaySeconds": 5 }, {"name":"get_product_service_details", "taskReferenceName":"product_service_details_task", "type": "HTTP", "uri":"http: / / ecommerce - system / product - service / details", "httpMethod": "GET", "inputParameters": { "faultInfo": "${detect_fault_task.output.faultInfo}" }, "retryCount": 3, "retryLogic": "FIXED", "retryDelaySeconds": 5, "dependsOn": ["detect_fault_task"] }, { "name": "get_downstream_services", "taskReferenceName": "downstream_services_task", "type": "HTTP", "uri": "http: / / ecommerce - system / product - service / downstream", "httpMethod": "GET", "inputParameters": { "productServiceId": "${product_service_details_task.output.productServiceId}" }, "retryCount": 3, "retryLogic": "FIXED", "retryDelaySeconds": 5, "dependsOn": ["product_service_details_task"] }, { "name": "get_upstream_services", "taskReferenceName": "upstream_services_task", "type": "HTTP", "uri": "http: / / ecommerce - system / product - service / upstream", "httpMethod": "GET", "inputParameters": { "productServiceId": "${product_service_details_task.output.productServiceId}" }, "retryCount": 3, "retryLogic": "FIXED", "retryDelaySeconds": 5, "dependsOn": ["product_service_details_task"] }, { "name": "aggregate_service_relations", "taskReferenceName": "aggregate_task", "type": "Java", "classPath": "com.example.AggregateCallChain", "inputParameters": { "downstream": "${downstream_services_task.output.downstreamServices}", "upstream": "${upstream_services_task.output.upstreamServices}" }, "retryCount": 3, "retryLogic": "FIXED", "retryDelaySeconds": 5, "dependsOn": ["downstream_services_task", "upstream_services_task"] } ], "inputParameters": {}, "outputParameters": { "faulty_product_service_call_chain": ${aggregate_task.output.callChain}" } } The above workflow consists of three tasks: the "downstream_services_task" task for acquiring the downstream services of the product service, the "upstream_services_task" task for acquiring the upstream services of the product service, and the "aggregate_task" task for aggregating upstream and downstream service relationships. The workflow is executed by the Conductor workflow engine to execute the "downstream_services_task" task for acquiring the downstream services of the product service, the "upstream_services_task" task for acquiring the upstream services of the product service, and the "aggregate_task" task for aggregating upstream and downstream service relationships. Ultimately, the output is the service call relationship chain for the product service.

[0027] S140: Based on the fault location reference information and the fault description information, the fault location model is used to locate the fault and obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning the large model.

[0028] It should be noted that artificial intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. AI also involves studying the design principles and implementation methods of various intelligent machines, enabling them to possess the capabilities of perception, reasoning, and decision-making.

[0029] Large models, also known as large pre-trained models, are deep learning models with massive parameters that are pre-trained on large amounts of data. Examples include the GPT series of models and the BERT series of models. Large models are designed to learn common patterns and knowledge in the data and then be fine-tuned or directly applied to various downstream tasks, such as text generation in natural language processing, question-answering systems, and image classification and object detection in computer vision.

[0030] In an embodiment of the present invention, a fault location model with fault location capability is obtained by pre-fine-tuning a large model. For example, historical fault description information of historical fault events that occurred in the business system can be obtained, as can the historical fault causes obtained by manually locating these historical fault events, and the historical fault location reference information used in the manual fault location process. For a piece of historical fault description information, a training sample is constructed based on the historical fault description information and its corresponding historical fault location reference information, and the historical fault cause corresponding to the historical fault description information is used as the training label of the training sample. In this way, multiple sets of training sample-training label pairs can be obtained. The large model is fine-tuned using these obtained training sample-training label pairs to obtain a fault location model with fault location capability. No specific restrictions are placed on the type of large model used.

[0031] Accordingly, in an embodiment of the present invention, after obtaining the above fault description information and fault location reference information, the obtained fault description information and fault location reference information are input into a fault location model. The fault location model is used to locate the fault event that occurred, thereby obtaining the fault cause of the fault event. For example, the fault description information and fault location reference information can be constructed as prompts and input into the fault location model. For example, the constructed prompts can be: "The following fault event {fill in the fault description here} has occurred in the business system you are responsible for operating. Please use the following reference information {fill in the fault location reference information here} to locate the possible root cause of the fault event." Optionally, in one embodiment, obtaining a workflow corresponding to the fault description information through a workflow agent includes: According to the fault description information, the fault event is classified to obtain the fault type of the fault event; The workflow corresponding to the fault description information is obtained through the workflow agent that matches the fault type.

