Diagnosis method and device of cloud service system, server and storage medium

Abstract

The invention discloses a diagnosis method and device of a cloud service system, a server and a storage medium, the diagnosis method of the cloud service system is applied to a diagnosis system, the diagnosis system comprises a task management unit and an analysis unit, and the method comprises the following steps: the task management unit responds to a diagnosis request of a target service in a target cloud service system; obtaining an initial diagnosis result of a target service in the target cloud service system; the analysis unit is used for determining a target processing suggestion matched with target abnormal information according to the target abnormal information existing in the target business in the initial diagnosis result; and the task management unit determines a target diagnosis result of the target business according to the initial diagnosis result and the target processing suggestion. According to the method, the target exception information and the target processing suggestion are taken as the diagnosis result, so that the user can process the exception in the cloud service system according to the target processing suggestion in the diagnosis result, and the exception processing difficulty of the cloud service system is reduced.

Description

Technical Field

[0001] The present application relates to the field of computer technologies, and more specifically, to a diagnosis method, apparatus, server, and storage medium for a cloud service system. Background Art

[0002] With the rapid development of cloud computing technologies, more and more enterprises and organizations have migrated their services to cloud service systems. However, due to the complexity and scale of these service systems, diagnosing problems such as security, faults, and performance has become more difficult. Summary of the Invention

[0003] In view of the above problems, the present application provides a diagnosis method, apparatus, server, and storage medium for a cloud service system.

[0004] In a first aspect, an embodiment of the present application provides a diagnosis method for a cloud service system, which is applied to a diagnosis system. The diagnosis system includes a task management unit and an analysis unit. The method includes: the task management unit, in response to a diagnosis request for a target service in a target cloud service system, obtains an initial diagnosis result of the target service in the target cloud service system; the analysis unit determines a target processing suggestion that matches the target abnormal information according to the target abnormal information existing in the target service in the initial diagnosis result; the task management unit determines a target diagnosis result of the target service according to the initial diagnosis result and the target processing suggestion.

[0005] In some embodiments, the analysis unit includes a large language model, and the large language model is connected to a knowledge base. The knowledge base includes professional information of multiple services and professional processing suggestions corresponding to different abnormal information of multiple services. The analysis unit determines a target processing suggestion that matches the target abnormal information according to the target abnormal information existing in the target service in the initial diagnosis result, including: the large language model in the analysis unit identifies the target abnormal information and obtains a matching result of a processing suggestion that matches the target abnormal information. The large language model is obtained by adjusting a target model trained for information identification and matching using sample abnormal information and sample processing suggestions that match the sample abnormal information; if the matching result indicates successful matching, the processing suggestion obtained by matching in the large language model is determined as the target processing suggestion; if the matching result indicates unsuccessful matching, the large language model obtains target professional information corresponding to the target abnormal information and a target professional processing suggestion corresponding to the target abnormal information from the knowledge base according to the target abnormal information; the large language model determines a target processing suggestion corresponding to the target abnormal information according to the target professional information and the target professional processing suggestion.

[0006] In some embodiments, if the matching result indicates unsuccessful matching, the large language model obtains, according to the target abnormal information, target professional information corresponding to the target abnormal information and target professional processing suggestions corresponding to the target abnormal information from the knowledge base, including: if the matching result indicates unsuccessful matching, the large language model determines an abnormal information vector corresponding to the target abnormal information according to the target abnormal information; and determines, according to the abnormal information vector, target professional information corresponding to the target abnormal information and target professional processing suggestions corresponding to the target abnormal information from the knowledge base.

[0007] In some embodiments, the analysis unit determines a target processing suggestion matching the target abnormal information according to the target abnormal information existing in the target service in the initial diagnosis result, including: the analysis unit preprocesses the initial diagnosis result to obtain initial abnormal information corresponding to the target service in the initial diagnosis result; the analysis unit formats the initial abnormal information according to a preset data format to obtain the target abnormal information; and the analysis unit determines a target processing suggestion matching the target abnormal information according to the target abnormal information.

[0008] In some embodiments, the task management unit, in response to a diagnosis request for a target service in a target cloud service system, obtains an initial diagnosis result of the target service in the target cloud service system, including: the task management unit, in response to the diagnosis request for the target service in the target cloud service system, obtains target virtual private cloud information corresponding to the target service; if there is no information about a running diagnostic agent cloud host in the target virtual private cloud information, issues a creation instruction for the diagnostic agent cloud host to create the diagnostic agent cloud host in the target virtual private cloud platform corresponding to the target virtual private cloud information, where the diagnostic agent cloud host is used to execute a diagnostic task in the target cloud service system; the task management unit issues a diagnostic task corresponding to the target service to the diagnostic agent cloud host and obtains the initial diagnosis result after the target diagnostic agent cloud host diagnoses the target task based on the diagnostic task; if there is information about a running diagnostic agent cloud host in the target virtual private cloud information, the task management unit issues a diagnostic task corresponding to the target service to the target diagnostic agent cloud host corresponding to the diagnostic agent cloud host information and obtains the initial diagnosis result after the target diagnostic agent cloud host diagnoses the target task based on the diagnostic task.

[0009] In some embodiments, before the task management unit obtains the target virtual private cloud information corresponding to the target service in response to a diagnostic request for the target service in the target cloud service system, the method further includes: the task management unit determines the cloud environment information of the target cloud service system in response to an information confirmation instruction for the target cloud service system; the task management unit performs an abstraction process on the communication interface corresponding to the cloud environment information to obtain an abstracted communication interface; the task management unit obtains the target virtual private cloud information corresponding to the target service in response to a diagnostic request for the target service in the target cloud service system, including: the task management unit obtains the target virtual private cloud information corresponding to the target service through the abstracted communication interface in response to a diagnostic request for the target service in the target cloud service system.

[0010] In a second aspect, an embodiment of the present application provides a diagnostic method for a cloud service system, which is applied to the cloud service system. The method includes: in response to a confirmation instruction for a target service, sending a diagnostic request for the target service to the task management unit of the diagnostic system; diagnosing the target service based on the diagnostic task corresponding to the diagnostic request sent by the task management unit to obtain an initial diagnostic result, and feeding back the initial diagnostic result to the task management unit, so that the analysis unit in the diagnostic system determines a target processing suggestion that matches the target abnormal information according to the target abnormal information existing in the target service in the initial diagnostic result, and the task management unit determines the target diagnostic result of the target service according to the initial diagnostic result and the target processing suggestion.

