CN117311777A - Automatic operation and maintenance platform and method - Google Patents

Abstract

The invention discloses an automatic operation and maintenance platform and method, and belongs to the technical field of operation and maintenance. The invention sets an operation monitoring module, an operation maintenance module and a script issuing module in an automatic operation maintenance platform, wherein the operation monitoring module collects the operation data of the system and analyzes the operation data to obtain the current operation state of the system, and determines whether the system needs to be automatically operated and maintained according to the current operation state; when the system needs to be automatically operated, the operation and maintenance module matches a corresponding fault processing template from a preset operation and maintenance knowledge base according to the current operation state and sends the fault processing template to the script issuing module; the script issuing module configures the corresponding operation and maintenance script according to the fault processing template and issues and loads the operation and maintenance script, so that the operation and maintenance script can be automatically updated and iterated according to actual needs, the operation and maintenance automation is realized, the operation and maintenance efficiency is improved, meanwhile, manual updating and iterating of a script library are not needed, and the operation cost is greatly saved.

Description

Automated operation and maintenance platform and method

Technical field

The present invention relates to the field of operation and maintenance technology, and in particular to an automated operation and maintenance platform and method.

Background technique

With the rapid development of big data and cloud computing, the scale and complexity of related IT infrastructure are increasing rapidly, and it has become a key guarantee to support the smooth development of businesses in various industries. Standard operation and maintenance of these IT infrastructures ensures their stability. , efficient and safe operation, ensuring the reliability of IT services to support business operations, and ensuring the agility of IT services to support digital transformation have become an important topic.

The existing automated operation and maintenance monitoring platform has a long construction period, high construction costs, and its own operation and maintenance work is also very complicated.

Contents of the invention

The main purpose of the present invention is to provide an automated operation and maintenance platform and method, aiming to solve the technical problems of high construction cost and low efficiency of the existing technical operation and maintenance platform.

In order to achieve the above purpose, the present invention provides an automated operation and maintenance platform, which includes: an operation monitoring module, an operation and maintenance module and a script publishing module;

The operation monitoring module is used to collect the operation data of the system, analyze the operation data, obtain the current operation status of the system, and determine whether the system requires automated operation and maintenance based on the current operation status;

The operation and maintenance module is used to match the corresponding fault processing template from the preset operation and maintenance knowledge base according to the current operating status when the system needs to perform automated operation and maintenance, and send the fault processing template to the Described script publishing module;

The script publishing module is used to configure the corresponding operation and maintenance script according to the fault processing template, and publish and load the operation and maintenance script.

Optionally, the script publishing module includes: an automatic script writing and publishing sub-module;

The automatic script writing and publishing sub-module is used to query the preset operation and maintenance script library for operation and maintenance scripts matching the fault type according to the fault processing template, and when the operation and maintenance script matching the fault type is queried, , publish and load the operation and maintenance script;

The automatic script writing and publishing sub-module is also used to automatically configure and generate an operation and maintenance script according to the fault handling template when no operation and maintenance script matching the fault type is queried, and to review the operation and maintenance script. and release loading.

Optionally, the automatic script writing and publishing sub-module is also used to determine a fault handling method according to the fault handling template, and determine a corresponding fault handling code structure according to the fault handling method;

The automatic script writing and publishing sub-module is also used to determine corresponding code parameters according to the fault handling code structure, fill in the code parameters into the fault handling code structure, and generate operations according to preset script generation rules. dimensional script.

Optionally, the script publishing module also includes: a script maintenance sub-module and a script update management sub-module;

The script maintenance sub-module is used to monitor the current status information of the system after the current operation and maintenance script is executed at a preset period, and detect whether the current status information is consistent with the preset system status;

The script maintenance sub-module is also used to generate update information to the script update management sub-module when the current status information is inconsistent with the preset system status;

The script update management submodule is configured to obtain a status difference between the current status information and the preset system status based on the update information, and update the current operation and maintenance script based on the status difference.

Optionally, the script maintenance sub-module is also used to obtain the operation and maintenance script running status monitoring model, and obtain fault information according to the current running status;

The script maintenance sub-module is also used to input the fault information and the current operation and maintenance script into the operation and maintenance script running status monitoring model to obtain the preset system state after executing the current operation and maintenance script.

Optionally, the script maintenance sub-module is also used to obtain historical operation and maintenance data, and preprocess the historical operation and maintenance data to obtain training data corresponding to the historical operation and maintenance data;

The script maintenance sub-module is also used to perform training based on the training data to obtain an operation and maintenance script running status monitoring model.

Optionally, the operation and maintenance module is also used to obtain system historical operation and maintenance data, and preprocess the system historical operation and maintenance data to obtain a text collection of operation and maintenance data;

The operation and maintenance module is also used to perform data association rule mining on the operation and maintenance data text collection, and convert historical operation and maintenance data in the operation and maintenance data text collection into operation and maintenance knowledge data;

The operation and maintenance module is also used to establish a preset operation and maintenance knowledge base according to the operation and maintenance knowledge data.

Optionally, the operation and maintenance module is also used to establish a search engine in the preset operation and maintenance knowledge base, and store the operation and maintenance knowledge data in the preset operation and maintenance knowledge base;

The operation and maintenance module is also used to obtain alarm information according to the operating status, and determine the fault type corresponding to the alarm information through semantic analysis;

The operation and maintenance module is also configured to match a fault processing template corresponding to the fault type in the preset operation and maintenance knowledge base according to the fault type, and send the fault processing template to the script publishing module .

Optionally, the operation monitoring module is also used to collect historical fault data and historical operating status data of the system;

The operation monitoring module is also used to perform training based on the historical fault data and the historical operating status data and establish an alarm prediction model;

The operation monitoring module is also used to input the operation data into the alarm prediction model to obtain the current operation status prediction value of the system;

The operation monitoring module is also used to compare the current operating status prediction value with a preset alarm threshold, and determine that the system needs to perform automated operation and maintenance when the current operating status prediction value is greater than the preset alarm threshold. ;

The operation monitoring module is also used to determine that the system does not require automated operation and maintenance when the current operating status prediction value is less than or equal to the preset alarm threshold.

In addition, in order to achieve the above purpose, the present invention also proposes an automated operation and maintenance method, which is applied to the automated operation and maintenance platform described above. The automated operation and maintenance method includes the following steps:

The operation monitoring module collects the operation data of the system, analyzes the operation data, obtains the current operation status of the system, and determines whether the system requires automated operation and maintenance based on the current operation status;

When the system needs to perform automated operation and maintenance, the operation and maintenance module matches the corresponding fault processing template from the preset operation and maintenance knowledge base according to the current operating status, and sends the fault processing template to the script publishing module;

The script publishing module configures the corresponding operation and maintenance script according to the fault handling template, and publishes and loads the operation and maintenance script.

