Detailed Description
In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first" and "second" and the like in the description and the claims of the present application and the above drawings are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
Fig. 1 is a flowchart of an intelligent operation and maintenance method according to a first embodiment of the present application, where the present embodiment is applicable to a case of performing intelligent operation and maintenance on a server cluster, and the method may be performed by an intelligent operation and maintenance device, which may be implemented by using software and/or hardware and specifically configured in operation and maintenance equipment of the server cluster, for example, a computer in an operation and maintenance platform.
Referring to the intelligent operation and maintenance method shown in fig. 1, the method specifically includes the following steps:
s110, acquiring corresponding original logs from various types of servers.
The server has different attention points for operation and maintenance due to different positions and functions, so that the original log of the corresponding type needs to be acquired according to the type of the server. Specifically, a data collection rule may be preset, and corresponding original logs are obtained from various types of servers according to the preset data collection rule. The data collection rules are used to define the types of raw logs that need to be obtained in each type of server.
By way of example, the types of raw logs may include at least one of a base log, a function log, a run component log, and the like. The base log may be a log that includes system resource load snapshot information generated by the server runtime. For example, the base log may include a running log of components such as a central processing unit (Central Processing Unit, CPU), memory, disk, and network. The function log may be a log generated during the operation of the function server-related function. For example, the function log may include an access log, a transaction detail log, an error log, and the like generated by the middleware server. The running component log may be component running log information that plays a key role in the normal operation of the system. For example, the running component log may be a security audit log.
The collection rules may be determined by a person of ordinary skill in the art based on experience or analysis of historical data, as the application is not particularly limited in this regard. For example, the collection rules may include a middleware server collecting a base log and a function log.
And S120, carrying out structural analysis on the original log to obtain a log to be analyzed.
The structured analysis is used for analyzing the corresponding data type and value of each field in the original log, and extracting key features to generate the log to be analyzed. The log to be analyzed can be a log after the original log is subjected to structural analysis standardization and is used for inputting a subsequent abnormal prediction model to predict abnormal risks.
The arrangement modes of fields in the original logs of different types may be different, the data types may be different, and not all contents in the original logs are useful for subsequent abnormal risk prediction, so that the original logs need to be subjected to structural analysis, data useful for subsequent abnormal risk prediction are extracted, the format is unified, and the data quality of the logs to be analyzed is improved.
In an alternative embodiment, the original log is subjected to structural analysis to obtain a log to be analyzed, wherein the log to be analyzed is obtained by deleting irrelevant information and supplementing context to the original log, key information in the log to be analyzed is extracted according to a preset structural quantization index, and the structural quantization index is determined according to the key information.
The irrelevant information deletion may be to remove the data irrelevant to the operation and maintenance analysis requirement in the original log according to the data cleaning rule defined by the user. For example, the irrelevant information deletion may be to reject output information at the time of startup, output service information, and the like. The context replenishment may be to complement missing information in the original log that is relevant to the operation and maintenance analysis requirements. Part of the original log lacks key information for subsequent structured quantization indexes, and therefore context supplementation is required. By way of example, the context-supplemented data types may include time of event occurrence, event level, event source, and the like. The context supplementation may be determined according to the type of data to be supplemented, which is not particularly limited by the present application. For example, in the absence of event occurrence time, the event occurrence time of the absence can be determined by an interpolation algorithm to perform context replenishment.
The log to be analyzed can be a log obtained by deleting irrelevant information and supplementing context to the original log, and the log to be analyzed is used for obtaining the log to be analyzed after structural analysis. The preset structured quantization index can be a preset quantization index which needs to be subjected to structured analysis, and is used for extracting key information in the log to be analyzed and carrying out quantization processing. And acquiring key information corresponding to a preset structured quantization index from the log to be analyzed, and carrying out quantization processing on the acquired key information according to quantization rules in the preset structured quantization index. The quantization process is used to convert the key information into a multi-dimensional quantization index that can be recognized and processed by a computer. Exemplary quantization processing may include normalization, feature engineering, etc., may normalize key information formats, improve quality of structured quantization indexes, and improve efficiency and accuracy of subsequent processing. For example, the data format in the log to be analyzed may be as follows:
[ event time ] [ hostname ] [ component ] [ primary index ] [ secondary index ] [ N-level index ]: index value 1] [ index value 2] [ index value N ].