[0011] In some embodiments, the cloud service system includes multiple virtual private cloud platforms, and at least one service runs in the virtual private cloud platform. The diagnosing the target service based on the diagnostic task corresponding to the diagnostic request sent by the task management unit to obtain an initial diagnostic result includes: if a diagnostic proxy cloud host for data diagnosis is included in the target virtual private cloud platform corresponding to the target service, using the diagnostic proxy cloud host to diagnose the target service and obtaining the initial diagnostic result; if the diagnostic proxy cloud host is not included in the target virtual private cloud platform, obtaining a creation instruction for the diagnostic proxy cloud host issued by the diagnostic system, and creating the diagnostic proxy cloud host according to the creation instruction; using the diagnostic proxy cloud host to diagnose the target service and obtaining the initial diagnostic result.

[0012] In a third aspect, an embodiment of the present application provides a diagnostic device for a cloud service system, which is applied to a diagnostic system. The diagnostic system includes a task management unit and an analysis unit. The device includes: an initial diagnostic result acquisition module, configured to, for the task management unit, in response to a diagnostic request for a target service in a target cloud service system, acquire an initial diagnostic result of the target service in the target cloud service system; a target processing suggestion matching module, configured to, for the analysis unit, determine a target processing suggestion matching the target abnormal information according to the target abnormal information existing in the target service in the initial diagnostic result; and a target diagnostic result determination module, configured to, for the task management unit, determine a target diagnostic result of the target service according to the initial diagnostic result and the target processing suggestion.

[0013] In a fourth aspect, an embodiment of the present application provides a diagnostic device for a cloud service system, which is applied to a cloud service system. The device includes: a diagnostic request sending module, configured to, in response to a confirmation instruction for a target service, send a diagnostic request for the target service to the task management unit of the diagnostic system; and an initial diagnostic result obtaining module, configured to, based on a diagnostic task corresponding to the diagnostic request sent by the task management unit, diagnose the target service to obtain an initial diagnostic result, and feed back the initial diagnostic result to the task management unit, so that the analysis unit in the diagnostic system determines a target processing suggestion matching the target abnormal information according to the target abnormal information existing in the target service in the initial diagnostic result, and the task management unit determines a target diagnostic result of the target service according to the initial diagnostic result and the target processing suggestion.

[0014] In a fifth aspect, an embodiment of the present application provides a server, which includes: one or more processors; a memory; and one or more application programs, where the one or more application programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs are configured to execute the diagnostic method for a cloud service system provided in the first aspect and the diagnostic method for a cloud service system provided in the second aspect.

[0015] In a sixth aspect, an embodiment of the present application provides a computer-readable storage medium, in which program code is stored, and the program code can be called by a processor to execute the diagnostic method for a cloud service system provided in the first aspect and the diagnostic method for a cloud service system provided in the second aspect.

[0016] In the solution provided by this application, after obtaining the initial diagnosis result of a diagnosis task, the processing suggestions matching the target abnormal information are determined according to the target abnormal information in the initial diagnosis result, and the target abnormal information and the target processing suggestions are used as the diagnosis result, so that the user can process the anomalies in the cloud business system according to the target processing suggestions in the diagnosis result, reducing the difficulty of anomaly processing in the cloud business system. Brief Description of the Drawings

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of this application. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 The flowchart of the diagnosis method for the cloud business system on the diagnosis system provided by the embodiment of this application is shown.

[0019] Figure 2 The flowchart of the diagnosis method for the cloud business system provided by another embodiment of this application is shown.

[0020] Figure 3 The flowchart of the diagnosis method for the cloud business system on the cloud business system provided by this application is shown.

[0021] Figure 4 The data interaction diagram of the diagnosis method for the cloud business system provided by the embodiment of this application is shown.

[0022] Figure 5 The data processing flowchart of the LLM model in the embodiment of this application is shown.

[0023] Figure 6 The structural block diagram of the diagnosis device for the cloud business system on the diagnosis system provided by the embodiment of this application is shown.

[0024] Figure 7 The structural block diagram of the diagnosis device for the cloud business system on the cloud business system provided by the embodiment of this application is shown.

[0025] Figure 8 The structural block diagram of the server for executing the diagnosis method for the cloud business system according to the embodiment of this application provided by the embodiment of this application is shown.

[0026] Figure 9 The storage medium for storing or carrying the program code for implementing the diagnosis method for the cloud business system according to the embodiment of this application provided by the embodiment of this application is shown. Detailed Embodiments

[0027] To enable those skilled in the art to better understand the solution of this application, the technical solution in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application.

[0028] In response to the problems in the background art, the inventors have proposed a diagnosis method, device, server, and storage medium for a cloud service system provided in the embodiments of this application. After obtaining the initial diagnosis result of a diagnosis task, a processing suggestion matching the target abnormal information in the initial diagnosis result is determined according to the target abnormal information, and the target abnormal information and the target processing suggestion are used as the diagnosis result, so that the user can process the abnormality in the cloud service system according to the target processing suggestion in the diagnosis result, reducing the difficulty of abnormal processing in the cloud service system.

[0029] The diagnosis method, device, server, and storage medium for a cloud service system provided in the embodiments of this application will be described in detail below in conjunction with the accompanying drawings.

[0030] Please refer to Figure 1 , Figure 1 which shows a schematic flowchart of the diagnosis method for a cloud service system on a diagnosis system provided in the embodiments of this application. The method is applied to a diagnosis system, and the diagnosis system includes a task management unit and an analysis unit. The diagnosis method for this cloud service system can be applied to a diagnosis device 300 for a cloud service system as shown in Figure 6 and a server 100 configured with the diagnosis device 300 for the cloud service system ( Figure 8 ).

[0031] The diagnosis method for the cloud service system may specifically include the following steps:

[0032] Step S110: The task management unit, in response to a diagnosis request for a target service in a target cloud service system, obtains an initial diagnosis result of the target service in the target cloud service system.

[0033] Among them, the diagnostic system can be a system with diagnostic functions, or a system with diagnostic result analysis, or a system with both diagnostic functions and diagnostic result analysis. In the diagnostic system, different units can perform different tasks. Taking the diagnostic system including a task management unit and an analysis unit as an example, as the name implies, the task management unit is a unit that allocates and manages different processing tasks in the diagnostic system. The analysis unit is a unit that analyzes tasks and / or task results. In the diagnostic system, the task management unit allocates tasks so that other execution units work according to the allocated tasks. In the analysis unit, when other execution units complete tasks and obtain execution results, the analysis unit can further analyze the execution results to determine whether the execution results of the execution units meet the expected results, so as to monitor the operation of the diagnostic system. The functions of the diagnostic system and the working units included in the diagnostic system are not specifically limited herein.