The present invention sets an operation monitoring module, an operation and maintenance module and a script publishing module in the automated operation and maintenance platform. The operation monitoring module is used to collect the operation data of the system and analyze the operation data to obtain the current operation of the system. status, and determine whether the system needs to perform automated operation and maintenance according to the current operating status; the operation and maintenance module is used to, when the system needs to perform automated operation and maintenance, obtain the preset operation and maintenance knowledge base from the preset operation and maintenance knowledge base according to the current operating status. Match the corresponding fault processing template in the fault processing template, and send the fault processing template to the script publishing module; the script publishing module is used to configure the corresponding operation and maintenance script according to the fault processing template, and send the operation and maintenance script to the fault processing template. The script is released and loaded, so that the operation and maintenance script can be automatically upgraded and iterated according to actual needs, realizing operation and maintenance automation and improving operation and maintenance efficiency. At the same time, there is no need to manually update and iterate the script library, which greatly saves operating costs.

Description of the drawings

Figure 1 is a schematic structural diagram of the first embodiment of the automated operation and maintenance platform of the present invention;

Figure 2 is a schematic structural diagram of the second embodiment of the automated operation and maintenance platform of the present invention;

Figure 3 is a schematic flow chart of the automatic script writing and publishing sub-module automatically configuring the operation and maintenance script in one embodiment of the automated operation and maintenance platform of the present invention;

Figure 4 is a schematic structural diagram of the third embodiment of the automated operation and maintenance platform of the present invention;

Figure 5 is a schematic diagram of establishing a preset operation and maintenance knowledge base in an embodiment of the automated operation and maintenance platform of the present invention;

Figure 6 is a schematic flow chart of the first embodiment of the automated operation and maintenance method of the present invention;

Figure 7 is a schematic diagram of the overall flow of the automated operation and maintenance method in one embodiment of the present invention.

Explanation of reference numbers:

label name label name 10 Run monitoring module 301 Automated scripting publishing submodules 20 Operation and maintenance module 302 Script maintenance submodule 30 Script publishing module 303 Script update management submodule

The realization of the purpose, functional features and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings.

Detailed ways

It should be understood that the specific embodiments described here are only used to explain the present invention and are not intended to limit the present invention.

Referring to Figure 1, Figure 1 is a schematic structural diagram of a first embodiment of an automated operation and maintenance platform.

In this embodiment, the automated operation and maintenance platform includes: an operation monitoring module 10, an operation and maintenance module 20, and a script publishing module 30.

It should be noted that the automated operation and maintenance platform of this embodiment mainly includes three parts, specifically including an operation monitoring module 10, an operation and maintenance module 20, and a script publishing module 30.

In a specific implementation, the operation monitoring module 10 is used to collect the operation data of the system, analyze the operation data, obtain the current operation status of the system, and determine whether the system needs to perform automated operation according to the current operation status. dimension.

The operation monitoring module 10 is mainly used to collect operation data of the platform or system. The operation data includes relevant operation parameters such as the operation status, business data and service data of the platform or system, and analyzes the collected operation data to obtain the system's operation data. Current running status. The operating status data of the system may include abnormal monitoring data of each application module in the system and usage data of each application module in the system.

Specifically, the abnormal monitoring data of the application module can be log monitoring data obtained based on the ELK real-time log analysis system. The log monitoring data is used to record the abnormal data of the processor, memory and storage of each application module (or device) in the system. , in the log monitoring data, 0 can be used to indicate no abnormality, and 1 can be used to indicate abnormality. Of course, according to actual needs, the monitored objects are not limited to processors, memory and storage, but can also include other components, such as network interfaces, power systems, etc. ELK consists of three open source tools: ElasticSearch, Logstash, and Kiabana. The specific method of using ELK to obtain log monitoring data will not be described here. In addition, the usage data of the application module can be host monitoring data obtained based on the Zabbix distributed monitoring system, such as the processor usage, memory usage, storage usage, etc. of the target device, or the usage data can also be Including application monitoring data of applications currently running on the system. For example, you can use tools such as Datadog or New Relic to read/write the memory, number of memory threads, disk IO, and index of the process in the Java virtual machine running on the system. Indicator parameters such as input operations are collected. The business data and service data of the system generally refer to data related to the business services provided externally by the system, and the data can be queried and obtained from the business log or service log of the system.

The current operating status of the system includes normal operating status, abnormal operating status, etc.

After the operation monitoring module 10 obtains the current operating status of the system, it can determine whether the system needs to perform automated operation and maintenance based on the current operating status. By performing risk prediction and business alarms on the current operating status, such as whether there are emergencies or business failures, etc., This determines whether the system requires automated operation and maintenance.

For example, if the current operating state of the system is a normal operating state, the system does not need to perform automated operation and maintenance. If the current operating state of the system is an abnormal operating state, the system needs to perform automated operation and maintenance.

In this embodiment, the operation and maintenance module 20 is used to match the corresponding fault processing template from the preset operation and maintenance knowledge base according to the current operating status when the system needs to perform automated operation and maintenance, and assign the corresponding fault processing template to the preset operation and maintenance knowledge base. The fault handling template is sent to the script publishing module 30.

It should be understood that when the system needs to perform automated operation and maintenance, the current operating state of the system is an abnormal operating state, and the operation monitoring module 10 can output the alarm information and other data of the system, and send the alarm information to the operation and maintenance module 20. The maintenance module 20 is mainly used to match the corresponding fault processing template in the preset operation and maintenance knowledge base according to the alarm information output by the operation monitoring module 10, thereby sending the fault processing template to the script publishing module 30.

The preset operation and maintenance knowledge base can be established in advance. The preset operation and maintenance knowledge base can be based on a machine learning algorithm, perform status mining and analysis on the collected historical operation and maintenance data, and then build an operation and maintenance database based on the analysis results.

There are multiple fault handling templates stored in the preset operation and maintenance knowledge base. When it is necessary to determine the fault handling template, the corresponding fault handling template can be matched from the multiple fault handling templates in the preset operation and maintenance knowledge base based on the alarm information in the current running status. Troubleshooting template.

Further, after receiving the alarm information in the current operating state output by the operation monitoring module 10, the fault type corresponding to the alarm information can be determined through semantic analysis, so that the corresponding fault processing can be retrieved in the preset operation and maintenance knowledge base. template, and sends the retrieved fault handling template to the script publishing module 30.

The script publishing module 30 is configured to configure the corresponding operation and maintenance script according to the fault processing template, and publish and load the operation and maintenance script.

It should be noted that the script publishing module 30 can configure the corresponding operation and maintenance script according to the fault processing template, thereby publishing and loading the operation and maintenance script.

It should be understood that before configuring the operation and maintenance script, the script publishing module 30 will also search whether there is an operation and maintenance script matching the current fault type in the existing preset operation and maintenance script library. When the query finds that there is a matching operation and maintenance script, When running an operation and maintenance script, execute the operation and maintenance script to handle the fault.

In this embodiment, the operation monitoring module 10, the operation and maintenance module 20 and the script publishing module 30 are set up in the automated operation and maintenance platform. The operation monitoring module 10 collects the operating data of the system and analyzes the operating data to obtain the current operating status of the system. And determine whether the system needs to perform automated operation and maintenance according to the current operating status; when the system needs to perform automated operation and maintenance, the operation and maintenance module 20 matches the corresponding fault processing template from the preset operation and maintenance knowledge base according to the current operating status, and handles the fault. The template is sent to the script publishing module 30; the script publishing module 30 configures the corresponding operation and maintenance script according to the fault processing template, and publishes and loads the operation and maintenance script, thereby realizing the automatic upgrade and iteration of the operation and maintenance script according to actual needs, and realizing operation and maintenance automation. , improves operation and maintenance efficiency, and does not require manual update and iteration of the script library, which greatly saves operating costs.