The method comprises the steps of deleting irrelevant information and supplementing context to an original log to obtain a log to be analyzed, reducing information redundancy of the log to be analyzed by deleting the irrelevant information, reducing interference of the irrelevant information to subsequent prediction, improving prediction accuracy, guaranteeing integrity of the log to be analyzed by supplementing context, providing multi-dimensional data support for subsequent prediction, extracting key information in the log to be analyzed according to preset structured quantization indexes, determining the structured quantization indexes according to the key information, and obtaining the log to be analyzed, so that the log to be analyzed is normalized, the subsequent prediction is facilitated, and the prediction efficiency and accuracy are improved.
S130, predicting whether the log to be analyzed has abnormal risk or not through a trained abnormal prediction model, and obtaining a prediction result.
The trained abnormal prediction model can be an abnormal prediction model with abnormal risk prediction capability obtained after monitoring and training through sample data with labels, and is used for predicting whether the abnormal risk exists in the log to be analyzed. The anomaly prediction model may be a deep learning model. By way of example, the anomaly prediction model may be a deep learning network such as a long-term memory network, a convolutional neural network, and a decision tree, which is not particularly limited in the present application. The trained anomaly prediction model may be packaged, and an inference interface of the trained anomaly prediction model may be derived using the HTTP (Hypertext Transfer Protocol )/TCP (TransmissionControl Protocol, transmission control protocol) protocol, through which the trained anomaly prediction model may be subsequently used.
And inputting the log to be analyzed into a trained abnormal prediction model, and outputting the log to be analyzed as a prediction result. The prediction results may include risk types and anomaly content. Risk types may include the presence of an abnormal risk and the absence of an abnormal risk. The abnormal content may be specific risk content corresponding to the risk type when the risk type is that there is an abnormal risk.
And S140, when the predicted result is that the abnormal risk exists, predicting whether manual maintenance is needed or not through a maintenance judging model according to the abnormal content in the predicted result.
The maintenance determination model may be a model for determining whether manual intervention is required based on the abnormal content. For example, the maintenance determination model may be an expert model, and determine, through a preset rule, whether the abnormal content determination requires manual intervention. The maintenance determination model may also be a deep learning model, for example, a large model, and automatically determines whether human intervention is required according to abnormal content after learning through a certain rule.
And when the predicted result is that the abnormal risk exists, inputting the abnormal content into a maintenance judging model, and determining whether the maintenance is needed or not.
And S150, if not, generating a target operation and maintenance strategy according to the preset operation and maintenance rule and the abnormal content, and controlling to execute the target operation and maintenance strategy.
If not, that is, no maintenance is needed, at this time, an automatic maintenance program is entered. The preset operation and maintenance rule may be a preset maintenance rule for an abnormal event, and is used for generating a target operation and maintenance policy. The operation and maintenance rules can be set by a professional technician according to related operation and maintenance guide books, related information of stipulation and historical operation and maintenance events, and the like, and the application is not particularly limited to the above. The target operation and maintenance policy may be a specific method of operation and maintenance, and may include operation and maintenance objects and operation contents. After the target operation and maintenance strategy is generated, the execution device is controlled to execute the target operation and maintenance strategy so as to realize automatic operation and maintenance.
If the operation and maintenance is needed, the abnormal content in the prediction result can be sent to the corresponding operation and maintenance personnel to prompt the operation and maintenance personnel to perform manual operation and maintenance.
In the operation and maintenance scenario of the server, an operation and maintenance manager typically uses various information generated based on the operation of the system to perform operation and maintenance operations such as monitoring, fault checking, resource management, and security maintenance on the system. These operations are critical to ensuring the stability and security of the information system, but at the same time suffer from a number of drawbacks. The manual operation and maintenance operation often requires a large amount of repeated labor, particularly in the aspects of monitoring and fault detection, has lower efficiency, needs more time for manual intervention when a system is in a problem, possibly has longer response and solving time of the problem, can have misoperation due to human factors such as fatigue, negligence and the like, increases the risk of the system, greatly increases the difficulty and complexity of the manual operation and maintenance along with the expansion of the scale of the system, is difficult to effectively expand, often lacks the prediction capability of the system on potential problems, is more post-treatment and has poor effect.
