Alarm message processing method, device and electronic equipment

By receiving and counting the IP addresses of the alert message, identifying and ignoring the alert message exceeding the threshold, the event storm problems caused by device failure are solved, the system paralysis and high operation and maintenance costs are avoided, and the alarm message is timely. Processing and system stability.

CN114338346BActive Publication Date: 2024-11-19INDUSTRIAL AND COMMERCIAL BANK OF CHINA
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111682310.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-29
Publication Date
2024-11-19
Estimated Expiration
2041-12-29

AI Technical Summary

Technical Problem

When the equipment fails, the existing technology causes a large number of alarm messages to be sent to the monitoring platform in a short time, causing an event storm, causing server memory overflow, monitoring platform stuck or paralyzed, and operation and maintenance personnel need to manually shut down, time and labor costs. High, unable to handle alarm messages from other equipment in a timely manner, which may cause security incidents.

Method used

By receiving the alarm message, determining its IP address, and counting the number of alarm messages of the same IP address within a predetermined time. If the threshold is exceeded, the alarm messages of the IP address will be ignored, and the IP address will be sent to the terminal for troubleshooting to avoid events Storm to prevent system paralysis.

Benefits of technology

It effectively avoids system paralysis caused by incident storms, reduces storage and operation pressure, ensures timely processing of alarm messages, and reduces operation and maintenance costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114338346B_ABST
    Figure CN114338346B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, device and electronic device for processing an alarm message. The method includes: receiving a first alarm message; determining a first Internet Protocol IP address of the first alarm message; determining the number of second alarm messages collected within a predetermined time, wherein the second alarm message is an alarm message with the same first IP address as the first alarm message; ignoring the alarm message sent by the first terminal corresponding to the second IP address when the number of collected second alarm messages exceeds a predetermined threshold, wherein the second IP address is the IP address of the second alarm message; sending the second IP address to the second terminal to perform fault processing on the first terminal corresponding to the second IP address. The present invention solves the technical problem of difficulty in timely processing of event storms when event storms occur in the related art.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of information security, and in particular to a method, device and electronic equipment for processing an alarm message. Background Art

[0002] At present, the alarm message monitoring solution process in the related technology is roughly as follows: 1) collect alarm messages sent by the device, 2) identify the alarm messages and define the level, and 3) send the high-level alarm messages to the monitoring platform page. The purpose of sending the alarm messages to the monitoring platform interface to process the alarm messages is achieved. However, when the solution in the related technology is used to solve the problem, when the device fails, millions of data are sent to the monitoring platform in a short period of time. When an event storm occurs, it is easy to cause the server memory to explode, the monitoring platform page to get stuck, and even the entire alarm message monitoring system to be paralyzed. The operation and maintenance personnel need to manually close the alarm messages from the monitoring platform page, which is too time-consuming and labor-intensive. And during this period, it is very likely that other normal device alarm messages cannot be sent to the monitoring platform page, resulting in other device alarm messages that cannot be processed in time, and even causing serious security incidents.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present invention provide a warning message processing method, device and electronic device to at least solve the technical problem in the related art that it is difficult to process the event storm in time when the event storm occurs.

[0005] According to one aspect of an embodiment of the present invention, there is provided a method for processing an alarm message, comprising: receiving a first alarm message; determining a first Internet Protocol IP address of the first alarm message; determining a collection quantity of second alarm messages within a predetermined time, wherein the second alarm message is an alarm message with the same first IP address as the first alarm message; when the collection quantity of the second alarm message exceeds a predetermined threshold, ignoring the alarm message sent by the first terminal corresponding to the second IP address, wherein the second IP address is the IP address of the second alarm message; and sending the second IP address to the second terminal to perform fault handling on the first terminal corresponding to the second IP address.

[0006] Optionally, it also includes: storing the received first warning message, and determining parameter information of the first warning message, wherein the parameter information includes at least one of the following: classification information of the first warning message, semantic information of the first warning message, and manufacturer information of the first warning message.