[0034] The cloud service system refers to a service system used to process cloud services. The cloud service system can be a server or a computer, and is not specifically limited herein.

[0035] Furthermore, the initial diagnostic result can be obtained after the target cloud service system diagnoses the target service in response to the diagnostic request of the target service. At this time, the diagnostic system obtains the initial diagnostic result through the communication channel with the target cloud service system. The initial diagnostic result can also be that after the diagnostic system receives the diagnostic request of the target service in the target cloud service system, the diagnostic system diagnoses the target service. The diagnostic method of the target service and the acquisition method of the initial diagnostic result are determined according to the actual situation and are not specifically limited herein.

[0036] Step S120: The analysis unit determines a target processing suggestion that matches the target abnormal information according to the target abnormal information existing in the target service in the initial diagnostic result.

[0037] Among them, the determination of the target abnormal information can be determined according to the abnormal identifier in the initial diagnostic result, or the initial diagnostic result can be matched with the preset diagnostic result, and the diagnostic result that fails to match is determined as the target abnormal information. The determination of the target abnormal information is not limited herein.

[0038] Multiple processing suggestions are stored in the analysis unit. When it is necessary to match the target exception information, the multiple processing suggestions are directly matched with the target exception information. A database storing multiple processing suggestions can also be connected to the analysis unit. When it is necessary to match the exception information, the multiple processing suggestions in the database are called to match with the target exception information. The analysis unit may further include a trained network model. The network model can be trained using multiple matching methods and multiple processing suggestions, or the network model can be obtained by fine-tuning a pre-trained large model using exception information samples, preset matching methods, and multiple processing suggestions. When it is necessary to match the processing suggestions for the target exception information, the network model is directly used to match the target exception information. The matching method between the target exception information and the processing suggestions is not limited herein.

[0039] Further, the exception information includes, but is not limited to, port exception information, vulnerability information, endpoint exception information, system status exception information, and other exception information.

[0040] Step S130: The task management unit determines the target diagnosis result of the target service according to the initial diagnosis result and the target processing suggestion.

[0041] In this embodiment, the task management unit determines the diagnosis result of the current diagnosis request according to the target exception information and the target processing suggestion corresponding to the target exception information, so that the user can perform exception handling on the target service according to the processing suggestion information in the target diagnosis result.

[0042] Among them, the processing suggestion information can be pre-set in the knowledge base, or the knowledge base can generate the processing suggestion information according to the exception information. The processing suggestion information is not specifically limited herein.

[0043] Further, the matching method can be matching according to a preset rule, or matching according to a regular expression, which is not specifically limited herein.

[0044] The solution provided in this embodiment, after obtaining the initial diagnosis result of the diagnosis task, will also determine the processing suggestion that matches the target exception information according to the target exception information in the initial diagnosis result, and use the target exception information and the target processing suggestion as the diagnosis result, so that the user can use the target processing suggestion in the diagnosis result to handle the exception in the cloud service system, reducing the difficulty of exception handling in the cloud service system.

[0045] Please refer to Figure 2 , Figure 2 shows a schematic flowchart of a diagnosis method for a cloud service system provided by another embodiment of the present application. The container group deployed in the server further includes a controller container group.

[0046] Step S201: The task management unit, in response to a diagnosis request for the target service in the target cloud service system, obtains target virtual private cloud information corresponding to the target service.

[0047] Among them, a virtual private cloud (VPC) refers to a virtual network environment created on a public cloud platform. The VPC provides a logically isolated network space, enabling tenants (users) to build their own virtual networks on the cloud platform, including subnets, route tables, network access control lists, etc. The VPC can be securely interconnected with the tenant's local data center or other cloud services to achieve a hybrid cloud or multi-cloud architecture. Through the public cloud VPC, tenants can create and manage their own virtual networks on the cloud platform, flexibly configure the network topology, and provide a secure and reliable network environment for applications. The VPC can also protect tenants' data and applications from unauthorized access through network isolation and access control.

[0048] It should be noted that the VPC is equivalent to a local area network. Cloud hosts within the VPC can access hosts outside the VPC, but hosts outside the VPC cannot access cloud hosts within the VPC by default. If a cloud host within the VPC is configured with an elastic IP, then hosts outside the VPC can access the cloud host through the elastic IP. For business systems within the tenant's VPC, for security reasons, usually only critical cloud hosts (such as API gateways) are configured with elastic IPs.

[0049] In some embodiments, to be compatible with more types of cloud service systems, before diagnosing the target cloud service system, before the task management unit, in response to a diagnosis request for the target service in the target cloud service system, obtains target virtual private cloud information corresponding to the target service, the method further includes: the task management unit, in response to an information confirmation instruction for the target cloud service system, determines the cloud environment information of the target cloud service system; the task management unit performs abstraction processing on the communication interface corresponding to the cloud environment information to obtain an abstracted communication interface. So that the task management unit, in response to a diagnosis request for the target service in the target cloud service system, obtains target virtual private cloud information corresponding to the target service through the abstracted communication interface.

[0050] Among them, abstraction processing refers to extracting the essence of the communication interface from multiple different communication interfaces, so that the abstracted communication interface can be compatible with different communication interfaces.

[0051] In the embodiments of the present application, the communication interface is abstracted, so that the solution provided by the present application can be compatible with a variety of different communication interfaces, and thus compatible with different cloud service systems, expanding the processing scope of the cloud service system, reducing the processing difficulty of the cloud service system, and saving the user's cost.

[0052] In the embodiments of the present application, the user needs to first create a project for the target cloud service system in the diagnosis system. The project information includes name, description, cloud environment type, tenant name (ID), access key ID / secret access key (AKSK), or username and password, etc. After adding is completed, the diagnosis system can obtain the detailed information of the target cloud service system, and the user can also view this information on the page. To improve the personalization of diagnosis, when creating a diagnosis task, the user can select the diagnosis items, and then determine the diagnosis plug-ins corresponding to the diagnosis items, so as to use the diagnosis plug-ins for diagnosis in the subsequent diagnosis process. There is no need to configure during the diagnosis process, improving the automation of diagnosis.

[0053] Exemplarily, the diagnosis items include port diagnosis, vulnerability diagnosis, full-link diagnosis of cloud observation, endpoint diagnosis, and system status chaos diagnosis, etc. Each diagnosis item corresponds to a diagnosis plug-in. The diagnosis plug-in corresponding to port diagnosis is a port scanning plug-in, and the diagnosis plug-in corresponding to full-link diagnosis of cloud observation is an application full-link observation plug-in. The quantity and category of diagnosis items and diagnosis plug-ins are not limited herein. The diagnosis results of each diagnosis item can be independent or open to each other. If the vulnerability scanning plug-in needs to perform on some ports opened by the cloud host, the diagnosis results of the port scanning plug-in can be used; the application full-link observation plug-in can identify the process list of each cloud host and the information of the used database / middleware, and the system status chaos diagnosis plug-in can read the results of the application full-link observation plug-in to arrange some specific test cases.