Referring to Figure 2, Figure 2 is a schematic structural diagram of the second embodiment of the automated operation and maintenance platform of the present invention.

Based on the above first embodiment, the script publishing module 30 in this embodiment includes: an automatic script writing and publishing sub-module 301; the automatic script writing and publishing sub-module 301 is used to perform preset operation and maintenance according to the fault handling template. Query the operation and maintenance script that matches the fault type in the script library, and when the operation and maintenance script that matches the fault type is found, publish and load the operation and maintenance script.

It should be noted that the default operation and maintenance script library is created in advance and stores multiple operation and maintenance scripts for handling faults.

After obtaining the fault processing template, the automatic script writing and publishing sub-module 301 can determine the operation and maintenance script matching the fault type in the preset operation and maintenance script library according to the fault processing template, thereby determining whether the operation and maintenance script matching the fault type is queried. dimensional script.

In a specific implementation, when it is queried that there is an operation and maintenance script matching the fault type in the preset operation and maintenance script library, the operation and maintenance script can be directly released and loaded.

In a specific implementation, the automatic script writing and publishing sub-module 301 is also used to automatically configure and generate an operation and maintenance script according to the fault handling template when the operation and maintenance script matching the fault type is not queried, and perform the operation and maintenance script on all the problems. The above operation and maintenance scripts are reviewed and released for loading.

It should be noted that when the automatic script writing and publishing sub-module 301 does not query the operation and maintenance script that matches the fault type, the automatic script writing and publishing sub-module 301 will automatically configure and generate an operation and maintenance script according to the fault processing target, and then perform the operation and maintenance script on the operation and maintenance script. After the maintenance script is reviewed and approved, the operation and maintenance script is released and loaded, and the operation and maintenance script is executed to implement fault handling.

The automatic script writing and publishing sub-module 301 automatically configures the operation and maintenance script to determine the corresponding code according to the fault processing template, thereby automatically configuring the operation and maintenance script. The automatic script writing and publishing sub-module 301 is also used to configure the operation and maintenance script according to the fault processing template. Determine the fault handling method, and determine the corresponding fault handling code structure according to the fault handling method; the automatic script writing and publishing sub-module 301 is also used to determine the corresponding code parameters according to the fault handling code structure, and write the The code parameters are filled in the fault handling code structure, and the operation and maintenance script is generated according to the preset script generation rules.

It should be understood that the automatic script writing and publishing sub-module 301 can determine the fault processing method according to the acquired fault processing template, and then determine the corresponding fault code structure according to the fault processing method. In the fault processing template determined according to the preset operation and maintenance knowledge base, It contains the processing method corresponding to the fault, and there is a large amount of reused code in the operation and maintenance script in the system, and the content and structure of some protocol codes and general method codes are fixed. In addition, the processing code structure corresponding to the same processing method is often the same. Therefore, the automatic script writing and publishing sub-module 301 can perform statistical analysis on historical operation and maintenance scripts to determine the code structures corresponding to different fault handling methods, and then build an operation and maintenance script code library. The operation and maintenance script code base includes several code snippet modules that are modularly encapsulated in advance.

Then, after determining that the fault processing template contains the processing method corresponding to the fault, the automatic script writing and publishing sub-module 301 can determine the code fragments corresponding to each processing method in the operation and maintenance script code library, and then combine these code fragments method to generate a fault handling code structure.

The automatic script writing and publishing sub-module 301 can determine the code parameters corresponding to each part in the fault handling code structure. The method of determining the code parameters can be determined by the current system status and fault type, or by manual input by operation and maintenance personnel. For determining the code parameters based on the current system status and fault type, the historical operation and maintenance parameters and historical system status data can be used as training samples to pre-train the operation and maintenance script release model, and then enter the current system status and fault type into the operation and maintenance script release Model, publish the output results of the model according to the operation and maintenance script, and determine the code parameters. Among them, the historical operation and maintenance data and historical system status data can include: system CPU, memory, network card monitoring data, server query rate per second, request delay, website login information, and other types of data can also be set according to business needs. Operation and maintenance data.

A training sample set is constructed based on these historical data, and a random forest model and a deep neural network model based on operation and maintenance system data are constructed by extracting the characteristic values of each training data in the training sample set. Use the training data in the training sample set to train the deep neural network model, and then obtain the operation and maintenance script release model. Enter the current system status and fault type into the operation and maintenance script release model to obtain the corresponding code parameters.

In addition to automatically generating code parameters through the above method, operation and maintenance personnel are also allowed to log in to the background operation and maintenance script writing page corresponding to the unified script release module through the bastion machine browser, and then fill in each part of the fault handling code structure through manual input. code parameters.

After obtaining the code parameters, the automatic script writing and publishing sub-module 301 can fill in the code parameters into the selected fault handling code structure, and then generate an operation and maintenance script according to the preset script production rules, where the preset script generation rules refer to It is the syntax rules of the script language and the combination rules of the modules. The syntax rules of the script language are used to ensure that the script will not have grammatical errors. The combination rules of the module are used to ensure that the script will not have logical errors. The syntax rules of the script language include, for example, Various variable definitions, function definitions, format requirements, etc. (specific rules can be defined according to the language used by the specific script), module combination rules such as the order of module combination, nested structure, etc.

As shown in Figure 3, Figure 3 is a schematic process diagram of the automatic configuration of operation and maintenance scripts by the automatic script writing and publishing sub-module 301. Code parameters can be automatically generated by operation and maintenance personnel manually inputting code parameters or models, and the automatic script writing and publishing sub-module 301 will The parameters are filled in the fault handling code structure to generate the operation and maintenance script, and the operation and maintenance script is reviewed, including security scanning, blacklist policy query, and business audit policy review. After the review is passed, the operation and maintenance script is released and loaded. .

In this embodiment, an automatic script writing and publishing sub-module 301 is set up in the script publishing module 30; the automatic script writing and publishing sub-module 301 is used to query and match the fault type in the preset operation and maintenance script library according to the fault processing template. The operation and maintenance script, and when the operation and maintenance script matching the fault type is queried, the operation and maintenance script is published and loaded; the automatic script writing and publishing sub-module 301 is also used to query the operation and maintenance script that matches the fault type. When describing an operation and maintenance script that matches the fault type, the operation and maintenance script is automatically configured and generated according to the fault processing template, and the operation and maintenance script is reviewed and released for loading, thereby automatically configuring the operation and maintenance script and improving the operation and maintenance script configuration. flexibility and applicability.

Referring to Figure 4, Figure 4 is a schematic structural diagram of the third embodiment of the automated operation and maintenance platform of the present invention.