In summary, with the rapid development of information technology, the complexity of system operation and maintenance is increasing, and the traditional operation and maintenance method relies on manual monitoring and analysis of system logs, which is inefficient, and is difficult to adapt to a large-scale and dynamic system environment due to easy omission of important information.
According to the technical scheme, the corresponding original logs are obtained from various types of servers, the original logs are subjected to structural analysis to obtain the logs to be analyzed, the logs to be analyzed can be normalized through structural analysis of the original logs, the data quality of the logs to be analyzed is improved, misjudgment caused by data non-standardization is avoided, the accuracy of subsequent prediction is improved, whether the logs to be analyzed have abnormal risks or not is predicted through a trained abnormal prediction model, prediction of potential problems of the servers can be achieved to improve the predictability of subsequent operation and maintenance, when the predicted results are abnormal risks, whether manual maintenance is needed is predicted through a maintenance judgment model according to abnormal content in the predicted results, if not, a target operation and maintenance strategy is generated according to preset operation and maintenance rules, execution of the target operation and maintenance strategy is controlled, operation and maintenance are automatically performed, and the accuracy of operation and maintenance efficiency are improved. Therefore, the technical scheme of the application solves the problems that the manual operation and maintenance often lacks the prediction capability for the potential problems of the system, the processing efficiency is low and the accuracy cannot be ensured, and achieves the effects of improving the predictability of the operation and maintenance and the processing efficiency and accuracy for the abnormal events.
Example two
Fig. 2 is a flowchart of an intelligent operation and maintenance method according to a second embodiment of the present application, where the technical solution of the present embodiment is further refined based on the technical solution.
Further, the method comprises the steps of predicting whether the log to be analyzed has abnormal risks through a trained abnormal prediction model, and the method comprises the steps of refining the log to be analyzed into a target abnormal prediction model determined from the trained candidate abnormal prediction model according to the server type corresponding to the log to be analyzed, and predicting whether the log to be analyzed has abnormal risks through the target abnormal prediction model so as to predict whether the log to be analyzed has abnormal risks.
Referring to fig. 2, an intelligent operation and maintenance method includes:
s210, acquiring corresponding original logs from various types of servers.
S220, carrying out structural analysis on the original log to obtain a log to be analyzed.
S230, determining a target abnormal prediction model from the trained candidate abnormal prediction models according to the server type corresponding to the log to be analyzed.
In the actual operation and maintenance process, the operation and maintenance requirements of different types of servers are different, so that the log information to be concerned is different, the influence factors and operation and maintenance targets to be considered when faults are encountered are also different, and the corresponding operation and maintenance actions to be adopted are different. For example, for a server of a resource class, monitoring of the resource is required, an operation and maintenance person mainly pays attention to the fluctuation condition of the resource, and when the use of the resource is too high or too low, corresponding operation and maintenance actions of capacity expansion or capacity shrinkage are adopted. For another example, for the monitoring of the transaction status such as that required by the middleware server, the operation and maintenance personnel may pay more attention to the indexes such as success rate and time consumption of the transaction. Thus, for different server types, it is necessary to train different candidate anomaly prediction models to meet different operational scenario requirements.
The candidate anomaly prediction model may be an anomaly prediction model corresponding to different server types, and is used for performing anomaly risk prediction according to logs to be analyzed of the corresponding server types. And according to the type of the server corresponding to the log to be analyzed, determining a candidate abnormal prediction model of the server of the type from the trained candidate abnormal prediction models as a target abnormal prediction model.
In an alternative embodiment, the trained candidate abnormal prediction models are obtained through the following training mode, wherein the history logs are respectively obtained based on different types of servers, log content of the history logs is subjected to structural analysis, corresponding abnormal content is marked to obtain a sample data set, and the corresponding abnormal prediction models are trained for the various types of servers based on the sample data set to obtain the trained candidate abnormal prediction models.
