Server fault log positioning auxiliary method
By using the CNN convolutional neural network model to analyze the BMC and OS logs in server fault location, the problems of cumbersome and inefficient fault location in the existing technology are solved, and more efficient fault handling and positioning are achieved.
Patent Information
- Application Number
- CN202411552530.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-01
- Publication Date
- 2025-03-25
AI Technical Summary
The existing technology has cumbersome processing procedures in the server fault location process, which affects the efficiency of maintenance personnel, and the fault phenomenon is unclear, resulting in difficulty in rapid location.
By collecting and preprocessing the server's historical BMC fault logs and OS logs, dividing them into training sets, test sets and verification sets, creating a CNN convolutional neural network model for training and verification, collecting logs in real time, cleaning and normalizing processing, and inputting the model for fault analysis and processing recommended output.
It effectively reduces the time for manual log review, improves the efficiency of fault handling and positioning accuracy, and simplifies the operation process of maintenance personnel.
Smart Images
Figure CN119690771A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of server fault location, and in particular relates to a server fault log location auxiliary method. Background Art
[0002] Server failure refers to the phenomenon that the server cannot provide services normally due to hardware, software or network reasons during normal operation. These failures may cause data loss, system crashes, business interruptions and other problems, which will have a serious impact on the normal operation of the enterprise. In order to ensure the stable operation of the server, it is necessary to regularly maintain and monitor the server and handle failures in a timely manner.
[0003] Currently, when a server fails, each component is individually processed for fault location, including hard disk, memory, raid card, motherboard, network card, power supply, etc. The existing component failure is handled separately, the problem handling process is cumbersome, affecting the efficiency of maintenance personnel, the fault phenomenon reported by customers on site is unclear, the professional capabilities of the front-line processing personnel are uneven, and it is not conducive to the rapid location of existing network problems.
[0004] Therefore, how to solve the failures of various components of existing servers with one click, simplify the operation process of maintenance personnel, and improve the timeliness of locating server failures is a technical problem that urgently needs to be solved. Summary of the invention
[0005] The purpose of the present invention is to provide a server fault log positioning auxiliary method, which is used to achieve one-click solution to the faults of various components of the existing server, simplify the operation process of maintenance personnel, and improve the timeliness of locating server faults.
[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is as follows:
[0007] A server fault log location auxiliary method includes the following steps:
[0008] S1: collect historical BMC fault logs and OS logs of each server, and pre-process the historical BMC fault logs and OS logs of each server;
[0009] S2: Divide the preprocessed historical BMC fault logs and OS logs into training set, test set and validation set;
[0010] S3: creating a fault log detection model, and training, testing and verifying the fault log detection model through a training set, a test set and a validation set in sequence, wherein the fault log detection model is a CNN convolutional neural network model;
[0011] S4: Compare the parameters of the verification result of the fault log detection model with the preset parameter indicators to determine whether the fault log detection model meets the requirements. If so, execute step S5; if not, collect more historical BMC fault logs and OS logs and execute steps S1 to S3 repeatedly until the verification result meets the requirements;
[0012] S5: collects the BMC logs and OS logs of the server in real time, and performs data cleaning and normalization processing on them;
[0013] S6: Input the cleaned and normalized BMC logs and OS logs into the fault log detection model, analyze the server fault through the fault log detection model, and output corresponding fault analysis results and fault handling suggestions.
[0014] Preferably, the BMC log in step S1 includes sel, fdm, systemcom, fru, version, and operate files; the OS log includes demsg, raid_com_log, and driver indicator data.
[0015] Preferably, the preprocessing process of the historical BMC fault log and OS log in step S1 includes cleaning and normalization processing, wherein the cleaning process uses a truth value discovery algorithm, and the normalization process uses a Min-Max normalization algorithm.
[0016] Preferably, step S3 includes the following specific process:
[0017] S31: inputting the training set into the fault log detection model to train the fault log detection model;
[0018] S32: input the test set into the trained fault log detection model, test the trained fault log detection model, and determine whether the model meets the requirements according to the test results. If yes, execute S33; if no, re-execute step S31;
[0019] S33: Input the verification set into the tested fault log detection model, verify the tested fault log detection model, and further verify whether the model meets the requirements.
[0020] Preferably, step S3 includes the following specific process:
[0021] S31: inputting the training set into the fault log detection model to train the fault log detection model;
[0022] S32: input the test set into the trained fault log detection model, test the trained fault log detection model, and determine whether the model meets the requirements according to the test results. If yes, execute S33; if no, re-execute step S31;
[0023] S33: Input the verification set into the tested fault log detection model, verify the tested fault log detection model, and further verify whether the model meets the requirements.
[0024] Preferably, the server is also preset with a fault handling case library, and step S6 includes the following specific process:
[0025] S61: Analyze the server fault through the fault log detection model according to the BMC log and the OS log, and analyze and determine the type of the server fault;
[0026] S62: According to the type of server fault, call the fault handling case library, match the corresponding processing method in the fault handling case library, and output the corresponding fault analysis result and fault handling suggestion according to the matched corresponding processing method.