[0054] Step S202: If there is no information about the running diagnosis agent cloud host in the target virtual private cloud information, issue a creation instruction for the diagnosis agent cloud host to create the diagnosis agent cloud host in the target virtual private cloud platform corresponding to the target virtual private cloud information. The diagnosis agent cloud host is used to execute diagnosis tasks in the target cloud service system.

[0055] Step S203: The task management unit issues a diagnosis task corresponding to the target service to the diagnosis agent cloud host, and obtains the initial diagnosis result after the target diagnosis agent cloud host diagnoses the target task based on the diagnosis task.

[0056] Among them, the diagnostic proxy cloud host information refers to the diagnostic proxy cloud host used to proxy the diagnostic system for diagnostic processing. The diagnostic proxy cloud host can be a cloud host only used for diagnosis, or a cloud host running a diagnostic proxy program. No specific limitation is imposed on the diagnostic proxy cloud host here.

[0057] In the embodiment, when there is no running diagnostic proxy cloud host in the target virtual private cloud platform, the diagnosis of the target service cannot be performed at this time. Therefore, the diagnostic system will send a creation instruction for the diagnostic proxy cloud host to the target virtual private cloud platform to newly create a diagnostic proxy cloud host.

[0058] Step S204: If there is running diagnostic proxy cloud host information in the target virtual private cloud information, the task management unit sends a diagnostic task corresponding to the target service to the target diagnostic proxy cloud host corresponding to the diagnostic proxy cloud host information, and obtains the initial diagnostic result after the target diagnostic proxy cloud host diagnoses the target task based on the diagnostic task.

[0059] In the embodiment, when there is a running diagnostic proxy cloud host in the target virtual private cloud platform, the diagnostic proxy cloud host is used for diagnosis and an initial diagnostic result is obtained.

[0060] Step S205: The analysis unit preprocesses the initial diagnostic result to obtain the initial abnormal information corresponding to the target service in the initial diagnostic result.

[0061] Among them, the preprocessing includes text cleaning, text annotation, and data augmentation. Text cleaning includes removing irrelevant information, such as privacy data, meaningless characters, etc. Text annotation includes classifying and annotating the problems in the report and providing corresponding handling suggestions. Data augmentation includes possibly needing to expand the dataset through data augmentation techniques, such as synonym replacement, sentence restructuring, etc. No specific limitation is imposed on the preprocessing method here.

[0062] Formatting processing refers to converting the initial abnormal information into data in a preset data format, and then obtaining the target abnormal data.

[0063] Step S206: The analysis unit performs formatting processing on the initial abnormal information according to the preset data format to obtain the target abnormal information.

[0064] Among them, the method for obtaining the abnormal information vector can be to analyze the target abnormal information according to grammar and semantics to obtain an abnormal information vector corresponding to the target abnormal information. The method for obtaining the abnormal information vector can also be to input the target abnormal information into a vector acquisition model to obtain the abnormal information vector. No specific limitation is imposed on the method for obtaining the abnormal information vector here.

[0065] Exemplarily, to improve the efficiency of vector retrieval, Redissearch can be used as the vector database, or ElasticSearch can be used as the vector database.

[0066] Furthermore, the matching method can be cosine similarity or Euclidean distance matching method. The matching method between the abnormal information vector and the target processing suggestion information is not specifically limited herein.

[0067] Step S207: The large language model in the analysis unit identifies the target abnormal information and obtains a matching result of the processing suggestion that matches the target abnormal information. The large language model is obtained by adjusting the target model trained for information identification and matching using sample abnormal information and sample processing suggestions that match the sample abnormal information. The analysis unit includes a large language model, and the large language model is connected to the knowledge base. The knowledge base includes professional information of multiple services and professional processing suggestions corresponding to different abnormal information of multiple services.

[0068] Among them, the large language model (Large Language Model, LLM) can understand and generate human language, including text, dialogue, and answers to questions. The LLM is fine-tuned using sample diagnostic results to obtain an information identification model for identifying abnormal information in the diagnostic results. The LLM can be the Llama3 model or other open-source large language models, such as Baichuan / Qwen, etc. The large language model is not specifically limited herein.

[0069] The sample processing suggestions can come from historical abnormal reports. The historical abnormal report information includes but is not limited to collecting historical system reports, including but not limited to security reports, performance monitoring logs, reliability analysis reports, etc. The public knowledge includes but is not limited to integrating relevant knowledge from open-source communities, technical forums, documents, etc. The practical solution information includes but is not limited to suggestions from experts in the field to which the target service belongs, disposal solutions, best practices, etc. The sample processing suggestions can also be preset processing suggestions. The acquisition of the sample processing suggestions is not specifically limited herein.

[0070] Step S208: If the matching result indicates a successful match, the processing established by the large language model through matching is determined as the target processing suggestion.

[0071] Step S209: If the matching result indicates an unsuccessful match, the large language model obtains the target professional information corresponding to the target abnormal information and the target professional processing suggestion corresponding to the target abnormal information from the knowledge base according to the target abnormal information.

[0072] In the embodiments of the present application, if the matching result indicates a successful match, it proves that the large language model includes a processing suggestion corresponding to the target abnormal information. If the matching result indicates an unsuccessful match, it means that the large language model does not include a processing suggestion corresponding to the target abnormal information. At this time, the large language model obtains the target professional knowledge and the target professional processing suggestion corresponding to the target abnormal information from the knowledge base according to the target abnormal information, and then determines the target processing suggestion corresponding to the target abnormal information, so that the user can process the target abnormal information of the diagnostic task according to the target processing suggestion.

[0073] Further, the method for obtaining the target professional knowledge and the target professional processing suggestion from the knowledge base can be to extract keywords from the target abnormal information, and match the target professional knowledge and the target professional processing suggestion from the knowledge base according to the extracted keywords. The obtaining method can also be a vector matching method. That is, according to the target abnormal information, an information vector corresponding to the target abnormal information is determined. And the information vectors of each professional knowledge and each professional processing suggestion in the knowledge base are obtained, and according to the similarity matching, the information vectors corresponding to the professional knowledge and the information vectors corresponding to the professional processing suggestion that meet the preset similarity threshold are determined. Then the target professional knowledge and the target professional processing suggestion matching the target abnormal information are determined. The method for obtaining the target professional knowledge and the target professional processing suggestion is not specifically limited herein.