[0032] In an embodiment of the present invention, a multi-agent architecture is used to implement workflow generation, wherein each workflow agent focuses on a specific failure type (such as network delay, service unavailability, etc.) and can generate targeted workflows based on the characteristics of the failure type it focuses on, thereby improving the flexibility of workflow generation.

[0033] Among them, when obtaining the workflow corresponding to the fault description information, the fault event is first classified according to the fault description information to determine the fault type of the fault event, and then the workflow corresponding to the fault description information is obtained through the workflow agent matching the fault type.

[0034] For example, a fault classification model for fault type classification can be trained using sample fault descriptions and the corresponding fault type labels. The model architecture of the fault classification model is not restricted. For example, a backbone network can be constructed using multiple Transformer blocks, and a classification head (which can be composed of a fully connected layer and a soft maximization layer) can be added to the backbone network to obtain the fault classification model. Accordingly, when classifying a fault event, the fault description of the fault event can be input into the fault classification model, and the fault classification model is used to perform fault classification and obtain the fault type of the fault event.

[0035] For example, please refer to Figure 3 Assume that there are pre-configured workflow agent A that matches fault type A, workflow agent B that matches fault type B, and workflow agent C that matches fault type C. First, the fault description information of the fault event is obtained and input into the fault classification model for fault classification, and the fault type of the fault event is obtained as fault type B. Then, the workflow corresponding to the fault description information is obtained through the workflow agent B that matches fault type B.

[0036] Optionally, in one embodiment, performing fault location using a fault location model based on the fault location reference information and the fault description information to obtain the fault cause of the fault event includes: Acquire fault location knowledge corresponding to the fault type, where the fault location knowledge records the fault location strategy corresponding to the fault type; According to the fault location reference information, fault description information and fault location knowledge, the fault location model is used to locate the fault and obtain the fault cause of the fault event.

[0037] Large model divergence refers to the phenomenon in which a large model deviates from expectations, lacks logic, or exhibits excessive freedom. Since the fault localization model in this invention is obtained by fine-tuning and training a large model, to avoid potential divergence, the present invention imposes constraints on the fault localization process of the fault localization model to ensure its accuracy.

[0038] A knowledge base is pre-built, and fault location knowledge corresponding to different fault types is maintained in the knowledge base. This fault location knowledge records the fault location strategy for the corresponding fault type. The fault location strategy describes the process steps of how to use the fault location reference information to locate the fault. The fault location strategy can be generated based on the fault location expert knowledge of the operation and maintenance personnel.

[0039] Accordingly, when fault location is performed through the fault location model based on the fault location reference information and the fault description information, the fault location knowledge corresponding to the fault type of the fault event is first obtained from the knowledge base, and then the fault location reference information, the fault description information and the fault location knowledge are constructed as prompt words and input into the fault location model. The fault location strategy recorded in the fault location knowledge constrains the fault location process of the fault location model, so that the fault location model can accurately locate the fault based on the fault location reference information and the fault description information to obtain the cause of the fault.

[0040] For example, the constructed prompt words can be: The business system you are responsible for operating and maintaining has experienced the following fault event {fill in the fault description here}. Please use the following reference information {fill in the fault location reference information here} and the following fault location strategy {fill in the fault location knowledge here} to locate the possible root cause of the fault event.

[0041] Optionally, in one embodiment, executing a workflow through a workflow engine to obtain fault location reference information includes: Execute the workflow through the workflow engine to obtain the service call relationship chain associated with the fault event, and obtain at least one of the service change record, service alarm record and resource usage record of the service in the service call relationship chain, At least one of the obtained service call relationship chain, service change record, service alarm record and resource usage record is used as fault location reference information.

[0042] The service change record details any changes made to the service, including but not limited to the following: Change identification: Each change has a unique identifier, such as a change number, to facilitate tracking and querying of changes; Change Type: This includes code changes, configuration changes, hardware upgrades, feature additions or deletions, etc. For example, code changes may be to fix vulnerabilities or optimize performance; configuration changes may involve adjusting server parameters or changing network settings, etc. Change description: Detailed description of the change, such as modifying the algorithm of a function or updating the database connection string in the configuration file; Change time: Accurately record the time when the change occurred, usually in the format of year-month-day hour:minute:second.

