Online application log analysis and problem positioning method based on large model

Abstract

The invention particularly relates to an online application log analysis and problem positioning method based on a large model. According to the online application log analysis and problem positioning method based on the large model, a front-end and back-end real-time communication channel is established through the WebSocket technology, operation and maintenance personnel achieve real-time interaction with a server side through a Web side management interface, and instruction execution and log information management are achieved through a server side processing program; log files are quickly positioned through the management server and system application configuration information; and in combination with the analysis capability of a large model technology, performing deep analysis, summarization and problem positioning on a system execution result or a system log. According to the online application log analysis and problem positioning method based on the large model, the efficiency and accuracy of problem positioning can be remarkably improved, the operation and maintenance cost can be greatly saved, and the positioning difficulty is reduced, so that the system operation and maintenance level is comprehensively improved, and more efficient, stable and reliable operation and maintenance management support is provided for enterprises.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent operation and maintenance of communication networks, and in particular to an online application log analysis and problem location method based on a large model. Background Art

[0002] In the field of application system operation and maintenance management at telecommunications companies, with the increasing complexity of business and diversity of technology, the requirements for system security, reliability, and robustness are becoming increasingly stringent, and management and control measures are becoming increasingly stringent. However, traditional operation and maintenance management methods often expose numerous shortcomings when faced with critical tasks such as fault location and log analysis, seriously affecting operation and maintenance efficiency and system stability.

[0003] Specifically, traditional fault location methods rely primarily on the personal experience and manual troubleshooting of operations and maintenance personnel. This is not only inefficient but also prone to missing critical information due to inexperience or negligence, leading to inaccurate problem location and increasing the risk of business interruption. Regarding log analysis, massive amounts of log data require line-by-line review and analysis, which is time-consuming and prone to errors. Furthermore, traditional methods lack the ability to deeply mine and intelligently analyze log data, making it difficult to quickly extract valuable information from massive amounts of data, further exacerbating the complexity and challenges of operations and maintenance management.

[0004] These problems not only lead to a waste of resources, but may also cause business interruptions, causing unnecessary economic losses and reputation damage to the company.

[0005] Based on the above situation, the present invention proposes an online application log analysis and problem location method based on a large model, aiming to improve the efficiency and accuracy of operation and maintenance management, reduce operation and maintenance costs, and enhance the stability and reliability of the system. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, the present invention provides a simple and efficient online application log analysis and problem location method based on a large model.

[0007] The present invention is achieved through the following technical solutions:

[0008] A large-scale model-based online application log analysis and problem location method uses WebSocket technology to establish a real-time communication channel between the front-end and back-end. Operations and maintenance personnel can interact with the server in real time through the Web management interface, issue server commands in real time through the Web, and view execution results immediately. The server-side handler can achieve efficient management of command execution and log information.

[0009] By managing server and system application configuration information, log files can be quickly located. By combining the powerful analytical capabilities of large-scale model technology, in-depth analysis, summary, and problem location can be performed on system execution results or system logs.

[0010] The following steps are involved:

[0011] Step S1: Establish a server and system application configuration information management function to collect host information, system process information, deployment location, and log document information deployed by each system module, so as to quickly locate the log file and avoid the trouble of repeated entry, thereby improving operation efficiency;

[0012] Step S2: Build a large model processing system, deploy a large model base (such as deepseek32B, Qwen32b), and preset log analysis prompt words;

[0013] In step S2, when constructing a large model processing system, the relevant base model within the enterprise is reused or the open source model is used as the large model base.

[0014] The open source model is a deepseek32B model or a Qwen32b model.

[0015] In step S2, when presetting log analysis prompt words, analysis prompt words are written according to the log type (such as redis, mysql, mongodb, nginx, palm run, etc.), and the preset log analysis prompt words are continuously optimized and adjusted during the application process to make the analysis more accurate and more usable.

[0016] Step S3: Establish a web-side management interface, establish a connection with the server through websocket, customize the selection of log files, input instructions or customize the selection of prompt words, and transmit the query results to the big model in real time for analysis and summary;

[0017] Step S4: Establish a server-side processing program, establish a socket connection with the web side, receive execution instructions, push execution results, send log information to the big model service, and manage prompt words and server information.

[0018] In step S3, the web management interface uses the websocket technology to establish a connection with the server and transmits information through streaming messages;

[0019] In step S4, the server-side processing program executes the instructions issued by the web side, pushes the instruction results to the web side in the form of streaming messages, packages the instruction execution results or log information and sends them to the big model service, and pushes the big model analysis results to the web side.

[0020] An online application log analysis and problem location system based on a large model, including a system configuration module, a large model processing module, a web-side management module, and a server-side processing module;

[0021] The system configuration module is responsible for collecting the host information, system process information, deployment location and log document information of each system module, so as to quickly locate the log file and avoid the trouble of repeated entry, thereby improving operation efficiency;

[0022] The large model processing module is responsible for deploying the large model base (such as deepseek32B, Qwen32b) and presetting log analysis prompt words;

[0023] The web-side management module is responsible for establishing a connection with the server through websocket, customizing the selection of log files, entering instructions or customizing the selection of prompt words, and transmitting the query results to the big model in real time for analysis and summary;

[0024] The server-side processing module is responsible for establishing a socket connection with the web side, receiving instructions issued by the web side, executing the instructions, pushing the instruction results to the web side in the form of streaming messages, packaging the instruction execution results or log information and sending them to the large model service, and pushing the large model analysis results to the web side, managing the prompt words of different log types, and managing information of different servers and different log locations.