[0074] In some embodiments, if the matching result indicates an unsuccessful match, the large language model obtains the target professional information corresponding to the target abnormal information and the target professional processing suggestion corresponding to the target abnormal information from the knowledge base according to the target abnormal information, including: if the matching result indicates an unsuccessful match, the large language model determines the abnormal information vector corresponding to the target abnormal information according to the target abnormal information. According to the abnormal information vector, the target professional information corresponding to the target abnormal information and the target professional processing suggestion corresponding to the target abnormal information are determined from the knowledge base.

[0075] In the embodiments of the present application, when the large language model fails to obtain the target processing suggestion matching the target abnormal information from the existing processing suggestions, the knowledge base connected to the large language model is used to further obtain the target processing suggestion corresponding to the target abnormal information. In the obtaining process, the large language model first vectorizes the target abnormal information to obtain the abnormal information vector corresponding to the target abnormal information, and then determines the target professional information and the target professional processing suggestion corresponding to the target abnormal information from the knowledge base according to the vector similarity matching method.

[0076] Step S210: The large language model determines a target handling suggestion corresponding to the target abnormal information according to the target professional information and the target professional handling suggestion.

[0077] In an embodiment of the present application, the target handling suggestion may include relevant professional information, so that when the user obtains the target handling suggestion, they can more clearly understand the relevant knowledge of the target abnormal information and improve the user's ability to handle abnormalities.

[0078] Step S211: The task management unit determines a target diagnosis result of the target service according to the initial diagnosis result and the target handling suggestion.

[0079] For a detailed description of step S211, please refer to step S130 and will not be elaborated here.

[0080] In some embodiments, after determining the target diagnosis result of the target service, to improve the user's visualization experience, the target diagnosis result is rendered to obtain a diagnosis result report for the user to preview and download.

[0081] The solution provided in this embodiment first uses the analysis unit to identify the formatted abnormal information in the initial diagnosis result, which can improve the identification efficiency and accuracy. Then, the large language model in the analysis unit is used to match the handling suggestion information of the abnormal information to obtain a handling suggestion corresponding to the target abnormal information, and the abnormal information and the handling suggestion information are jointly used as the diagnosis result. Using the large language model and the knowledge base connected to the large language model to determine the handling suggestion corresponding to the abnormal information in the diagnosis result can improve the diagnosis efficiency and accuracy.

[0082] Please refer to Figure 3 , Figure 3 shows a schematic flowchart of a diagnosis method for a cloud service system provided by the present application. This method is applied to a cloud service system, and the diagnosis method of the cloud service system may specifically include the following steps:

[0083] Step S310: In response to a confirmation instruction for a target service, send a diagnosis request for the target service to the task management unit of the diagnosis system.

[0084] Among them, the confirmation instruction may be obtained by the cloud service system in response to the user's selection instruction, or the confirmation instruction may be generated when the cloud service system periodically diagnoses the target service. The specific method for obtaining the confirmation instruction is not specifically limited here.

[0085] Step S320: Based on the diagnostic task corresponding to the diagnostic request sent by the task management unit, diagnose the target service to obtain an initial diagnostic result, and feedback the initial diagnostic result to the task management unit, so that the analysis unit in the diagnostic system determines a target processing suggestion that matches the target abnormal information in the initial diagnostic result according to the target abnormal information existing in the target service, and the task management unit determines the target diagnostic result of the target service according to the initial diagnostic result and the target processing suggestion.

[0086] Among them, the initial diagnostic result can be obtained by the cloud service system diagnosing the target service for the diagnostic request, or by the cloud host for diagnosis in the cloud service system diagnosing the target service. Preferably, it is obtained by using the cloud host for diagnosis in the cloud service system to diagnose the target service.

[0087] In some embodiments, the cloud service system includes multiple virtual private cloud platforms, and at least one service runs in the virtual private cloud platform. The diagnosing the target service based on the diagnostic task corresponding to the diagnostic request sent by the task management unit to obtain an initial diagnostic result includes: if the target virtual private cloud platform corresponding to the target service includes a diagnostic proxy cloud host for data diagnosis, then use the diagnostic proxy cloud host to diagnose the target service and obtain the initial diagnostic result; if the diagnostic proxy cloud host is not included in the target virtual private cloud platform, then obtain the creation instruction for creating the diagnostic proxy cloud host issued by the diagnostic system, and create the diagnostic proxy cloud host according to the creation instruction; use the diagnostic proxy cloud host to diagnose the target service and obtain the initial diagnostic result.

[0088] In the embodiments of the present application, when the target virtual private cloud platform includes a diagnostic proxy cloud host for diagnosis, the diagnostic proxy cloud host is used for diagnosis. When the diagnostic proxy cloud host for diagnosis does not exist in the target virtual private cloud platform, the creation instruction issued by the diagnostic system is obtained to create the diagnostic proxy cloud host for diagnosis. For the specific methods of the diagnostic method and the diagnostic proxy cloud host, please refer to the foregoing embodiments and will not be elaborated here.

[0089] In the solution of the embodiment of the present application, the information recognition model is used to recognize the target abnormal information in the initial diagnostic result, and then the knowledge base configured in the information recognition model is used to match the processing suggestion information of the abnormal information, and the abnormal information and the processing suggestion information are jointly used as the diagnostic result. Using the knowledge base to determine the solution corresponding to the abnormal information in the diagnostic result can improve the diagnostic efficiency and accuracy.