[0043] Change Object: Indicates the service in which the change occurred.

[0044] Service alert records refer to notifications and records automatically generated by the business system when a service anomaly occurs or may affect the normal operation of the service, including but not limited to the following: Alarm ID: Each alarm has a unique number to facilitate tracking and querying of the alarm.

[0045] Alarm time: Accurately record the time when the alarm occurs.

[0046] Alarm severity: Typically categorized as severe, important, or general, these levels reflect the urgency of the alarm and the severity of the potential service impact. For example, an alarm indicating complete service unavailability might be classified as severe, while an alarm indicating a slight anomaly in a performance indicator might be classified as general.

[0047] Alarm description: Describes the specific content of the alarm in detail and clearly points out the problem with the service, such as the server CPU usage exceeding 80%.

[0048] Associated Services: Indicates the services that are affected.

[0049] Resource usage records record the usage of various resources during the operation of the service, including hardware resources (such as CPU, memory, disk, network, etc.) and software resources (such as database connections, threads, file handles, etc.), including but not limited to the following: Recording time: The time when resource usage data is collected.

[0050] Service ID: Indicates which service the resource usage record belongs to.

[0051] Resource type: Specify the type of resource, such as CPU resources, memory resources, disk resources, network resources, etc.

[0052] Usage values: Records the specific usage values ​​of various resources, such as the number of CPU cores and CPU usage rate of CPU resources, the usage and usage rate of memory resources, the usage rate, usage, read and write latency of disk resources, and the usage rate, usage, and network latency of network resources.

[0053] In an embodiment of the present invention, the fault location reference information required by the fault location model for fault location includes the service call relationship chain associated with the fault event, and also includes at least one of the service change record, service alarm record and resource usage record of the service in the service call relationship chain.

[0054] Among them, the workflow is executed by the workflow engine to obtain the service call relationship chain associated with the fault event (that is, the service call relationship chain where the service where the fault event occurs is located). In addition, at least one of the following is obtained: a service change record of the service in the service call relationship chain that matches the fault type within a first preset period (which can be configured according to actual needs, for example, the first preset period is configured as: seven natural days with the fault time point of the fault event as the end time point), a service alarm record that matches the fault type within a second preset period (which can also be configured according to actual needs, can be configured to be the same as the first preset period, or can be configured to be different from the first preset period, for example, the second preset period is configured as: three natural days with the fault time point of the fault event as the end time point), and a resource usage record that matches the fault type at the fault time point.

[0055] Alternatively, in one embodiment, please refer to Figure 4 Based on the fault location reference information and fault description information, the fault location model is used to locate the fault. After obtaining the fault cause of the fault event, the following steps are also included: Determine the location accuracy of the fault cause. If the location accuracy does not reach the preset location accuracy, the workflow is updated through the workflow agent based on the fault cause and fault description information. Execute the update workflow through the workflow engine to obtain updated fault location reference information; Based on the updated fault location reference information, fault description information and fault cause, the fault location is performed again using the fault location model until the target fault cause is obtained with a location accuracy that reaches a preset location accuracy.

[0056] Location accuracy refers to the degree of precision with which the root cause of a fault can be pinpointed. For example, consider the following three fault causes: service change, service code change, and service response logic optimization. For the "service change" fault cause, while it's clear that the fault was caused by a service change, ruling out other possible causes, it's still unclear what type of service change it was, and even less clear which service change it was. Further investigation is still needed, resulting in the lowest location accuracy of the three. For the "service code change" fault cause, it's clear that the fault was caused by both a service change and a single service code change, but it's still unclear which service code change it was, and further investigation is also needed, giving it an intermediate location accuracy. For the "service response logic optimization" fault cause, it's clear that the fault was caused by both a service change and a single service code change, specifically a service code change to optimize the service response logic. This directly identifies the root cause of the fault, eliminating the need for further investigation and resulting in the highest location accuracy of the three.