The history log can be related log in a period of history time when the operation and maintenance process is carried out manually. And carrying out structural analysis on the history log, and marking the corresponding abnormal content to obtain a sample data set. Through structural analysis, irrelevant information in a history log is removed and key information is complemented, structural quantization indexes are determined, and the quality of data in a sample data set is improved. And marking the abnormal content in the history log according to the historical operation and maintenance operation to obtain a sample data set with labels. The sample data set may be divided into a training set and a validation set, and the division ratio may be determined by a skilled artisan according to experiments or experience, which is not particularly limited in the present application. For example, the ratio of training set to validation set may be 8:2.
Because the attention content of different types of servers is different, the actions and targets of the operation and maintenance are different, and therefore, corresponding anomaly prediction models are respectively trained for each type of server.
The history log is usually time series data, and is characterized by time-sequential influence among the data, and for prediction and classification of the data, a variant Long-Term Memory (LSTM) algorithm model of a cyclic neural network (Recurrent Neural Network, RNN) model can be used as an initial candidate anomaly prediction model.
In the structure of the LSTM model, training set data of various servers can be input into a corresponding initial LSTM model to train, the input sequence is as [ index 1 (t), index 2 (t)..index N (t) ], [ index 1 (t+1), index 2 (t+1)..index N (t+1) ]..[ index 1 (t+n), index 2 (t+n)..index N (t+n) ], and model parameters obtained through training according to the sequence are abnormal prediction models in the current operation and maintenance environment. And verifying the trained abnormal prediction models based on the corresponding verification set, and obtaining the trained candidate abnormal prediction models after verification.
The method comprises the steps of respectively obtaining history logs based on different types of servers, carrying out structural analysis on log content of the history logs, marking corresponding abnormal content to obtain a sample data set, training corresponding abnormal prediction models for the different types of servers based on the sample data set to obtain trained candidate abnormal prediction models, training the corresponding abnormal prediction models for the different types of servers, and effectively improving accuracy of risk prediction for the different types of servers.
S240, predicting whether the log to be analyzed has abnormal risk or not through a target abnormal prediction model.
And inputting the log to be analyzed into a target abnormality prediction model to obtain an abnormality risk prediction result.
S250, when the prediction result is that the abnormal risk exists, predicting whether manual maintenance is needed or not through a maintenance judging model according to the abnormal content in the prediction result.
And S260, if not, generating a target operation and maintenance strategy according to the preset operation and maintenance rule and the abnormal content, and controlling to execute the target operation and maintenance strategy.
According to the technical scheme, the target abnormality prediction model is determined from the trained candidate abnormality prediction models according to the server types corresponding to the logs to be analyzed, whether the logs to be analyzed have abnormal risks or not is predicted through the target abnormality prediction model, the target abnormality prediction model can predict corresponding server types through training the corresponding candidate abnormality prediction models for different server types, and accuracy of predicting whether the logs to be analyzed have abnormal risks or not is improved.
Example III
Fig. 3 is a flowchart of an intelligent operation and maintenance method according to a third embodiment of the present application, where the technical solution of the present embodiment is further refined on the basis of the technical solution described above.
Further, the method comprises the steps of generating a target operation and maintenance strategy according to preset operation and maintenance rules and abnormal contents, controlling and executing the target operation and maintenance strategy, and refining the steps of acquiring relevant contexts of the abnormal contents, sending the relevant contexts of the abnormal contents and the abnormal contents to an operation and maintenance strategy generation model, determining the target operation and maintenance rules from the preset operation and maintenance rules according to the abnormal contents, generating the target operation and maintenance strategy according to the target operation and maintenance rules and the relevant contexts, and controlling an execution module corresponding to the target operation and maintenance strategy to execute the target operation and maintenance strategy so as to automatically realize the target operation and maintenance strategy.
Referring to fig. 3, an intelligent operation and maintenance method includes:
s310, acquiring corresponding original logs from various types of servers.
S320, carrying out structural analysis on the original log to obtain a log to be analyzed.
S330, predicting whether the log to be analyzed has abnormal risk or not through a trained abnormal prediction model, and obtaining a prediction result.
And S340, when the predicted result is that the abnormal risk exists, predicting whether the maintenance is needed manually or not through a maintenance judging model according to the abnormal content in the predicted result.