[0090] Combined with the foregoing embodiments, please refer to Figure 4 ,Figure 4 The data interaction diagram of the diagnostic method of the cloud business system provided by the embodiment of the present application is shown. The diagnostic system includes a project management unit, a cloud environment abstraction unit, a task management unit, an analysis unit and a report management unit. Among them, the diagnostic system supports the diagnosis of multiple cloud business systems at the same time, so projects are used to correspond to different cloud business systems. Before the user diagnoses the cloud business system, it is necessary to create a project for it in the diagnostic system and configure the corresponding cloud environment information, including the cloud environment type, tenant ID, AKSK or username password, etc. During the subsequent diagnosis, the diagnostic system obtains the detailed information of the corresponding business system based on the above information, including the cloud host list, etc. The project management unit is used to manage the various pending and executed projects in the diagnostic system. The cloud environment abstraction unit is used for the diagnostic system to support business systems in different cloud environments, and the cloud environment is abstracted to facilitate subsequent expansion. The essence is to implement a series of specific interfaces based on the APIs of various cloud environments, such as obtaining the cloud host list in the tenant VPC, executing orchestration tasks in the cloud host, etc., to shield the differences in the underlying cloud environment. The diagnostic plug-in management module is used in the diagnostic system. In order to support the comprehensive diagnosis of the business system, a plug-in design is used. A plug-in is usually used to diagnose a certain aspect of the business system, which is convenient for the subsequent expansion of various diagnostic plug-ins. Each plug-in is actually a compressed package packaged as required, which contains the executable file, configuration file, etc. of the plug-in. The task management unit allows users to select any project on the diagnostic system to perform diagnostic tasks. After a simple configuration on the task wizard and clicking OK, a diagnostic task will be created, and the diagnostic system will execute the task in the background. Users can see the real-time execution status of each task on the diagnostic system page, such as success or failure, reason for failure, which step is executed, etc. In order to intelligently interpret the diagnostic results, the analysis unit gives professional disposal suggestions for various problems such as security, performance, reliability, etc., and conducts a comprehensive assessment of the overall situation of the business system. The diagnosis system adopts a solution that combines the currently popular LLM model with Retrieval-Augmented Generation (RAG). It pre-trains an LLM model and then enhances the model's capabilities based on RAG. RAG is essentially equivalent to plugging in a knowledge database to the LLM model. The content of the knowledge database can be continuously enriched, thereby improving the knowledge depth and breadth of the expert model without retraining the model. After each diagnostic task is successful, the user can preview the diagnostic results on the diagnostic system. In order to allow users to see more friendly diagnostic results, the diagnostic system will render the diagnostic results into beautiful reports, and users can also choose to download the report.

[0091] The cloud service system includes multiple virtual private cloud platforms. When the task management unit receives a diagnostic request for a target service in virtual private cloud platform 1, if it senses that there is a running diagnostic agent cloud host in virtual private cloud platform 1, it will send the diagnostic plugin and diagnostic task corresponding to the target service to the diagnostic agent cloud host in virtual private cloud platform 1 of the cloud service system, so that the diagnostic agent cloud host in virtual private cloud platform 1 can perform the diagnosis. Virtual private cloud platform 1 will periodically report logs during the diagnosis and update the task progress. After the diagnosis is completed, the diagnostic result file will be uploaded. If it does not sense that there is a running diagnostic agent cloud host in virtual private cloud platform 1, it will send the diagnostic agent cloud host program to virtual private cloud platform 1 to create a diagnostic agent cloud host. Different cloud hosts perform different diagnostic actions, and the diagnostic results of each cloud host can be shared, so as to facilitate further diagnosis based on the shared information during the diagnosis process and improve the accuracy of the diagnosis. When the diagnostic agent cloud host diagnoses the target service, it will send the diagnostic result to the report management unit, so that the report management unit can render a diagnostic result report based on the diagnostic result.

[0092] Please refer to Figure 5 , Figure 5 which shows the schematic diagram of the data processing flow of the LLM model in the embodiment of the present application. When the task management unit obtains the diagnostic result file uploaded by virtual private cloud platform 1, it will send the diagnostic result file to the analysis unit, first preprocess the diagnostic result file using the fine-tuning model in the analysis unit, and identify the list of abnormal information in the diagnostic result file. Based on the matching of the abnormal phenomena in the list of abnormal information with the knowledge base, the processing suggestion information corresponding to the abnormal information is obtained, and the association between the abnormal information and the processing suggestion information is established to obtain the final diagnostic result, and the final diagnostic result is sent to the report management system to render a diagnostic report according to the final diagnostic result for the user to preview and download.

[0093] Please refer to Figure 6, which shows a structural block diagram of a diagnostic device 300 for a cloud service system provided by an embodiment of the present application. The diagnostic device 300 for the cloud service system is applied to a diagnostic system, the diagnostic system includes a task management unit and an analysis unit, the diagnostic device 300 for the cloud service system is applied to a server 100, and the diagnostic device 300 for the cloud service system includes: an initial diagnostic result acquisition module 310, configured to, for the task management unit, in response to a diagnostic request for a target service in a target cloud service system, acquire an initial diagnostic result of the target service in the target cloud service system; a target processing suggestion matching module 320, configured to, for the analysis unit, determine a target processing suggestion matching the target abnormal information according to the target abnormal information existing in the target service in the initial diagnostic result; a target diagnostic result determination module 330, configured to, for the task management unit, determine a target diagnostic result of the target service according to the initial diagnostic result and the target processing suggestion.

[0094] In some embodiments of the present application, the analysis unit includes a large language model, and the large language model is connected to a knowledge base. The knowledge base of the large language model includes professional information of multiple services and professional processing suggestions corresponding to different abnormal information of multiple services. The target processing suggestion matching module 320 includes: a matching result determination sub-module, configured to, for the large language model in the analysis unit, identify the target abnormal information and obtain a matching result of a processing suggestion matching the target abnormal information. The large language model is obtained by adjusting a target model trained for information identification and matching by using sample abnormal information and sample processing suggestions matching the sample abnormal information; a first target processing suggestion determination sub-module, configured to, if the matching result indicates successful matching, determine the processing established by the large language model as the target processing suggestion; a target processing suggestion acquisition sub-module, configured to, if the matching result indicates unsuccessful matching, the large language model, according to the target abnormal information, obtain target professional information corresponding to the target abnormal information and target professional processing suggestions corresponding to the target abnormal information from the knowledge base; a second target processing suggestion determination sub-module, configured to, for the large language model, determine a target processing suggestion corresponding to the target abnormal information according to the target professional information and the target professional processing suggestions.

[0095] In some embodiments of the present application, the target processing suggestion acquisition sub-module includes: an abnormal information vector determination component, configured to, if the matching result indicates unsuccessful matching, the large language model, according to the target abnormal information, determine an abnormal information vector corresponding to the target abnormal information; a target processing suggestion acquisition determination component, configured to, according to the abnormal information vector, determine target professional information corresponding to the target abnormal information and target professional processing suggestions corresponding to the target abnormal information from the knowledge base.

[0096] In some embodiments of the present application, the diagnosis device 300 of the Chuyun business system further includes: an initial abnormal information acquisition module, which is used for the analysis unit to preprocess the initial diagnosis result to obtain the initial abnormal information corresponding to the target business in the initial diagnosis result; a formatting processing module, which is used for the analysis unit to format the initial abnormal information according to a preset data format to obtain the target abnormal information.