Based on the above first embodiment, the script publishing module 30 in this embodiment also includes: a script maintenance sub-module 302 and a script update management sub-module 303;

The script maintenance sub-module 302 is used to monitor the current status information of the system after the current operation and maintenance script is executed at a preset period, and detect whether the current status information is consistent with the preset system status;

It should be noted that when the automatic script writing and publishing sub-module 301 publishes the operation and maintenance script, since the loaded operation and maintenance script is automatically generated based on the preset operation and maintenance knowledge base and the operation and maintenance script publishing model, due to model training, it may There will be a problem that the automatically generated operation and maintenance script is not applicable to the current system failure, which will lead to the problem that the system cannot be restored to the desired state after executing the operation and maintenance script. Therefore, a script maintenance sub-module 302 is also set up in the script publishing module 30, so that the script can be maintained The sub-module 302 monitors the operation and maintenance scripts to ensure that there are no faults in the operation and maintenance scripts.

The preset period can be set in hours or days. For example, the current status information of the system after the execution of the current operation and maintenance script is monitored every 3 hours. The current status information data of the system after the execution of the current operation and maintenance script can be obtained to determine whether the current status information is It is consistent with the preset system status to determine whether the operation and maintenance script is valid. The default system state is the ideal or expected state of the system after the current operation and maintenance script is executed.

In this embodiment, the preset system status can be obtained through the operation and maintenance script running status monitoring model, and the script maintenance sub-module 302 is also used to obtain the operation and maintenance script running status monitoring model, and obtain Fault information; the script maintenance sub-module 302 is also used to input the fault information and the current operation and maintenance script into the operation and maintenance script running status monitoring model to obtain the preset system status after executing the current operation and maintenance script.

It should be noted that after the operation and maintenance script is determined, the operation and maintenance script and fault information can be input into the operation and maintenance script running status monitoring model, so that the expected system state after executing the operation and maintenance script can be predicted.

Furthermore, the operation and maintenance script running status monitoring model can be established in advance, so the script maintenance sub-module 302 is also used to obtain historical operation and maintenance data, and preprocess the historical operation and maintenance data to obtain training corresponding to the historical operation and maintenance data. Data; the script maintenance sub-module 302 is also used to perform training based on the training data to obtain an operation and maintenance script running status monitoring model.

It should be noted that historical operation and maintenance data includes: fault type, fault system status information, operation and maintenance scripts, and system status information after executing the operation and maintenance scripts. By preprocessing historical operation and maintenance data, training data corresponding to historical operation and maintenance data is obtained. Among them, preprocessing mainly includes: filling missing values, checking and removing noise values and outliers through data cleaning technology; using data dimensionality reduction technology, using data supervision methods to achieve the dimensionality reduction effect of the target data; using text cleaning technology, Delete redundant features or cluster to eliminate redundant data; use data discretization technology to reduce data storage space and convert continuous values into categorical features for specific machine learning methods; use data normalization processing technology to extract feature vectors Perform normalization processing. The preprocessed operation and maintenance feature data set can be divided into two parts, of which 70-90% of the data set is used as a training set to train the model, and 10-30% of the data set is used as a test set to evaluate the model effect. Then use 70%-90% of the data set as the training set to train the operation and maintenance script running status monitoring model.

Specifically, the decision tree CART analysis method can be used, using the Gini coefficient as a measure of feature division. When predicting, a certain attribute value is used to make a judgment at the internal node of the tree, and based on the judgment result, it is decided which branch node to enter until the leaf node is reached. at , get the classification results. The CART pruning algorithm is used to subtract some subtrees from the bottom of the fully grown decision tree to simplify the decision tree model, thereby enabling more accurate predictions of unknown data.

According to the training data, starting from the root node, each node is recursively operated by inputting the training data set and calculating the stopping condition, and then outputs a binary decision tree; according to the CART pruning algorithm, the binary decision tree is As input, the optimal decision tree is output. Through the above process, the training of the operation and maintenance script running status monitoring model is completed.

Through the trained operation and maintenance script running status monitoring model, it is possible to predict the normal status information of the system in each period after the operation and maintenance script is executed and when the operation and maintenance script runs normally.

For example, the abnormal system fault detected is: "Insufficient system memory (abnormal occupancy rate is 90%), resulting in memory read and write failure", and the unified script publishing platform generated operation and maintenance script a based on this fault. After operation and maintenance script a is executed, , through the operation and maintenance script running status monitoring model, the system status expected by the user in each period after executing the operation and maintenance script can be predicted. The assumptions are: "Predicted status a: The operation and maintenance script is executed for one hour, and the system memory usage is 80%. ; Prediction state b: The operation and maintenance script has been executed for two hours, and the system memory usage is 60%; Measurement status c: The operation and maintenance script has been executed for three hours, and the system memory usage is 50%; Prediction status d: The operation and maintenance script has been executed for four hours, and the system memory usage is 50% "The occupancy rate is 40%, memory reading and writing is normal, and the fault is eliminated", then the unified script publishing platform can periodically collect the current system status data according to the time period of the predicted status after the operation and maintenance script is released and executed, and then compare the current system status Data and predicted status are used to determine whether the operation and maintenance script is valid. When the judgment result is yes, the operation and maintenance script continues to be executed and periodic monitoring continues until the fault is eliminated, operation and maintenance is completed, and the unified script publishing platform can Publish the operation and maintenance script and the corresponding fault type to the preset operation and maintenance script library for storage, so that when similar faults occur later, the corresponding operation and maintenance script can be directly selected from the preset operation and maintenance script library for processing.

In this embodiment, the script maintenance sub-module 302 is also used to generate update information to the script update management sub-module 303 when the current status information is inconsistent with the preset system status; the script update management sub-module 303 Module 303 is configured to obtain a status difference value based on the current status information and the preset system status based on the update information, and update the current operation and maintenance script according to the status difference value.

When the current status information is inconsistent with the preset system status, the current operation and maintenance script needs to be updated. Therefore, the script maintenance sub-module 302 can generate update information to the script update management sub-module 303, so that the current operation and maintenance script can be updated through the script update management sub-module 303. Dimension scripts are updated.

After the script update management sub-module 303 receives the update information, it can obtain the current system status value and the predicted system status value based on the current status information and the preset system status, thereby calculating the difference between the current system status value and the predicted system status value. , obtain the status difference, and update the current operation and maintenance script through the status difference, terminate the execution of the previous operation and maintenance script, re-release and execute the updated operation and maintenance script, and execute the preset cycle through the script maintenance sub-module 302 The process of monitoring the current status information of the system after the current operation and maintenance script is executed, and detecting whether the current status information is consistent with the preset system status.

Specifically, the script update management sub-module 303 can use the pre-trained operation and maintenance script status monitoring model to determine the corresponding optimization parameters (including optimization of the operation and maintenance code structure) based on the difference between the current system status value and the predicted system status value. Optimization parameters and optimization of specific code parameters). Then the unified script publishing module can update and iterate the code of the original operation and maintenance script according to the optimization parameters, thereby obtaining the updated and iterated operation and maintenance script, and iteratively publish the operation and maintenance script to update the original operation and maintenance script in the script library. Dimension scripts are updated iteratively.