[0057] In an embodiment of the present invention, after fault location is performed through a fault location model based on fault location reference information and fault description information to obtain the fault cause of the fault event, the positioning accuracy of the fault cause is further determined, and it is judged whether the positioning accuracy of the fault cause reaches the preset positioning accuracy. If the positioning accuracy of the fault cause does not reach the preset positioning accuracy, a prompt word corresponding to the workflow agent is constructed based on the fault cause and the fault description information, and the prompt word is input into the workflow agent, thereby generating a new workflow through the workflow agent, which is recorded as an updated workflow.

[0058] For example, the prompt words constructed corresponding to the workflow agent can be: "The following fault event occurred in the business system {fill in the fault description information here}, and the fault location model has preliminarily located the cause of the fault {fill in the fault location reference information used by the fault location model for the last fault location} based on the fault location reference information {fill in the fault location reference information used by the fault location model for the last fault location}. Now it is necessary to improve the accuracy of locating the cause of the fault. Please regenerate a workflow that is adapted to the workflow engine {fill in the description information of the workflow engine here} so that the workflow engine can execute the newly generated workflow to obtain new fault location reference information for the fault location model to perform fault location again to improve the accuracy of locating the cause of the fault."

[0059] The workflow engine executes the update workflow to obtain new fault location reference information, which is recorded as the updated fault location reference information. Based on the updated fault location reference information, the fault description information and the fault cause, the fault location is performed again through the fault location model to update the fault cause until the target fault cause with the positioning accuracy reaching the preset positioning accuracy is obtained.

[0060] For example, the fault description information, updated fault location reference information, and the fault cause obtained from the previous fault location can be constructed as prompts and input into the fault location model. For example, the constructed prompts can be: "The following fault event {fill in the fault description information here} has occurred in the business system you are responsible for operating and maintaining, and the cause of the fault was preliminarily located in the previous location process {fill in the fault cause obtained by the fault location model during the last fault location}. Now the accuracy of locating the fault cause needs to be improved. Please use the following reference information {fill in the updated fault location reference information here} to perform fault location again to improve the accuracy of locating the fault cause."

[0061] In other embodiments, if the location accuracy of the fault cause obtained by the fault location model reaches a preset location accuracy, the fault cause is directly used as the target fault cause.

[0062] Optionally, in one embodiment, after performing fault location again using the fault location model based on the updated fault location reference information and fault description information until a target fault cause with a location accuracy reaching a preset location accuracy is obtained, the method further includes: According to the target fault cause, the fault location model is used to obtain the fault elimination strategy corresponding to the fault event.

[0063] In the embodiment of the present invention, after obtaining the target fault cause with a positioning accuracy that reaches a preset positioning accuracy, a fault troubleshooting strategy corresponding to the fault event is further generated through the fault location model to assist operation and maintenance personnel in quickly troubleshooting the fault.

[0064] For example, the following prompt words can be generated based on the target fault cause to instruct the fault location model to generate a troubleshooting strategy for troubleshooting the fault: "The business system you are responsible for operating and maintaining has experienced the following fault event {fill in the fault description here}. The fault cause {fill in the target fault cause here} has been located and now needs to be corrected. Please provide suggestions for resolving the fault." Optionally, in one embodiment, after performing fault location again using the fault location model based on the updated fault location reference information and fault description information until a target fault cause with a location accuracy reaching a preset location accuracy is obtained, the method further includes: Obtaining an input correction operation for a target fault cause, and correcting the target fault cause according to the correction operation to obtain a corrected fault cause; Update the fault location model based on the fault location reference information, fault description information, and correct the fault cause.

[0065] For example, after obtaining a target fault cause with a positioning accuracy that reaches a preset positioning accuracy, on the one hand, a fault display interface including the target fault cause can be displayed to the operation and maintenance personnel (the fault display interface can also include information related to the fault event, such as fault description information and fault location reference information), for confirmation or further investigation by the operation and maintenance personnel. If the operation and maintenance personnel determine that the target fault cause is inconsistent with the fault cause obtained through manual investigation, they can enter a correction operation through the fault display interface. When the correction operation is executed, it is used to correct the target fault cause to the fault cause obtained through manual investigation by the operation and maintenance personnel. On the other hand, after receiving the correction operation input for the target fault cause through the fault display interface, the target fault cause is corrected according to the correction operation, and the corrected target fault cause is recorded as the corrected fault cause. The corrected fault cause is also the fault cause obtained through manual investigation by the operation and maintenance personnel. Furthermore, after the cumulative number of correction operations reaches a threshold, for a correction operation, a new training sample is constructed based on the fault location reference information and fault description information corresponding to the correction operation, and the correction fault cause corresponding to the correction operation is used as the training label of the new training sample. In this way, multiple sets of new training sample-training label pairs can be obtained; these new training samples-training labels are used to update the fault location model.