[0007] Optionally, the first warning message is input into a classifier for classification to obtain classification information of the first warning message, wherein the classifier includes: a word embedding module, a bidirectional long short word memory Bi-LSTM module, and a conditional random field CRF module.

[0008] Optionally, the step of inputting the first alarm message into a classifier for classification to obtain classification information of the first alarm message includes: inputting the first alarm message into a classifier to obtain matrix parameters corresponding to the first alarm message, and feature parameters corresponding to the first alarm message; and obtaining classification information of the first alarm message based on the matrix parameters and the feature parameters.

[0009] Optionally, when the parameter information of the first alarm message includes classification information of the first alarm message, it includes: determining the classification of the first alarm message and the number of alarm messages included in each classification based on the classification information of the first alarm message; determining the cause of the fault based on the classification and the number of alarm messages included in each classification.

[0010] Optionally, ignoring the warning message sent by the first terminal corresponding to the second IP address includes: ignoring the warning message sent by the first terminal corresponding to the second IP address; deleting a third warning message, wherein the third warning message is a warning message in which the first IP address is the second IP address in the stored first warning messages.

[0011] According to one aspect of an embodiment of the present invention, there is provided an alarm message processing device, comprising: a receiving module for receiving a first alarm message; a first determining module for determining a first Internet Protocol IP address of the first alarm message; a second determining module for determining the number of second alarm messages collected within a predetermined time, wherein the second alarm message is an alarm message with the same first IP address as the first alarm message; an ignoring module for ignoring the alarm message sent by a first terminal corresponding to a second IP address when the number of second alarm messages collected exceeds a predetermined threshold, wherein the second IP address is the IP address of the second alarm message; and a sending module for sending the second IP address to a second terminal so as to perform fault handling on the first terminal corresponding to the second IP address.

[0012] According to one aspect of an embodiment of the present invention, there is provided an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement any one of the above-mentioned alarm message processing methods.

[0013] According to one aspect of an embodiment of the present invention, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute any of the above-mentioned alarm message processing methods.

[0014] According to one aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for processing warning messages.

[0015] In an embodiment of the present invention, a first alarm message is received, and a first Internet Protocol IP address of the first alarm message is determined. During the receiving process, the number of alarm messages collected with the same IP address within a predetermined time is determined, that is, the number of second alarm messages collected with the IP address being the second IP address is determined. When the number of second alarm messages collected exceeds a predetermined threshold, it is considered that too many alarm messages are sent by the first terminal corresponding to the second IP address, resulting in an event storm. Therefore, the alarm message sent by the first terminal corresponding to the second IP address is ignored, so as to achieve the purpose of not receiving the alarm message sent to the terminal corresponding to the second IP address, thereby avoiding the phenomenon of system paralysis caused by the event storm, and sending the second IP address to the second terminal to perform fault processing on the first terminal corresponding to the second IP address, so that the first terminal no longer sends too many alarm messages, thereby solving the technical problem of difficulty in timely processing of event storms in related technologies when event storms occur. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0017] Figure 1 is a flowchart of a method for processing an alarm message according to an embodiment of the present invention;

[0018] Figure 2 is a flow chart of an event blocking method provided by an optional embodiment of the present invention;

[0019] Figure 3 is a flow chart of an event storm blocking module provided by an optional implementation manner of the present invention;

[0020] Figure 4 It is a structural block diagram of an alarm message processing device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following explanations:

[0024] Event storm: A device sends a large number of warning messages in a short period of time. For example, if a device fails, millions of warning messages will be sent within 1 second.

[0025] Filebeat component: A lightweight log collector implemented in Golang that can read logs at corresponding locations according to the configuration and report them to the corresponding locations.

[0026] CRF: Conditional Random Field, conditional random field, is often used in sequence labeling, including part-of-speech tagging, word segmentation, named entity recognition and other fields.

[0027] BiLSTM: BILSTM-Attention, attention–bidirectional long short-term memory, long short-term memory network.