In this embodiment, a script maintenance sub-module 302 and a script update management sub-module 303 are also provided in the script publishing module 30; the script maintenance sub-module 302 is used to obtain the current status information of the system after the current operation and maintenance script is executed in a preset period Monitor and detect whether the current status information is consistent with the preset system status; the script maintenance sub-module 302 is also used to generate update information to the script update when the current status information is inconsistent with the preset system status. Management sub-module 303; the script update management sub-module 303 is used to obtain the status difference between the current status information and the preset system status based on the update information, and perform current operation and maintenance based on the status difference. The script is updated, and the script maintenance sub-module 302 automatically evaluates and analyzes the working effect of the currently running operation and maintenance script based on the system status and historical operation and maintenance data. When it is judged that the currently running operation and maintenance script cannot be solved as expected by the system In the event of a failure, the script update management sub-module 303 can automatically update and upgrade the operation and maintenance script, and re-release the updated operation and maintenance script, so that the operation and maintenance script can be automatically upgraded and iterated according to actual needs, thereby Realize operation and maintenance automation.

Based on the above first embodiment, a fourth embodiment of the automated operation and maintenance platform of the present invention is proposed.

Based on the above first embodiment, the operation and maintenance module 20 in this embodiment is also used to obtain system historical operation and maintenance data, and preprocess the system historical operation and maintenance data to obtain an operation and maintenance data text set; The operation and maintenance module 20 is also used to perform data association rule mining on the operation and maintenance data text set, and convert the historical operation and maintenance data in the operation and maintenance data text set into operation and maintenance knowledge data; the operation and maintenance module 20 is also used to Establishing a preset operation and maintenance knowledge base based on the operation and maintenance knowledge data.

It should be noted that the operation and maintenance module 20 can establish a preset operation and maintenance knowledge base based on the system's historical operation and maintenance data. The system's historical operation and maintenance data includes: fault text description, operation and maintenance processing plan, and processing results corresponding to the operation and maintenance processing plan. . As shown in Figure 5, Figure 5 is a schematic diagram for establishing a preset operation and maintenance knowledge base. By collecting system historical operation and maintenance data, including IP, asset chemistry, internal and external data, etc., and preprocessing system historical operation and maintenance data, you can Perform preprocessing such as unified format conversion and data cleaning on the acquired historical operation and maintenance data to obtain the initial operation and maintenance data text collection; then stop words can be removed for the data in the operation and maintenance data text collection to obtain the reference operation and maintenance data. Text collection; next, mark the text in the reference operation and maintenance text collection. You can use the BIO (BeginIntermediate Other) annotation method in natural language processing technology to sequence the operation and maintenance statements in the reference operation and maintenance data text collection. Mark the corresponding category information for each corpus entity. For example: "Process detection failure" can be marked as: progress (B-NA) process (B-NA) detection (I-VB) detection (I-VB) reason (I -NB) barrier (I-NB) to obtain the operation and maintenance data text collection. Before performing association relationship mining operations on the operation and maintenance data text collection, each marked entity is represented in the form of a vector in advance in the form of a vector, including its position, label and other information in the original data set. The marked text set can be input into the word2vec model. , using the Skip-Gram and CBOW (Continuous Bag-of-Words) models under the word2vec framework to convert the input data into vector form, and convert the operation and maintenance data text into an operation and maintenance word vector set.

The operation and maintenance data text collection can be used for data association rule mining, and historical operation and maintenance data can be converted into operation and maintenance knowledge data. Specifically, the operation and maintenance word vector set can be converted from the operation and maintenance data text and input into the association mining algorithm (such as generateRules function, or Apriori algorithm) to obtain frequent itemsets and confidence. Input the operation and maintenance word vector set into the Bi-LSTM model to obtain a word vector matrix with contextual information; input the word vector matrix into the TransE model, and then jointly calculate it through the Euclidean formula to obtain the primary relationship threshold; The calculation of the primary relationship threshold is repeated, and then the critical value of the relationship threshold is determined based on multiple relationship threshold results; the frequent item set, confidence level and critical value are jointly passed into the association rule mining algorithm model.

According to the theorem "For each non-empty subset x of frequent itemset l, calculate confidence(x→(l-x)), if confidence(x→(l-x))confmin, then the rule x→(l-x)" holds, therefore, the association The rule mining algorithm obtains each subset by scanning frequent item sets, and calculates the confidence. When the confidence meets the conditions (that is, greater than or equal to the minimum confidence), a rule is generated, and finally each operation and maintenance in the operation and maintenance data text collection is obtained. Association relationships between entities, and then determine operation and maintenance knowledge data based on the association relationships. As shown in Table 1 below, Table 1 is the obtained operation and maintenance knowledge data table.

Table 1

serial number accident details type Solutions result 1 Process interruption A plan 1 recover 2 block B Scenario 2 recover 3 load C Option 3 recover

It should be understood that after the operation and maintenance knowledge data is obtained, a preset operation and maintenance knowledge base can be established.

In this embodiment, the operation and maintenance module 20 is also used to establish a search engine in the preset operation and maintenance knowledge base, and store the operation and maintenance knowledge data in the preset operation and maintenance knowledge base. The operation and maintenance module 20 is also used to obtain alarm information according to the operating status, and determine the fault type corresponding to the alarm information through semantic analysis; the operation and maintenance module 20 is also used to obtain alarm information according to the fault type. The fault processing template corresponding to the fault type is matched in the preset operation and maintenance knowledge base, and the fault processing template is sent to the script publishing module 30 .

It can be understood that the operation and maintenance module 20 is also used to establish a search engine in the preset operation and maintenance knowledge base, and use production representation to save the acquired operation and maintenance knowledge data into the preset operation and maintenance knowledge base.

After the operation and maintenance module 20 receives the running status, the alarm information can be obtained according to the running status, and the fault type corresponding to the alarm information can be determined through semantic analysis, and then the corresponding fault processing template can be retrieved in the preset operation and maintenance knowledge base. Send the fault handling template to the script publishing module 30.

The operation and maintenance module 20 in this embodiment is also used to obtain system historical operation and maintenance data, and preprocess the system historical operation and maintenance data to obtain an operation and maintenance data text collection; the operation and maintenance module 20 is also used to obtain Conduct data association rule mining on the operation and maintenance data text set, and convert the historical operation and maintenance data in the operation and maintenance data text set into operation and maintenance knowledge data; the operation and maintenance module 20 is also used to analyze the operation and maintenance knowledge data based on the operation and maintenance knowledge data. Establish a preset operation and maintenance knowledge base. By establishing a preset operation and maintenance knowledge base in advance, you can quickly match the corresponding fault processing template in the preset operation and maintenance knowledge base according to the fault type to improve processing efficiency.

Based on the above first embodiment, a fifth embodiment of the automated operation and maintenance platform of the present invention is proposed.

Based on the above first embodiment, the operation monitoring module 10 in this embodiment is also used to collect historical fault data and historical operating status data of the system.

In a specific implementation, after obtaining the operation data, the operation monitoring module 10 can also analyze the operation data to determine the current operating status of the system. The current operating status of the system can be detected according to the alarm prediction model, and the alarm prediction model can be established in advance. , therefore the operation monitoring module 10 is also used to collect historical fault data and historical operating status data when establishing an alarm prediction model, where the historical fault data may include: fault types or alarm names that have occurred during the historical operation of the system, etc. .

The operation monitoring module 10 is also used to perform training based on the historical fault data and the historical operating status data to establish an alarm prediction model.