[0066] As can be seen from the above, the fault location solution provided by the present invention obtains fault description information of a fault event that occurs in a business system; obtains the workflow corresponding to the fault description information through a workflow agent; executes the workflow through a workflow engine to obtain fault location reference information; and performs fault location based on the fault location reference information and the fault description information through a fault location model to obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning and training a large model. In this way, when a fault event occurs in a business system, the fault description information and the fault location reference information can be automatically obtained, and the fault location model can complete the fault location of the fault event and obtain the fault cause of the fault event. This achieves fully automated fault location for fault events that occur in the business system, improving fault location efficiency compared to manual fault location methods.

[0067] In one embodiment, a fault location device is provided, which corresponds one-to-one to the fault location method in the above embodiment. Figure 5 As shown, the fault location device includes a description information acquisition module 210, a workflow acquisition module 220, a reference information acquisition module 230 and a fault location module 240. The functional modules are described in detail as follows: Description information acquisition module 210, used to obtain fault description information of fault events occurring in the business system; The workflow acquisition module 220 is used to acquire the workflow corresponding to the fault description information through the workflow agent; Reference information acquisition module 230, used to execute the workflow through the workflow engine to obtain fault location reference information; The fault location module 240 is used to locate the fault based on the fault location reference information and the fault description information through a fault location model to obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning the large model.

[0068] Optionally, in one embodiment, the workflow acquisition module 220 is used to classify fault events according to the fault description information to obtain the fault type of the fault event; and obtain the workflow corresponding to the fault description information through the workflow agent matching the fault type.

[0069] Optionally, in one embodiment, the fault location module 240 is used to obtain fault location knowledge corresponding to the fault type, where the fault location knowledge records the fault location strategy corresponding to the fault type; based on the fault location reference information, the fault description information and the fault location knowledge, the fault location is performed through the fault location model to obtain the fault cause of the fault event.

[0070] Optionally, in one embodiment, the reference information acquisition module 230 is used to execute the workflow through the workflow engine to obtain the service call relationship chain associated with the fault event, and obtain at least one of the service change records, service alarm records and resource usage records of the service in the service call relationship chain, and use the obtained service call relationship chain, service change records, service alarm records and resource usage records as fault location reference information.

[0071] Optionally, in one embodiment, the workflow acquisition module 220 is also used to determine the positioning accuracy of the fault cause. If the positioning accuracy does not reach the preset positioning accuracy, the updated workflow is obtained through the workflow agent based on the fault cause and fault description information; the reference information acquisition module 230 is also used to execute the updated workflow through the workflow engine to obtain updated fault location reference information; the fault location module 240 is also used to perform fault location again through the fault location model based on the updated fault location reference information, fault description information and fault cause, until the target fault cause with the positioning accuracy reaching the preset positioning accuracy is obtained.

[0072] Optionally, in one embodiment, the fault locating device further includes a troubleshooting strategy acquisition module, configured to acquire a fault troubleshooting strategy corresponding to the fault event through the fault locating model according to the target fault cause.

[0073] Optionally, in one embodiment, the fault location device further includes a model updating module, which is used to obtain an input correction operation for a target fault cause, and to correct the target fault cause according to the correction operation to obtain a corrected fault cause; and to update the fault location model according to the fault location reference information, the fault description information and the corrected fault cause.

[0074] The specific definition of the fault location device can be found in the definition of the fault location method above and will not be repeated here. Each module in the aforementioned fault location device may be implemented in whole or in part via software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in hardware form, or may be stored in a computer device memory in software form, allowing the processor to call and execute the corresponding operations of each module.

[0075] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, memory, network interface, and database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client via a network connection. When the computer program is executed by the processor, the steps of the fault location method in the above embodiment are implemented.

[0076] In one embodiment, a computer device is provided. The computer device may be a client, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, memory, a network interface, a display screen, and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile 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 the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with a target external server via a network connection. When executed by the processor, the computer program implements the steps of the fault location method in the above embodiment.