[0027] Preferably, step S62 includes the following specific processes:
[0028] S621: Match the fault handling case library according to the server fault type, and sort the cases in the fault handling case library according to similarity when searching;
[0029] S622: Determine the priority according to the similarity ranking, extract the preset fault judgment conditions one by one and match them with the uploaded log files to determine whether there is a case hit, if so, execute step S623, if not, execute step S624;
[0030] S623: Outputting corresponding fault analysis results and fault handling suggestions;
[0031] S624: directly output the fault analysis result, and the fault is handled manually.
[0032] The beneficial effects of the present invention include:
[0033] The server fault log location auxiliary method provided by the present invention first collects the historical BMC fault logs and OS logs of each server and performs preprocessing; then divides the preprocessed logs into a training set, a test set and a verification set; creates a fault log detection model, and trains, tests and verifies through the training set, the test set and the verification set in turn; compares the parameters of the verification result of the fault log detection model with the preset parameter index to determine whether the fault log detection model meets the requirements, and if so, collects the BMC logs and OS logs of the server in real time, performs data cleaning and normalization processing on them, and inputs them into the fault log detection model, analyzes the server faults through the fault log detection model, and outputs the corresponding fault analysis results and fault handling suggestions. It effectively reduces the manual log review, improves the fault handling efficiency through case priority matching, and has higher fault location efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 The figure is a flow chart of the server fault log location assistance method of the present invention. Figure 2 The figure is a schematic diagram of the training process of the fault log detection model of the present invention. Figure 3 The figure is a flow chart of server failure handling of the present invention. DETAILED DESCRIPTION
[0035] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application usually described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.
[0036] The present application will present various aspects, embodiments or features around a system that may include multiple devices, components, modules, etc. It should be understood and appreciated that each system may include additional devices, components, modules, etc., and / or may not include all of the devices, components, modules, etc. discussed in conjunction with the figures. In addition, combinations of these schemes may also be used.
[0037] In addition, in the embodiments of the present application, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" in the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present concepts in a specific way.
[0038] The following is combined with Figure 1-Figure 3 The present invention is further described in detail:
[0039] See attached Figure 1 As shown, a server fault log location auxiliary method includes the following steps:
[0040] S1: Collect historical BMC fault logs and OS logs of each server, and preprocess them. The preprocessing process includes cleaning and normalization. The cleaning process uses the truth discovery algorithm, and the normalization process uses the Min-Max normalization algorithm.
[0041] S2: Divide the preprocessed historical BMC fault logs and OS logs into a training set, a test set and a validation set. The training set is used as the training output of the subsequent fault log detection model to implement the training of the fault log detection model. The test set is used as the test output of the fault log detection model to implement the testing of the fault log detection model and evaluate the accuracy, fit and adaptability of the model. The validation set is used as the validation input of the fault log detection model to verify the accuracy of the test results.
[0042] S3: Create a fault log detection model, and train, test, and verify the fault log detection model through a training set, a test set, and a validation set in sequence, so that the fault log detection model can accurately judge server faults;
[0043] S4: Compare the parameters of the verification result of the fault log detection model with the preset parameter indicators to determine whether the fault log detection model meets the requirements. If so, execute step S5; if not, collect more historical BMC fault logs and OS logs and execute steps S1 to S3 repeatedly until the verification result meets the requirements;
[0044] S5: collects the server's BMC logs and OS logs in real time, and performs data cleaning and normalization on them. Data cleaning is used to ensure that all subsequent input data is valid. Normalization is used to scale the data to the same range and eliminate the dimensional differences between features, thereby improving the performance and stability of the model.
[0045] S6: Input the cleaned and normalized BMC logs and OS logs into the fault log detection model, analyze the server fault through the fault log detection model, and output corresponding fault analysis results and fault handling suggestions.
[0046] In the above scheme, the BMC log in step S1 includes sel, fdm, systemcom, fru, version and operate files; the OS log includes demsg, raid_com_log and driver indicator data. The preprocessing process of the historical BMC fault log and OS log in step S1 includes cleaning and normalization, wherein the cleaning process uses the truth value discovery algorithm, and the normalization process uses the Min-Max normalization algorithm.
[0047] See also Figure 2 , step S3 includes the following specific processes:
[0048] S31: inputting the training set into the fault log detection model, training the fault log detection model, so that the model has a certain learning ability, and can subsequently accurately identify server faults based on real-time BMC logs and OS logs;
[0049] S32: input the test set into the trained fault log detection model, test the trained fault log detection model, and determine whether the model meets the requirements according to the test results. If yes, execute S33; if no, re-execute step S31;
[0050] S33: Input the verification set into the tested fault log detection model, verify the tested fault log detection model, and further verify whether the model meets the requirements.