Since a wide variety of applications are integrated on the cloud service system, the system will also receive a wide variety and a huge amount of fault data. In order to better deepen the analysis of fault data by operation and maintenance personnel and testers, so that it can be based on the Build an alarm prediction model based on some historical fault data. After obtaining the historical fault data, it is necessary to classify and manage the collected historical fault data. By filtering, normalizing, statistical analyzing, and labeling the historical fault data, we can obtain the fault (or alarm) data. ) types and fault alarm characteristic quantities. Subsequently, the collected historical fault data can be preprocessed according to the fault alarm knowledge base to obtain a training sample set.

In this solution, the pre-collected historical fault (alarm) data and historical operating status data of the system can be used as training samples, and the long short-term memory network (Long Short-Term Memory, LSTM) can be used to pre-train the alarm prediction model to establish Alarm prediction model.

The alarm prediction model can be trained as follows: first, build an alarm prediction model based on the LSTM neural network, and then input the historical operating parameters and corresponding fault types in the training sample set into the LSTM neural network for training, and output the predicted fault type. The predicted fault type and the actual fault type are evaluated to obtain the loss value. Based on the loss value, it is judged whether the model training effect is ideal. When the final loss value is less than the preset threshold, the training is completed and a fault prediction model that meets the requirements is obtained.

The operation monitoring module 10 is also used to input the operation data into the alarm prediction model to obtain the current operating status prediction value of the system; the operation monitoring module 10 is also used to input the current operating status prediction value to Compare with the preset alarm threshold, and when the current operating status prediction value is greater than the preset alarm threshold, it is determined that the system needs to perform automated operation and maintenance.

It should be noted that after the alarm prediction model is established, the current operating status of the system can be predicted based on the operating data collected during the current monitoring cycle based on the alarm prediction model. By inputting the operating data into the alarm prediction model, the current operation of the system can be obtained. Status prediction value.

The preset alarm threshold can be set according to actual needs. The preset alarm threshold reflects the critical state of system operation. When the predicted value of the current operating status is greater than the preset alarm threshold, the system is currently abnormal and requires alarm processing, that is, automated operation is required. dimension. When it is determined based on the prediction results that there is currently an abnormality in the system that requires alarm processing, the operation monitoring module 10 will send alarm information to the operation and maintenance module 20 and input the fault type corresponding to the alarm information at the same time.

In a specific implementation, the operation monitoring module 10 is also used to determine that the system does not need to perform automated operation and maintenance when the current operating status prediction value is less than or equal to the preset alarm threshold.

When the current operating status prediction value is less than or equal to the preset alarm threshold, there is no abnormality in the system, and automated operation and maintenance is not required, and the system can continue to be monitored through the operation monitoring module 10 .

The operation monitoring module 10 in this embodiment is also used to collect historical fault data and historical operating status data of the system; the operation monitoring module 10 is also used to perform monitoring according to the historical fault data and the historical operating status data. training to establish an alarm prediction model; the operation monitoring module 10 is also used to input the operation data into the alarm prediction model to obtain the current operating status prediction value of the system; the operation monitoring module 10 is also used to The current operating status prediction value is compared with the preset alarm threshold, and when the current operating status prediction value is greater than the preset alarm threshold, it is determined that the system needs to perform automated operation and maintenance; the operation monitoring module 10 also uses When the predicted value of the current operating state is less than or equal to the preset alarm threshold, it is determined that the system does not need to perform automated operation and maintenance, and the operating data is input into the established alarm prediction model to quickly determine the predicted value of the current operating state of the system. , thereby further determining whether the system needs automated operation and maintenance, and improving the automation of operation and maintenance.

An embodiment of the present invention provides an automated operation and maintenance method. Refer to FIG. 6 , which is a schematic flow chart of the first embodiment of the automated operation and maintenance method of the present invention. The automated operation and maintenance method is applied to the automated operation and maintenance platform described above.

In this embodiment, the automated operation and maintenance method includes the following steps:

Step S10: The operation monitoring module collects the operation data of the system, analyzes the operation data, obtains the current operation status of the system, and determines whether the system requires automated operation and maintenance based on the current operation status.

The operation monitoring module is mainly used to collect the operation data of the platform or system. The operation data includes the operation status, business data and service data of the platform or system and other related operation parameters. It analyzes the collected operation data to obtain the current status of the system. Operating status. The operating status data of the system may include abnormal monitoring data of each application module in the system and usage data of each application module in the system.

Specifically, the abnormal monitoring data of the application module can be log monitoring data obtained based on the ELK real-time log analysis system. The log monitoring data is used to record the abnormal data of the processor, memory and storage of each application module (or device) in the system. , in the log monitoring data, 0 can be used to indicate no abnormality, and 1 can be used to indicate abnormality. Of course, according to actual needs, the monitored objects are not limited to processors, memory and storage, but can also include other components, such as network interfaces, power systems, etc. ELK consists of three open source tools: ElasticSearch, Logstash, and Kiabana. The specific method of using ELK to obtain log monitoring data will not be described here. In addition, the usage data of the application module can be host monitoring data obtained based on the Zabbix distributed monitoring system, such as the processor usage, memory usage, storage usage, etc. of the target device, or the usage data can also be Including application monitoring data of applications currently running on the system. For example, you can use tools such as Datadog or New Relic to read/write the memory, number of memory threads, disk IO, and index of the process in the Java virtual machine running on the system. Indicator parameters such as input operations are collected. The business data and service data of the system generally refer to data related to the business services provided externally by the system, and the data can be queried and obtained from the business log or service log of the system.

The current operating status of the system includes normal operating status, abnormal operating status, etc.

When the operation monitoring module obtains the current operating status of the system, it can determine whether the system requires automated operation and maintenance based on the current operating status. It can perform risk prediction and business alarms on the current operating status, such as whether there are emergencies or business failures, etc. Determine whether the system requires automated operation and maintenance.

For example, if the current operating state of the system is a normal operating state, the system does not need to perform automated operation and maintenance. If the current operating state of the system is an abnormal operating state, the system needs to perform automated operation and maintenance.

Step S20: When the system needs to perform automated operation and maintenance, the operation and maintenance module matches the corresponding fault processing template from the preset operation and maintenance knowledge base according to the current operating status, and sends the fault processing template to the script publishing module .

It should be understood that when the system requires automated operation and maintenance, the current operating state of the system is an abnormal operating state, and the operation monitoring module can output the alarm information and other data of the system, and send this alarm information to the operation and maintenance module. The operation and maintenance module It is mainly used to match the corresponding fault processing template in the preset operation and maintenance knowledge base based on the alarm information output by the operation monitoring module, thereby sending the fault processing template to the script publishing module.

The preset operation and maintenance knowledge base can be established in advance. The preset operation and maintenance knowledge base can be based on a machine learning algorithm, perform status mining and analysis on the collected historical operation and maintenance data, and then build an operation and maintenance database based on the analysis results.

There are multiple fault handling templates stored in the preset operation and maintenance knowledge base. When it is necessary to determine the fault handling template, the corresponding fault handling template can be matched from the multiple fault handling templates in the preset operation and maintenance knowledge base based on the alarm information in the current running status. Troubleshooting template.