S350, if not, acquiring the related context of the abnormal content, and transmitting the abnormal content and the related context to the operation and maintenance strategy generation model.
The relevant context may be relevant information for operating and maintaining the server corresponding to the abnormal content, and is used for generating a target operation and maintenance policy. By way of example, the relevant context may include information such as server address, configuration, current system load, and running process, as the application is not limited in detail. The related context can be obtained through the original log corresponding to the abnormal content, and the abnormal content can be obtained through the output of the abnormal prediction model.
The operation and maintenance strategy generation model is used for determining an operation and maintenance strategy based on operation and maintenance actions corresponding to abnormal contents in a rule base maintained in advance. Illustratively, the operation and maintenance policy generation model may be a deep learning network. For example, the operation and maintenance strategy generation model can be a large model or a decision tree, etc. And sending the abnormal content and the related context of the abnormal content to the operation and maintenance strategy generation model so that the operation and maintenance strategy generation model generates a target operation and maintenance strategy.
S360, an operation and maintenance strategy generation model determines a target operation and maintenance rule from preset operation and maintenance rules according to abnormal content, and generates the target operation and maintenance strategy according to the target operation and maintenance rule and related context.
The preset operation and maintenance rule can be a relevant rule for operation and maintenance and can be obtained through a relevant rule file. The target operation rule may be a rule for handling abnormal contents among preset operation rules. The preset operation and maintenance rules can be maintained through a special rule base. And the operation and maintenance strategy generation model determines a corresponding operation and maintenance rule for processing the abnormal content according to a preset operation and maintenance rule, and takes the operation and maintenance rule as a target operation and maintenance rule. The target operation and maintenance policy may include specific operation and maintenance steps and operation and maintenance objects for processing abnormal contents. Specifically, according to the target operation and maintenance rule and the related context, a specific strategy for processing the abnormal content, namely, the target operation and maintenance strategy can be determined.
S370, controlling an execution module corresponding to the target operation and maintenance strategy to execute the target operation and maintenance strategy.
The execution module may be a module for executing a target operation and maintenance policy, and is configured to perform operation and maintenance processing on the abnormal content. The execution modules can be multiple, after the target operation and maintenance strategy is determined, the target operation and maintenance strategy can be added into the execution queue, and the idle execution modules are controlled to sequentially acquire and execute the target operation and maintenance strategy in the execution queue, so that the automatic operation and maintenance of abnormal content is realized.
In an optional embodiment, after the execution module corresponding to the control target operation and maintenance policy executes the target operation and maintenance policy, the method further comprises the steps of obtaining an execution result of the execution module after executing the target operation and maintenance policy, and feeding back the execution result to the operation and maintenance policy generation model.
The execution result may be a change of the relevant target index after executing the target operation and maintenance policy on the operation and maintenance object. For example, if the target operation policy is capacity expansion, the execution result may be whether capacity is expanded. Optionally, the target operation and maintenance policy may include an operation and maintenance target, where the operation and maintenance target may include specific target index data. After executing the target operation and maintenance strategy, the execution result can be obtained according to the target index data.
And feeding back the execution result to the operation and maintenance strategy generation model, wherein the operation and maintenance strategy generation model can perform incremental training according to the execution result, and the accuracy of the operation and maintenance strategy generation model is improved. An execution report may be generated based on the execution result and the target index data. And feeding back the execution report to the operation and maintenance strategy generation model, so that the operation and maintenance strategy generation model can perform incremental training on indexes corresponding to the target index data according to the target operation and maintenance strategy and operation and maintenance results.
The execution result after the execution module executes the target operation and maintenance strategy is acquired, and the execution result is fed back to the operation and maintenance strategy generation model, so that the operation and maintenance strategy generation model can be used for performing incremental training, the operation and maintenance strategy generation model can be continuously optimized based on a feedback learning mode, the accuracy of the target operation and maintenance strategy is improved, and the effectiveness and the intelligence of operation and maintenance are improved.