[0077] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the fault location method in the above embodiment are implemented, such as: Obtain fault description information of fault events that occur in the business system; Obtaining the workflow corresponding to the fault description information through the workflow agent; Execute the workflow through the workflow engine to obtain fault location reference information; According to the fault location reference information and fault description information, the fault location model is used to locate the fault and obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning the large model.

[0078] In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the steps of the fault location method in the above embodiment are implemented, such as: Obtain fault description information of fault events that occur in the business system; Obtaining the workflow corresponding to the fault description information through the workflow agent; Execute the workflow through the workflow engine to obtain fault location reference information; According to the fault location reference information and fault description information, the fault location model is used to locate the fault and obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning the large model.

[0079] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can be found in the relevant descriptions of the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0080] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware using a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0081] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0082] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

[0083] It should be noted that the non-Company's software tools or components appearing in the embodiments of the present invention are merely examples and do not represent actual use.

Claims

1. A fault location method, characterized in that: include: Obtain fault description information of fault events that occur in the business system; Acquire a workflow corresponding to the fault description information through a workflow agent; Executing the workflow through a workflow engine to obtain fault location reference information; According to the fault location reference information and the fault description information, fault location is performed using a fault location model to obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning and training a large model.

2. The fault location method according to claim 1, characterized in that: The obtaining of the workflow corresponding to the fault description information by a workflow agent includes: Classifying the fault event according to the fault description information to obtain the fault type of the fault event; The workflow corresponding to the fault description information is obtained through a workflow agent that matches the fault type.

3. The fault location method according to claim 2, characterized in that: The performing fault location by using a fault location model according to the fault location reference information and the fault description information to obtain the fault cause of the fault event includes: Acquire fault location knowledge corresponding to the fault type, wherein the fault location knowledge records a fault location strategy corresponding to the fault type; According to the fault location reference information, the fault description information and the fault location knowledge, the fault location model is used to perform fault location to obtain the fault cause of the fault event.

4. The fault location method according to claim 1, characterized in that: The step of executing the workflow through a workflow engine to obtain fault location reference information includes: Executing the workflow through a workflow engine to obtain a service call relationship chain associated with the fault event, and obtaining at least one of a service change record, a service alarm record, and a resource usage record of a service in the service call relationship chain, The obtained service call relationship chain, at least one of the service change record, service alarm record and resource usage record is used as the fault location reference information.

5. The fault location method according to any one of claims 1 to 4, characterized in that: After performing fault location by using a fault location model based on the fault location reference information and the fault description information to obtain the fault cause of the fault event, the method further includes: Determining the positioning accuracy of the fault cause, and if the positioning accuracy does not reach a preset positioning accuracy, obtaining an updated workflow through the workflow agent according to the fault cause and the fault description information; Executing the update workflow through the workflow engine to obtain updated fault location reference information; According to the updated fault location reference information, the fault description information and the fault cause, the fault location is performed again using the fault location model until a target fault cause with a location accuracy reaching the preset location accuracy is obtained.

6. The fault location method according to claim 5, characterized in that: The method further includes: performing fault location again using the fault location model according to the updated fault location reference information and the fault description information until a target fault cause with a location accuracy reaching the preset location accuracy is obtained; According to the target fault cause, a fault elimination strategy corresponding to the fault event is obtained through the fault location model.

7. The fault location method according to claim 6, characterized in that: The method further includes: performing fault location again using the fault location model according to the updated fault location reference information and the fault description information until a target fault cause with a location accuracy reaching the preset location accuracy is obtained; Obtaining an input correction operation for the target fault cause, and correcting the target fault cause according to the correction operation to obtain a corrected fault cause; The fault location model is updated according to the fault location reference information, the fault description information and the corrected fault cause.

8. A fault location device, characterized in that: include: Description information acquisition module, used to obtain fault description information of fault events occurring in the business system; A workflow acquisition module, configured to acquire the workflow corresponding to the fault description information through a workflow agent; A reference information acquisition module, configured to execute the workflow through a workflow engine to obtain fault location reference information; A fault location module is used to locate the fault based on the fault location reference information and the fault description information through a fault location model to obtain the fault cause of the fault event. The fault location model is obtained by fine-tuning the large model.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the fault location method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the fault location method according to any one of claims 1 to 7 is implemented.

Images