Further, after receiving the alarm information in the current operating state output by the operation monitoring module, the fault type corresponding to the alarm information can be determined through semantic analysis, so that the corresponding fault processing template can be retrieved in the preset operation and maintenance knowledge base. , and sends the retrieved fault handling template to the script publishing module.

Step S30: The script publishing module configures the corresponding operation and maintenance script according to the fault processing template, and publishes and loads the operation and maintenance script.

It should be noted that the script publishing module can configure the corresponding operation and maintenance script according to the fault handling template, thereby publishing and loading the operation and maintenance script.

It should be understood that before configuring the operation and maintenance script, the script publishing module will also search in the existing preset operation and maintenance script library to see if there is an operation and maintenance script that matches the current fault type. When the query finds that there is a matching operation and maintenance script, When running an operation and maintenance script, execute the operation and maintenance script to handle faults.

As shown in Figure 7, Figure 7 is a schematic diagram of the overall process of the automated operation and maintenance method. Step 1: Monitor and collect operation-related data such as the system's operating status, business data, and service data through the operation monitoring module; Step 2: The operation monitoring module monitors and collects Analyze operation-related data to determine the current operating status of the system, and determine whether automated operation and maintenance is required based on the current operating status of the system; Step 3: The operation and maintenance module uses the pre-built operation and maintenance knowledge based on the fault type carried in the alarm prompt. Perform fault matching in the library to determine the corresponding fault handling template; Step 4: The script publishing module configures the operation and maintenance script according to the fault handling template obtained through step 3, and publishes and loads the operation and maintenance script; Step 5: The script publishing module will According to the preset cycle, monitor the status information of the system after the execution of the operation and maintenance script, and determine whether the system status after the execution of the operation and maintenance script is consistent with the expected system status. When the system status after the execution of the operation and maintenance script is consistent with the expected system status, Continue to monitor the system status information after the operation and maintenance script is executed. When the system status after the operation and maintenance script is executed is inconsistent with the expected system status, step 6: Determine the difference between the current system status value and the predicted system status value, based on This difference will update the current operation and maintenance script.

This embodiment collects the operating data of the system through the operation monitoring module, analyzes the operating data, obtains the current operating status of the system, and determines whether the system requires automated operation and maintenance based on the current operating status; the operation and maintenance module is located where When the system needs to perform automated operation and maintenance, the corresponding fault processing template is matched from the preset operation and maintenance knowledge base according to the current operating status, and the fault processing template is sent to the script publishing module; the script publishing module Configure the corresponding operation and maintenance script in the above fault handling template, and publish and load the operation and maintenance script, so that the operation and maintenance script can be automatically upgraded and iterated according to actual needs, realize operation and maintenance automation, improve operation and maintenance efficiency, and do not need to modify the operation and maintenance script. The script library performs manual update and iteration, which greatly saves operating costs.

In some embodiments, the script publishing module includes: an automatic script writing and publishing sub-module, the script publishing module configures the corresponding operation and maintenance script according to the fault handling template, and publishes and loads the operation and maintenance script. Specifically: the automatic script writing and publishing sub-module queries the preset operation and maintenance script library for operation and maintenance scripts that match the fault type according to the fault processing template, and when the operation and maintenance script that matches the fault type is queried, , publish and load the operation and maintenance script; when the automatic script writing and publishing sub-module does not query the operation and maintenance script that matches the fault type, it automatically configures and generates the operation and maintenance script according to the fault handling template, and The operation and maintenance script is reviewed and released for loading.

In some embodiments, the steps of automatically configuring and generating an operation and maintenance script according to the fault handling template, and reviewing and publishing and loading the operation and maintenance script specifically include: the automatic script writing and publishing sub-module according to the fault The processing template determines the fault processing method, and determines the corresponding fault processing code structure according to the fault processing method; the automatic script writing and publishing sub-module determines the corresponding code parameters according to the fault processing code structure, and fills in the code parameters. into the fault handling code structure, and generate operation and maintenance scripts according to preset script generation rules.

In some embodiments, the script publishing module also includes: a script maintenance sub-module and a script update management sub-module; the script publishing module configures the corresponding operation and maintenance script according to the fault handling template, and sends the operation and maintenance script to After publishing and loading, it also includes: the script maintenance sub-module monitors the current status information of the system after the current operation and maintenance script is executed at a preset period, and detects whether the current status information is consistent with the preset system status; the script maintenance The sub-module generates update information to the script update management sub-module when the current status information is inconsistent with the preset system status; the script update management sub-module generates update information based on the update information according to the current status information and the preset system status. It is assumed that the system status obtains the status difference value, and the current operation and maintenance script is updated according to the status difference value.

In some embodiments, the script maintenance sub-module monitors the current status information of the system after the current operation and maintenance script is executed at a preset period, and before detecting whether the current status information is consistent with the preset system status, the method further includes: The script maintenance sub-module obtains the operation and maintenance script running status monitoring model, and obtains fault information according to the current running status; the script maintenance sub-module inputs the fault information and the current operation and maintenance script to the operation and maintenance script running status monitoring model to obtain the preset system state after executing the current operation and maintenance script.

In some embodiments, the steps for the script maintenance sub-module to obtain the operation and maintenance script running status monitoring model specifically include: the script maintenance sub-module obtains historical operation and maintenance data, and preprocesses the historical operation and maintenance data to obtain Training data corresponding to historical operation and maintenance data; the script maintenance sub-module performs training based on the training data to obtain an operation and maintenance script running status monitoring model.

In some embodiments, when the system needs to perform automated operation and maintenance, the operation and maintenance module matches the corresponding fault processing template from the preset operation and maintenance knowledge base according to the current operating status, and adds the fault processing template to the preset operation and maintenance knowledge base. Before sending to the script publishing module, it also includes: the operation and maintenance module obtains the system historical operation and maintenance data, and preprocesses the system historical operation and maintenance data to obtain the operation and maintenance data text set; the operation and maintenance module The operation and maintenance data text collection performs data association rule mining and converts the historical operation and maintenance data in the operation and maintenance data text collection into operation and maintenance knowledge data; the operation and maintenance module establishes a preset operation and maintenance knowledge base based on the operation and maintenance knowledge data.

In some embodiments, after the operation and maintenance module establishes a preset operation and maintenance knowledge base according to the operation and maintenance knowledge data, it further includes: the operation and maintenance module establishes a search engine in the preset operation and maintenance knowledge base, and Store the operation and maintenance knowledge data in the preset operation and maintenance knowledge base; match the corresponding fault processing template from the preset operation and maintenance knowledge base according to the current operating status, and send the fault processing template to the script publishing module The steps specifically include: the operation and maintenance module obtains alarm information according to the operating status, and determines the fault type corresponding to the alarm information through semantic analysis; the operation and maintenance module obtains alarm information according to the fault type according to the preset operation status. Match the fault processing template corresponding to the fault type in the dimension knowledge base, and send the fault processing template to the script publishing module.