[0025] The large model processing module reuses the relevant base model within the enterprise or uses the open source model as the large model base.

[0026] The open source model is a deepseek32B model or a Qwen32b model.

[0027] When the large model processing module presets log analysis prompt words, it writes analysis prompt words according to the log type (such as redis, mysql, mongodb, nginx, palm run, etc.), and continuously optimizes and adjusts the preset log analysis prompt words during the application process to make the analysis more accurate and more usable.

[0028] The beneficial effects of the present invention are as follows: the online application log analysis and problem location method based on a large model can not only significantly improve the efficiency and accuracy of problem location, but also save a lot of operation and maintenance costs, reduce the difficulty of location, thereby comprehensively improving the system operation and maintenance level, and providing enterprises with more efficient, stable and reliable operation and maintenance management support. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are 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 work.

[0030] Figure 1 Schematic diagram of the online application log analysis and problem location method based on a large model of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present invention.

[0032] This large-scale model-based online application log analysis and problem location method uses WebSocket technology to establish a real-time communication channel between the front-end and back-end. Operations and maintenance personnel can interact with the server in real time through the Web-side management interface, issue server commands in real time through the Web, and view execution results immediately. The server-side processing program can achieve efficient management of command execution and log information.

[0033] By managing server and system application configuration information, log files can be quickly located. By combining the powerful analytical capabilities of large-scale model technology, in-depth analysis, summary, and problem location can be performed on system execution results or system logs.

[0034] The following steps are involved:

[0035] Step S1: Establish a server and system application configuration information management function to collect host information, system process information, deployment location, and log document information deployed by each system module, so as to quickly locate the log file and avoid the trouble of repeated entry, thereby improving operation efficiency;

[0036] The server and system application configuration information management function supports multi-user concurrent operations and provides user authority management function.

[0037] Step S2: Build a large model processing system, deploy a large model base (such as deepseek32B, Qwen32b), and preset log analysis prompt words;

[0038] In step S2, when constructing a large model processing system, the relevant base model within the enterprise is reused or the open source model is used as the large model base.

[0039] The open source model is a deepseek32B model or a Qwen32b model.

[0040] The preset log analysis prompt words support custom adjustment according to actual application conditions.

[0041] In step S2, when presetting log analysis prompt words, analysis prompt words are written according to the log type (such as redis, mysql, mongodb, nginx, palm run, etc.), and the preset log analysis prompt words are continuously optimized and adjusted during the application process to make the analysis more accurate and more usable.

[0042] Step S3: Establish a web-side management interface, establish a connection with the server through websocket, customize the selection of log files, input instructions or customize the selection of prompt words, and transmit the query results to the big model in real time for analysis and summary;

[0043] The web-based management interface supports real-time display of analysis results of large models and provides interactive dialogue functions.

[0044] Step S4: Establish a server-side processing program, establish a socket connection with the web side, receive execution instructions, push execution results, send log information to the big model service, and manage prompt words and server information.

[0045] In step S3, the web management interface uses the websocket technology to establish a connection with the server and transmits information through streaming messages;

[0046] In step S4, the server-side processing program executes the instructions issued by the web side, pushes the instruction results to the web side in the form of streaming messages, packages the instruction execution results or log information and sends them to the big model service, and pushes the big model analysis results to the web side.

[0047] The server-side processing program supports dynamic management of prompt words of different log types.

[0048] The server-side processing program supports real-time updating and maintenance of server and log location information.

[0049] The online application log analysis and problem location system based on large models includes a system configuration module, a large model processing module, a web-side management module and a server-side processing module;

[0050] The system configuration module is responsible for collecting the host information, system process information, deployment location and log document information of each system module, so as to quickly locate the log file and avoid the trouble of repeated entry, thereby improving operation efficiency;

[0051] The large model processing module is responsible for deploying the large model base (such as deepseek32B, Qwen32b) and presetting log analysis prompt words;

[0052] The web-side management module is responsible for establishing a connection with the server through websocket, customizing the selection of log files, entering instructions or customizing the selection of prompt words, and transmitting the query results to the big model in real time for analysis and summary;

[0053] The server-side processing module is responsible for establishing a socket connection with the web side, receiving instructions issued by the web side, executing the instructions, pushing the instruction results to the web side in the form of streaming messages, packaging the instruction execution results or log information and sending them to the large model service, and pushing the large model analysis results to the web side, managing the prompt words of different log types, and managing information of different servers and different log locations.

[0054] The large model processing module reuses the relevant base model within the enterprise or uses the open source model as the large model base.