[0097] In some embodiments of the present application, the initial diagnosis result acquisition module 310 includes: a target virtual private cloud information acquisition sub-module, which is used for the task management unit to respond to the diagnosis request of the target business in the target cloud business system and acquire the target virtual private cloud information corresponding to the target business; a proxy cloud host creation sub-module, which is used to issue a creation instruction for the diagnostic proxy cloud host if there is no running diagnostic proxy cloud host information in the target virtual private cloud information, so as to create the diagnostic proxy cloud host in the target virtual private cloud platform corresponding to the target virtual private cloud information, and the diagnostic proxy cloud host is used to execute the diagnostic task in the target cloud business system; a diagnostic task issuing sub-module, which is used for the task management unit to issue a diagnostic task corresponding to the target business to the diagnostic proxy cloud host and acquire the initial diagnosis result after the target diagnostic proxy cloud host diagnoses the target task based on the diagnostic task; an initial diagnosis result acquisition sub-module, which is used for the task management unit to issue a diagnostic task corresponding to the target business to the target diagnostic proxy cloud host corresponding to the diagnostic proxy cloud host information if there is running diagnostic proxy cloud host information in the target virtual private cloud information, and acquire the initial diagnosis result after the target diagnostic proxy cloud host diagnoses the target task based on the diagnostic task.

[0098] In some embodiments of the present application, the initial diagnosis result acquisition module 310 further includes: a cloud environment information acquisition sub-module, which is used for the task management unit to determine the cloud environment information of the target cloud business system in response to an information confirmation instruction for the target cloud business system; an abstraction processing sub-module, which is used for the task management unit to perform abstraction processing on the communication interface corresponding to the cloud environment information to obtain the abstracted communication interface; the target virtual private cloud information acquisition sub-module includes: a target virtual private cloud information acquisition component, which is used for the task management unit to acquire the target virtual private cloud information corresponding to the target business through the abstracted communication interface in response to the diagnosis request of the target business in the target cloud business system.

[0099] Please refer to Figure 7, which shows a structural block diagram of a diagnostic device 400 for a cloud service system provided by an embodiment of the present application. The diagnostic device 400 for the cloud service system is applied to the server 100, and the diagnostic device for the cloud service system is applied to the cloud service system. The diagnostic device 400 for the cloud service system includes: a diagnostic request sending module 410, configured to send a diagnostic request for the target service to the task management unit of the diagnostic system in response to a confirmation instruction of the target service; an initial diagnostic result obtaining module 420, configured to diagnose the target service based on the diagnostic task corresponding to the diagnostic request sent by the task management unit, obtain an initial diagnostic result, and feedback the initial diagnostic result to the task management unit, so that the analysis unit in the diagnostic system determines a target processing suggestion matching the target abnormal information according to the target abnormal information existing in the target service in the initial diagnostic result, and the task management unit determines a target diagnostic result of the target service according to the initial diagnostic result and the target processing suggestion.

[0100] In some embodiments of the present application, the initial diagnostic result obtaining module 420 includes: a first initial diagnostic result obtaining sub-module, configured to, if a diagnostic proxy cloud host for data diagnosis is included in the target virtual private cloud platform corresponding to the target service, use the diagnostic proxy cloud host to diagnose the target service and obtain the initial diagnostic result; a diagnostic proxy cloud host creating module, configured to, if the diagnostic proxy cloud host is not included in the target virtual private cloud platform, obtain a creation instruction for creating a diagnostic proxy cloud host issued by the diagnostic system, and create the diagnostic proxy cloud host according to the creation instruction; a second initial diagnostic result obtaining sub-module, configured to use the diagnostic proxy cloud host to diagnose the target service and obtain the initial diagnostic result.

[0101] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0102] In several embodiments provided by the present application, the coupling between modules may be electrical, mechanical or other forms of coupling.

[0103] In addition, in each embodiment of the present application, each functional module may be integrated in a processing module, or each module may exist physically alone, or two or more modules may be integrated in one module. The above integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0104] A structural block diagram of a server provided by an embodiment of the present application. Please refer to Figure 8, which shows a structural block diagram of a server provided by an embodiment of the present application. The server 100 in the present application may include one or more of the following components: a processor 101, a memory 102, and one or more application programs, where one or more application programs may be stored in the memory 102 and configured to be executed by one or more processors 101, and one or more programs are configured to execute the methods described in the foregoing method embodiments.

[0105] The processor 101 may include one or more processing cores. The processor 101 connects various parts within the entire server 100 using various interfaces and lines. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 102, and by calling data stored in the memory 102, it executes various functions of the server 100 and processes data. Optionally, the processor 101 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 101 may integrate one or several combinations of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, and application programs, etc.; the GPU is responsible for rendering and drawing display content; the modem is used to process wireless communication. It can be understood that the above modem may not be integrated into the processor 101 and may be implemented separately through a communication chip.

[0106] The memory 102 may include random access memory (RAM) and may also include read-only memory. The memory 102 can be used to store instructions, programs, codes, code sets, or instruction sets. The memory 102 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for implementing at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the following various method embodiments, etc. The data storage area may also store data created during the use of the server 100 (such as phone books, audio and video data, chat record data, etc.).

[0107] Please refer to Figure 9, which shows a structural block diagram of a computer-readable storage medium provided by an embodiment of the present application. Program code is stored in the computer-readable storage medium 200, and the program code can be called by a processor to execute the method described in the above method embodiment.

[0108] The computer-readable storage medium 200 may be an electronic memory such as a flash memory, EEPROM (electrically erasable programmable read-only memory), EPROM, hard disk, or ROM. Optionally, the computer-readable storage medium 200 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 200 has a storage space for the program code 210 that executes any of the method steps in the above method. These program codes can be read out from or written into one or more computer program products. The program code 210 can be compressed in a suitable form, for example.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A diagnostic method for a cloud business system, characterized in that: Applied to a diagnostic system, the diagnostic system includes a task management unit and an analysis unit, and the method includes: The task management unit obtains an initial diagnosis result of the target service in the target cloud service system in response to a diagnosis request of the target service in the target cloud service system; The analyzing unit determines, according to target abnormality information existing in the target business in the initial diagnosis result, a target processing suggestion matching the target abnormality information; The task management unit determines a target diagnosis result of the target business according to the initial diagnosis result and the target processing suggestion.