In some embodiments, the operation monitoring module collects the operation data of the system, analyzes the operation data, obtains the current operation status of the system, and determines whether the system requires automated operation and maintenance based on the current operation status. Specifically, it includes: the operation monitoring module collects historical fault data and historical operation status data of the system; the operation monitoring module performs training based on the historical fault data and the historical operation status data, and establishes an alarm prediction model; the operation monitoring module The module inputs the operating data into the alarm prediction model to obtain the current operating status prediction value of the system; the operation monitoring module compares the current operating status prediction value with the preset alarm threshold and performs When the predicted status value is greater than the preset alarm threshold, it is determined that the system needs to perform automated operation and maintenance; when the predicted current operating status value is less than or equal to the preset alarm threshold, the operation monitoring module determines that the system does not need to be automated. dimension.

It should be understood that the above are only examples and do not constitute any limitation on the technical solution of the present invention. In specific applications, those skilled in the art can make settings as needed, and the present invention does not impose any limitations on this.

It should be noted that the workflow described above is only illustrative and does not limit the scope of the present invention. In practical applications, those skilled in the art can select some or all of them for implementation according to actual needs. The purpose of this embodiment is not limited here.

In addition, for technical details that are not described in detail in this embodiment, please refer to the automated operation and maintenance method provided by any embodiment of the present invention, and will not be described again here.

Furthermore, it should be noted that, as used herein, the terms "include", "comprises" or any other variation thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or system that includes a list of elements includes not only those elements, but also other elements not expressly listed or elements inherent to the process, method, article or system. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

The above serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments.

Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product that is essentially or contributes to the existing technology. The computer software product is stored in a storage medium (such as a read-only memory). , ROM)/RAM, magnetic disk, optical disk), including several instructions to cause a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the method described in various embodiments of the present invention.

The above are only preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the description and drawings of the present invention may be directly or indirectly used in other related technical fields. , are all similarly included in the scope of patent protection of the present invention.

Claims (10)

1. An automated operation and maintenance platform, characterized in that the automated operation and maintenance platform comprises: the system comprises an operation monitoring module, an operation maintenance module and a script issuing module;

the operation monitoring module is used for collecting operation data of the system, analyzing the operation data to obtain the current operation state of the system, and determining whether the system needs to be subjected to automatic operation or not according to the current operation state;

the operation and maintenance module is used for matching a corresponding fault processing template from a preset operation and maintenance knowledge base according to the current operation state when the system needs to be automatically operated and maintained, and sending the fault processing template to the script issuing module;

and the script issuing module is used for configuring the corresponding operation and maintenance script according to the fault processing template and issuing and loading the operation and maintenance script.

2. The automated operation and maintenance platform of claim 1, wherein the script issuing module comprises: an automatic script writing and issuing sub-module;

the automatic script writing and issuing sub-module is used for inquiring an operation and maintenance script matched with the fault type in a preset operation and maintenance script library according to the fault processing template, and issuing and loading the operation and maintenance script when inquiring the operation and maintenance script matched with the fault type;

And the automatic script writing and issuing sub-module is also used for automatically configuring and generating an operation and maintenance script according to the fault processing template when the operation and maintenance script matched with the fault type is not inquired, and auditing, issuing and loading the operation and maintenance script.

3. The automated operation and maintenance platform according to claim 2, wherein the automated script writing and publishing sub-module is further configured to determine a fault handling method according to the fault handling template, and determine a corresponding fault handling code structure according to the fault handling method;

the automatic script writing and issuing sub-module is further used for determining corresponding code parameters according to the fault processing code structure, filling the code parameters into the fault processing code structure and generating an operation and maintenance script according to a preset script generation rule.

4. The automated operation and maintenance platform of claim 1, wherein the script issuing module further comprises: a script maintenance sub-module and a script update management sub-module;

the script maintenance submodule is used for monitoring current state information of the system after the current operation and maintenance script is executed in a preset period and detecting whether the current state information is consistent with a preset system state or not;

The script maintenance sub-module is further configured to generate update information to the script update management sub-module when the current state information is inconsistent with a preset system state;

the script updating management sub-module is used for obtaining a state difference value according to the current state information and the preset system state based on the updating information, and updating the current operation and maintenance script according to the state difference value.

5. The automated operation and maintenance platform according to claim 4, wherein the script maintenance sub-module is further configured to obtain an operation and maintenance script operation state monitoring model, and obtain fault information according to the current operation state;

the script maintenance sub-module is further configured to input the fault information and the current operation and maintenance script into the operation and maintenance script running state monitoring model, so as to obtain a preset system state after the current operation and maintenance script is executed.

6. The automated operation and maintenance platform according to claim 5, wherein the script maintenance sub-module is further configured to obtain historical operation and maintenance data, and perform preprocessing on the historical operation and maintenance data to obtain training data corresponding to the historical operation and maintenance data;

and the script maintenance sub-module is also used for training according to the training data to obtain an operation and maintenance script running state monitoring model.

7. The automated operation and maintenance platform according to claim 1, wherein the operation and maintenance module is further configured to obtain system historical operation and maintenance data, and perform preprocessing on the system historical operation and maintenance data to obtain an operation and maintenance data text set;

the operation and maintenance module is further used for carrying out data association rule mining on the operation and maintenance data text set and converting historical operation and maintenance data in the operation and maintenance data text set into operation and maintenance knowledge data;

the operation and maintenance module is also used for establishing a preset operation and maintenance knowledge base according to the operation and maintenance knowledge data.

8. The automated operation and maintenance platform according to claim 7, wherein the operation and maintenance module is further configured to build a search engine in the preset operation and maintenance knowledge base and store operation and maintenance knowledge data in the preset operation and maintenance knowledge base;

the operation and maintenance module is further used for obtaining alarm information according to the operation state and determining a fault type corresponding to the alarm information through semantic analysis;

the operation and maintenance module is further configured to match a fault handling template corresponding to the fault type in the preset operation and maintenance knowledge base according to the fault type, and send the fault handling template to the script issuing module.

9. The automated operation and maintenance platform of any one of claims 1 to 8, wherein the operation monitoring module is further configured to collect historical fault data and historical operating status data of the system;

the operation monitoring module is also used for training according to the historical fault data and the historical operation state data, and establishing an alarm prediction model;

the operation monitoring module is also used for inputting the operation data into the alarm prediction model to obtain a current operation state predicted value of the system;

the operation monitoring module is further used for comparing the current operation state predicted value with a preset alarm threshold value and determining that the system needs to perform automatic operation and maintenance when the current operation state predicted value is greater than the preset alarm threshold value;

the operation monitoring module is further configured to determine that the system does not need to perform automated operation and maintenance when the predicted value of the current operation state is less than or equal to the preset alarm threshold.

10. An automated operation and maintenance method, wherein the automated operation and maintenance method is applied to the automated operation and maintenance platform according to any one of claims 1 to 9, and the automated operation and maintenance method comprises:

The operation monitoring module collects operation data of the system, analyzes the operation data to obtain the current operation state of the system, and determines whether the system needs to be automatically operated and maintained according to the current operation state;

when the system needs to be automatically operated, the operation and maintenance module matches a corresponding fault processing template from a preset operation and maintenance knowledge base according to the current operation state and sends the fault processing template to a script issuing module;

and the script issuing module configures a corresponding operation and maintenance script according to the fault processing template, and issues and loads the operation and maintenance script.