According to the technical scheme, the method comprises the steps of obtaining relevant contexts of abnormal content, sending the abnormal content and the relevant contexts of the abnormal content to an operation and maintenance strategy generation model, determining a target operation and maintenance rule from preset operation and maintenance rules according to the abnormal content, generating a target operation and maintenance strategy according to the target operation and maintenance rule and the relevant contexts, accurately determining the target operation and maintenance strategy through the operation and maintenance strategy generation model, controlling an execution module corresponding to the target operation and maintenance strategy to execute the target operation and maintenance strategy, automatically determining and executing the target operation and maintenance strategy, realizing automatic operation and maintenance, reducing the workload of manual operation and maintenance, reducing the labor cost of operation and maintenance, and improving the operation and maintenance efficiency and accuracy.
Example IV
Fig. 4 is a schematic structural diagram of an intelligent operation and maintenance device according to a fourth embodiment of the present application, where the embodiment is applicable to a case of performing intelligent operation and maintenance on a server cluster, and is configured in operation and maintenance equipment of the server cluster, and the specific structure of the intelligent operation and maintenance device is as follows:
An original log obtaining module 410, configured to obtain corresponding original logs from servers of various types;
The log to be analyzed obtaining module 420 is configured to perform structural analysis on the original log to obtain a log to be analyzed;
the prediction result obtaining module 430 is configured to predict whether the log to be analyzed has an abnormal risk through a trained abnormal prediction model, so as to obtain a prediction result;
The manual maintenance judging module 440 is configured to predict whether maintenance is needed manually according to the abnormal content in the prediction result when the prediction result is that the abnormal risk exists;
And the target operation and maintenance policy executing module 450 is configured to generate a target operation and maintenance policy according to the preset operation and maintenance rule and the abnormal content if not, and control to execute the target operation and maintenance policy.
According to the technical scheme, the corresponding original logs are obtained from various types of servers, the original logs are subjected to structural analysis to obtain the logs to be analyzed, the logs to be analyzed can be normalized through structural analysis of the original logs, the data quality of the logs to be analyzed is improved, misjudgment caused by data non-standardization is avoided, the accuracy of subsequent prediction is improved, whether the logs to be analyzed have abnormal risks or not is predicted through a trained abnormal prediction model, prediction of potential problems of the servers can be achieved to improve the predictability of subsequent operation and maintenance, when the predicted results are abnormal risks, whether manual maintenance is needed is predicted through a maintenance judgment model according to abnormal content in the predicted results, if not, a target operation and maintenance strategy is generated according to preset operation and maintenance rules, execution of the target operation and maintenance strategy is controlled, operation and maintenance are automatically performed, and the accuracy of operation and maintenance efficiency are improved. Therefore, the technical scheme of the application solves the problems that the manual operation and maintenance often lacks the prediction capability for the potential problems of the system, the processing efficiency is low and the accuracy cannot be ensured, and achieves the effects of improving the predictability of the operation and maintenance and the processing efficiency and accuracy for the abnormal events.
Optionally, the prediction result obtaining module 430 includes:
The target abnormal prediction model determining unit is used for determining a target abnormal prediction model from the trained candidate abnormal prediction models according to the server type corresponding to the log to be analyzed;
and the abnormal risk prediction unit is used for predicting whether the log to be analyzed has abnormal risk or not through the target abnormal prediction model.
Optionally, the intelligent operation and maintenance device further includes:
The history log acquisition module is used for respectively acquiring history logs based on different types of servers;
the sample data set acquisition module is used for carrying out structural analysis on log content of the history log and marking corresponding abnormal content to obtain a sample data set;
And the candidate abnormal prediction model training module is used for training the corresponding abnormal prediction models for the servers of various types based on the sample data set to obtain each trained candidate abnormal prediction model.
Optionally, the target operation and maintenance policy executing module 450 includes:
The related context acquisition unit is used for acquiring the related context of the abnormal content and sending the abnormal content and the related context of the abnormal content to the operation and maintenance strategy generation model;
The target operation and maintenance strategy generation unit is used for generating an operation and maintenance strategy generation model, determining a target operation and maintenance rule from preset operation and maintenance rules according to abnormal contents, and generating a target operation and maintenance strategy according to the target operation and maintenance rule and related contexts;
and the target operation and maintenance strategy executing unit is used for controlling the executing module corresponding to the target operation and maintenance strategy to execute the target operation and maintenance strategy.