[0055] The open source model is a deepseek32B model or a Qwen32b model.

[0056] When the large model processing module presets log analysis prompt words, it writes analysis prompt words according to the log type (such as redis, mysql, mongodb, nginx, palm run, etc.), and continuously optimizes and adjusts the preset log analysis prompt words during the application process to make the analysis more accurate and more usable.

[0057] Compared with existing technologies, this large-model-based online application log analysis and problem location method has the following advantages:

[0058] First, it improves operation and maintenance efficiency: through real-time communication and intelligent analysis, it reduces the time for manual investigation and analysis, and improves the efficiency of operation and maintenance operations.

[0059] Second, it reduces operation and maintenance costs: it reduces reliance on manual experience, reduces the difficulty of locating problems caused by lack of experience, and saves operation and maintenance resources.

[0060] Third, enhanced system stability: By quickly and accurately locating problems, business interruption time is reduced and the overall stability of the system is improved.

[0061] Fourth, it improves the level of operation and maintenance management: By leveraging the analytical capabilities of large model technology, it provides more comprehensive and in-depth operation and maintenance data support, helping operation and maintenance personnel make more informed decisions.

[0062] In summary, this large-scale model-based online application log analysis and problem location method is committed to providing an efficient, intelligent, and automated solution for the operation and maintenance management of communication companies to adapt to the rapidly developing business needs and technical environment.

[0063] The embodiment described above is only one specific implementation of the present invention. Common changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A large-scale model-based online application log analysis and problem location method, characterized by: By using WebSocket technology to establish a real-time communication channel between the front-end and back-end, operation and maintenance personnel can interact with the server in real time through the Web management interface, issue server commands in real time through the Web, and view the execution results immediately. The server-side processing program can manage the execution of commands and log information. By managing server and system application configuration information, log files can be quickly located. By combining the analytical capabilities of large-scale model technology, in-depth analysis, summary, and problem location can be performed on system execution results or system logs.

2. The large model-based online application log analysis and problem location method according to claim 1, characterized in that: The following steps are involved: Step S1: Establish a server and system application configuration information management function to collect host information, system process information, deployment location, and log document information deployed by each system module, so as to quickly locate the log file and avoid the trouble of repeated entry, thereby improving operation efficiency; Step S2: Build a large model processing system, deploy a large model base, and preset log analysis prompts; Step S3: Establish a web-side management interface, establish a connection with the server through websocket, customize the selection of log files, input instructions or customize the selection of prompt words, and transmit the query results to the big model in real time for analysis and summary; Step S4: Establish a server-side processing program, establish a socket connection with the web side, receive execution instructions, push execution results, send log information to the big model service, and manage prompt words and server information.

3. The large model-based online application log analysis and problem location method according to claim 2, characterized in that: In step S2, when constructing a large model processing system, the relevant base model within the enterprise is reused or the open source model is used as the large model base.

4. The large model-based online application log analysis and problem location method according to claim 3 is characterized by: The open source model is a deepseek32B model or a Qwen32b model.

5. The large model-based online application log analysis and problem location method according to claim 2, characterized in that: In step S2, when presetting log analysis prompt words, analysis prompt words are written according to the log type, and the preset log analysis prompt words are continuously optimized and adjusted during the application process.

6. The large model-based online application log analysis and problem location method according to claim 2, characterized in that: In step S3, the web management interface uses the websocket technology to establish a connection with the server and transmits information through streaming messages; In step S4, the server-side processing program executes the instructions issued by the web side, pushes the instruction results to the web side in the form of streaming messages, packages the instruction execution results or log information and sends them to the big model service, and pushes the big model analysis results to the web side.

7. An online application log analysis and problem location system based on a large model, characterized by: Including system configuration module, large model processing module, web management module and server processing module; The system configuration module is responsible for collecting the host information, system process information, deployment location and log document information of each system module, so as to quickly locate the log file and avoid the trouble of repeated entry, thereby improving operation efficiency; The large model processing module is responsible for deploying the large model base and presetting log analysis prompt words; The web-side management module is responsible for establishing a connection with the server through websocket, customizing the selection of log files, entering instructions or customizing the selection of prompt words, and transmitting the query results to the big model in real time for analysis and summary; The server-side processing module is responsible for establishing a socket connection with the web side, receiving instructions issued by the web side, executing the instructions, pushing the instruction results to the web side in the form of streaming messages, packaging the instruction execution results or log information and sending them to the large model service, and pushing the large model analysis results to the web side, managing the prompt words of different log types, and managing information of different servers and different log locations.

8. The large model-based online application log analysis and problem location system according to claim 7, characterized in that: The large model processing module reuses the relevant base model within the enterprise or uses the open source model as the large model base.

9. The large model-based online application log analysis and problem location system according to claim 8, characterized in that: The open source model is a deepseek32B model or a Qwen32b model.

10. The large model-based online application log analysis and problem location system according to claim 7, characterized in that: When the large model processing module presets log analysis prompt words, it writes analysis prompt words according to the log type and continuously optimizes and adjusts the preset log analysis prompt words during the application process.

Images