[0051] Before the server works, a fault handling case library is preset in the server, and the fault handling case library includes various server fault data that have occurred in history. The server fault data includes the log, category and corresponding fault handling method of the server fault. Step S6 includes the following specific process:
[0052] S61: Analyze the server fault through the fault log detection model according to the BMC log and the OS log, and determine the type of the server fault;
[0053] S62: According to the type of server fault, call the fault handling case library, match the corresponding processing method in the fault handling case library, and output the corresponding fault analysis result and fault handling suggestion according to the matched corresponding processing method.
[0054] Step S62 includes the following specific processes:
[0055] S621: Match the fault handling case library according to the server fault type, and sort the cases in the fault handling case library according to similarity when searching;
[0056] S622: Determine the priority according to the similarity ranking, extract the preset fault judgment conditions one by one and match them with the uploaded log files to determine whether there is a case hit, if so, execute step S623, if not, execute step S624;
[0057] S623: Outputting corresponding fault analysis results and fault handling suggestions;
[0058] S624: directly output the fault analysis result, and the fault is handled manually.
[0059] The priorities may be identical, 80% similar, 60% similar, 40% similar, 20% similar, and completely different.
[0060] In summary, the server fault log location auxiliary method provided by the present invention first collects the historical BMC fault logs and OS logs of each server and performs preprocessing; then divides the preprocessed logs into a training set, a test set and a verification set; creates a fault log detection model, and trains, tests and verifies through the training set, the test set and the verification set in turn; compares the parameters of the verification result of the fault log detection model with the preset parameter indicators to determine whether the fault log detection model meets the requirements, and if so, collects the BMC logs and OS logs of the server in real time, performs data cleaning and normalization processing on them, and inputs them into the fault log detection model, analyzes the server faults through the fault log detection model, and outputs the corresponding fault analysis results and fault handling suggestions. It effectively reduces the manual review of logs, improves the fault handling efficiency through case priority matching, and has a higher fault location efficiency.
[0061] The above-mentioned embodiments only express the specific implementation method of the present application, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the protection scope of the present application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the technical solution concept of the present application, and these all belong to the protection scope of the present application.
Claims
1. A server fault log location assistance method, characterized in that: The following steps are involved: S1: Collect historical BMC fault logs and OS logs of each server, and pre-process the historical BMC fault logs and OS logs of each server; S2: Divide the preprocessed historical BMC fault logs and OS logs into training set, test set and validation set; S3: Create a fault log detection model, and train, test, and verify the fault log detection model through the training set, test set, and validation set in turn; S4: Compare the parameters of the verification result of the fault log detection model with the preset parameter indicators to determine whether the fault log detection model meets the requirements. If so, execute step S5; if not, collect more historical BMC fault logs and OS logs and execute steps S1 to S3 repeatedly until the verification result meets the requirements; S5: collects the BMC logs and OS logs of the server in real time, and performs data cleaning and normalization processing on them; S6: Input the cleaned and normalized BMC logs and OS logs into the fault log detection model, analyze the server fault through the fault log detection model, and output corresponding fault analysis results and fault handling suggestions.
2. A server fault log location assistance method according to claim 1, characterized in that: The BMC log in step S1 includes sel, fdm, systemcom, fru, version, and operate files; the OS log includes demsg, raid_com_log, and driver indicator data.
3. A server fault log location assistance method according to claim 2, characterized in that: The preprocessing process of the historical BMC fault log and OS log in step S1 includes cleaning and normalization processing, wherein the cleaning process uses a truth value discovery algorithm, and the normalization process uses a Min-Max normalization algorithm.
4. A server fault log location assistance method according to claim 1, characterized in that: Step S3 includes the following specific processes: S31: inputting the training set into the fault log detection model to train the fault log detection model; S32: input the test set into the trained fault log detection model, test the trained fault log detection model, and determine whether the model meets the requirements according to the test results. If yes, execute S33; if no, re-execute step S31; S33: Input the verification set into the tested fault log detection model, verify the tested fault log detection model, and further verify whether the model meets the requirements.
5. A server fault log location assistance method according to claim 1, characterized in that: The server also has a preset fault handling case library. Step S6 includes the following specific processes: S61: Analyze the server fault through the fault log detection model according to the BMC log and the OS log, and analyze and determine the type of the server fault; S62: According to the type of server fault, call the fault handling case library, match the corresponding processing method in the fault handling case library, and output the corresponding fault analysis result and fault handling suggestion according to the matched corresponding processing method.
6. A server fault log location assistance method according to claim 5, characterized in that: Step S62 includes the following specific processes: S621: Match the fault handling case library according to the server fault type, and sort the cases in the fault handling case library according to similarity when searching; S622: Determine the priority according to the similarity ranking, extract the preset fault judgment conditions one by one and match them with the uploaded log files to determine whether there is a case hit, if so, execute step S623, if not, execute step S624; S623: Outputting corresponding fault analysis results and fault handling suggestions; S624: directly output the fault analysis result, and the fault is handled manually.
7. A server fault log location assistance method according to claim 1, characterized in that: The fault log detection model is a CNN convolutional neural network model.