Optionally, the target operation and maintenance policy executing module 450 further includes:
The execution result acquisition unit is used for acquiring an execution result after the execution module executes the target operation and maintenance strategy;
and the execution result feedback unit is used for feeding back the execution result to the operation and maintenance strategy generation model.
Optionally, the log to be analyzed obtaining module 420 includes:
the log to be analyzed determining unit is used for deleting irrelevant information and supplementing context to the original log to obtain the log to be analyzed;
The structured quantization index determining unit is used for extracting key information in the log to be analyzed according to a preset structured quantization index, and determining the structured quantization index according to the key information to obtain the log to be analyzed.
The intelligent operation and maintenance device provided by the embodiment of the application can execute the intelligent operation and maintenance method provided by any embodiment of the application, and has the corresponding functional modules and beneficial effects of executing the intelligent operation and maintenance method.
According to embodiments of the present invention, the present invention also provides an electronic device, a readable storage medium and a computer program product.
Example five
Fig. 5 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present application, where, as shown in fig. 5, the electronic device includes a processor 510, a memory 520, an input device 530 and an output device 540, where the number of the processors 510 in the electronic device may be one or more, and in fig. 5, one processor 510 is taken as an example, and the processor 510, the memory 520, the input device 530 and the output device 540 in the electronic device may be connected by a bus or other manners, and in fig. 5, the connection is taken as an example by a bus.
The memory 520 is a computer readable storage medium, and may be used to store software programs, computer executable programs, and modules, such as program instructions/modules (e.g., an original log obtaining module 410, a log to be analyzed obtaining module 420, a prediction result obtaining module 430, a manual maintenance determining module 440, and a target operation and maintenance policy executing module 450) corresponding to the intelligent operation and maintenance method in the embodiment of the present application. The processor 510 executes various functional applications of the electronic device and data processing, i.e., implements the intelligent operation and maintenance method described above, by running software programs, instructions, and modules stored in the memory 520.
The memory 520 may mainly include a storage program area which may store an operating system, application programs required for at least one function, and a storage data area which may store data created according to the use of the terminal, etc. In addition, memory 520 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some examples, memory 520 may further include memory located remotely from processor 510, which may be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input means 530 may be used to receive input character information and to generate key signal inputs related to user settings and function control of the electronic device. The output 540 may include a display device such as a display screen.
Example six
The sixth embodiment of the application also provides a storage medium containing computer executable instructions, which when executed by a computer processor, are used for executing an intelligent operation and maintenance method, the method comprises the steps of obtaining corresponding original logs from various types of servers; the method comprises the steps of carrying out structural analysis on an original log to obtain a log to be analyzed, predicting whether the log to be analyzed has abnormal risks through a trained abnormal prediction model to obtain a prediction result, predicting whether manual maintenance is needed through a maintenance judgment model according to abnormal contents in the prediction result when the prediction result is that the abnormal risks exist, and if not, generating a target operation and maintenance strategy according to preset operation and maintenance rules and abnormal contents and controlling execution of the target operation and maintenance strategy.
Of course, the storage medium containing the computer executable instructions provided in the embodiments of the present application is not limited to the method operations described above, and may also perform the related operations in the intelligent operation and maintenance method provided in any embodiment of the present application.
From the above description of embodiments, it will be clear to a person skilled in the art that the present application may be implemented by means of software and necessary general purpose hardware, but of course also by means of hardware, although in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a FLASH Memory (FLASH), a hard disk, or an optical disk of a computer, etc., and include several instructions for causing an electronic device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments of the present application.
It should be noted that in the embodiment of the intelligent operation and maintenance device, each unit and module included are only divided according to the functional logic, but not limited to the above division, so long as the corresponding functions can be implemented, and in addition, specific names of each functional unit are only for convenience of distinguishing each other, and are not used for limiting the protection scope of the present application.
Note that the above is only a preferred embodiment of the present application and the technical principle applied. It will be understood by those skilled in the art that the present application is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the application. Therefore, while the application has been described in connection with the above embodiments, the application is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the application, which is set forth in the following claims.