2. The method according to claim 1, characterized in that The analysis unit includes a large language model, and the large language model is connected to a knowledge base, the knowledge base includes professional information of multiple businesses and professional processing suggestions corresponding to different abnormal information of the multiple businesses, and the analysis unit determines a target processing suggestion matching the target abnormal information according to the target abnormal information existing in the target business in the initial diagnosis result, including: The large language model in the analysis unit identifies the target abnormal information and obtains a matching result of a processing suggestion matching the target abnormal information, wherein the large language model is obtained by adjusting a target model that has been trained for information identification and matching using sample abnormal information and a sample processing suggestion matching the sample abnormal information; If the matching result indicates a successful match, the processing establishment obtained by matching in the large language model is determined as the target processing suggestion; If the matching result indicates that the match is unsuccessful, the large language model obtains target professional information corresponding to the target abnormal information and target professional processing suggestions corresponding to the target abnormal information from the knowledge base according to the target abnormal information; The large language model determines the target processing suggestion corresponding to the target abnormal information according to the target professional information and the target professional processing suggestion.

3. The method according to claim 2, characterized in that If the matching result indicates that the match is unsuccessful, the large language model obtains target professional information corresponding to the target abnormal information and target professional processing suggestions corresponding to the target abnormal information from the knowledge base according to the target abnormal information, including: If the matching result indicates that the match is unsuccessful, the large language model determines, based on the target abnormal information, an abnormal information vector corresponding to the target abnormal information; According to the abnormal information vector, target professional information corresponding to the target abnormal information and target professional processing suggestions corresponding to the target abnormal information are determined from the knowledge base.

4. The method according to claim 1, characterized in that: Before the analyzing unit determines, according to the target abnormality information existing in the target business in the initial diagnosis result, a target processing suggestion matching the target abnormality information, the method further includes: The analyzing unit pre-processes the initial diagnosis result to obtain initial abnormal information corresponding to the target business in the initial diagnosis result; The analysis unit formats the initial abnormal information according to a preset data format to obtain the target abnormal information.

5. The method according to any one of claims 1 to 4, characterized in that: The task management unit, in response to a diagnosis request of a target service in a target cloud service system, obtains an initial diagnosis result of a target service in the target cloud service system, including: The task management unit, in response to a diagnosis request of the target service in the target cloud service system, obtains target virtual private cloud information corresponding to the target service; If the target virtual private cloud information does not contain any running diagnostic proxy cloud host information, a diagnostic proxy cloud host creation instruction is issued to create the diagnostic proxy cloud host in the target virtual private cloud platform corresponding to the target virtual private cloud information, and the diagnostic proxy cloud host is used to perform diagnostic tasks in the target cloud business system; The task management unit sends a diagnostic task corresponding to the target business to the diagnostic proxy cloud host, and obtains the initial diagnostic result after the target diagnostic proxy cloud host diagnoses the target task based on the diagnostic task; If the target virtual private cloud information contains information about a running diagnostic proxy cloud host, the task management unit sends a diagnostic task corresponding to the target business to the target diagnostic proxy cloud host corresponding to the diagnostic proxy cloud host information, and obtains the initial diagnostic result after the target diagnostic proxy cloud host diagnoses the target task based on the diagnostic task.

6. The method according to claim 5, characterized in that Before the task management unit, in response to the diagnosis request of the target service in the target cloud service system, obtains the target virtual private cloud information corresponding to the target service, the method further includes: The task management unit determines the cloud environment information of the target cloud business system in response to the information confirmation instruction of the target cloud business system; The task management unit performs abstract processing on the communication interface corresponding to the cloud environment information to obtain the communication interface after abstract processing; The task management unit, in response to a diagnosis request of the target service in the target cloud service system, obtains target virtual private cloud information corresponding to the target service, including: The task management unit, in response to a diagnosis request of the target service in the target cloud service system, obtains target virtual private cloud information corresponding to the target service through the abstracted communication interface.

7. A data diagnosis method for a cloud business system, characterized in that: Applied to a cloud business system, the method includes: In response to a confirmation instruction of the target service, sending a diagnosis request for the target service to a task management unit of the diagnosis system; Based on the diagnostic task corresponding to the diagnostic request sent by the task management unit, the target business is diagnosed to obtain an initial diagnostic result, and the initial diagnostic result is fed back to the task management unit, so that the analysis unit in the diagnostic system determines a target processing suggestion that matches the target abnormality information based on the target abnormality information existing in the target business in the initial diagnostic result, and the task management unit determines the target diagnostic result of the target business based on the initial diagnostic result and the target processing suggestion.

8. The method according to claim 7, characterized in that The cloud service system includes a plurality of virtual private cloud platforms, at least one service is running in the virtual private cloud platform, and the diagnostic task corresponding to the diagnostic request sent by the task management unit is used to diagnose the target service to obtain an initial diagnostic result, including: If the target virtual private cloud platform corresponding to the target service includes a diagnostic proxy cloud host for data diagnosis, the target service is diagnosed using the diagnostic proxy cloud host to obtain the initial diagnostic result; If the target virtual private cloud platform does not include the diagnostic proxy cloud host, obtaining a diagnostic proxy cloud host creation instruction issued by the diagnostic system, and creating the diagnostic proxy cloud host according to the creation instruction; The target service is diagnosed using the diagnostic proxy cloud host, and the initial diagnostic result is obtained.

9. A data diagnostic device for a cloud business system, characterized in that: Applied to a diagnostic system, the diagnostic system includes a task management unit and an analysis unit, and the device includes: An initial diagnosis result acquisition module, used in the task management unit, responding to a diagnosis request of a target service in the target cloud service system, to acquire an initial diagnosis result of a target service in the target cloud service system; A target processing suggestion matching module, used in the analysis unit, determines a target processing suggestion matching the target abnormality information according to the target abnormality information existing in the target business in the initial diagnosis result; The target diagnosis result determination module is used in the task management unit to determine the target diagnosis result of the target business according to the initial diagnosis result and the target processing suggestion.

10. A data diagnostic device for a cloud business system, characterized in that: Applied to a cloud business system, the device comprises: A diagnosis request sending module, configured to send a diagnosis request for the target service to a task management unit of the diagnosis system in response to a confirmation instruction of the target service; An initial diagnostic result acquisition module is used to diagnose the target business based on the diagnostic task corresponding to the diagnostic request sent by the task management unit, obtain an initial diagnostic result, and feed back the initial diagnostic result to the task management unit, so that the analysis unit in the diagnostic system determines the target processing suggestion that matches the target abnormality information based on the target abnormality information existing in the target business in the initial diagnostic result, and the task management unit determines the target diagnostic result of the target business based on the initial diagnostic result and the target processing suggestion.

11. A server, characterized in that: The server comprises: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the method according to any one of claims 1 to 6 or the method according to any one of claims 7 to 8.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method according to any one of claims 1 to 6 or the method according to any one of claims 7 to 8.

Images