[0028] Word embedding: word embedding module, word embedding consists of three different parts: 1) Glove word vector, 2) character embedding, 3) syntactic features based on one-hot encoding. It can capture the relationship between words more accurately. Character embedding can effectively solve the problem of out-of-vocabulary (OOV) and can provide effective help for texts with strong professionalism. Syntactic features based on one-hot encoding can obtain more grammatical information, which helps to query professional terms more effectively.

[0029] Example 1

[0030] According to an embodiment of the present invention, an embodiment of a method for processing an alarm message is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0031] Figure 1 FIG. 1 is a flowchart of a method for processing an alarm message according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0032] Step S102, receiving a first warning message;

[0033] Step S104, determining a first Internet Protocol IP address of the first warning message;

[0034] Step S106, determining the number of second alarm messages collected within a predetermined time, wherein the second alarm message is an alarm message having the same first IP address as the first alarm message;

[0035] Step S108, when the number of collected second alarm messages exceeds a predetermined threshold, ignoring the alarm message sent by the first terminal corresponding to the second IP address, wherein the second IP address is the IP address of the second alarm message;

[0036] Step S110: sending the second IP address to the second terminal to perform fault processing on the first terminal corresponding to the second IP address.

[0037] Through the above steps, the first alarm message is received, and the first Internet Protocol IP address of the first alarm message is determined. During the receiving process, the number of alarm messages collected with the same IP address within a predetermined time is determined, that is, the number of second alarm messages collected with the IP address being the second IP address is determined. When the number of second alarm messages collected exceeds the predetermined threshold, it is considered that the first terminal corresponding to the second IP address has sent too many alarm messages, resulting in an event storm. Therefore, the alarm message sent by the first terminal corresponding to the second IP address is ignored, so as to achieve the purpose of not receiving the alarm message sent to the terminal corresponding to the second IP address, thereby avoiding the phenomenon of system paralysis caused by the event storm, and sending the second IP address to the second terminal to perform fault processing on the first terminal corresponding to the second IP address, so that the first terminal no longer sends too many alarm messages, thereby solving the technical problem of difficulty in timely processing of event storms in related technologies.

[0038] As an optional embodiment, a first warning message is received, and a first Internet Protocol IP address of the first warning message is determined, wherein the first warning message may be a warning message containing warning information sent by multiple terminal devices, and by determining the first Internet Protocol IP address of the first warning message, it is possible to know which terminal device sent the warning message. By receiving the first warning message, the alarm conditions of multiple terminal devices can be known, so that based on the first warning message, the alarms of the multiple terminal devices can be processed in a timely manner.

[0039] As an optional embodiment, the number of collected alarm messages with the same first IP address in the first alarm message within a predetermined time is determined, that is, the second alarm message. When the number of collected second alarm messages exceeds a predetermined threshold, it is considered that a large number of alarm messages have been sent by the terminal device corresponding to the IP address of the second alarm message in a short period of time. If a large number of alarm messages sent by the terminal device corresponding to the IP address of the second alarm message are continuously received, it may cause system memory to explode, the system to freeze, or even paralyze. In order to avoid this phenomenon, the alarm messages sent by the terminal device corresponding to the IP address of the second alarm message can be ignored at this time, and a large number of alarm message information is not received directly, which directly blocks the failure impact of the event storm on the system. It should be noted that, in this case, the alarm message in which the first IP address is the second IP address in the stored first alarm message can also be deleted, because the second IP address is subsequently sent directly to the terminal used by the operation and maintenance personnel for processing, that is, the second IP address can be sent to the second terminal, so that the operation and maintenance personnel using the second terminal can perform troubleshooting on the terminal device corresponding to the second IP address. Therefore, there is no need to store the alarm message sent by the terminal device corresponding to the second IP address for further analysis and processing, which not only saves storage space but also reduces operational pressure.

[0040] As an optional embodiment, the received first alarm message can also be stored. After storing the first alarm message, various parameter information of the first alarm message can be determined, such as: classification information of the first alarm message, semantic information of the first alarm message, and manufacturer information of the first alarm message. That is, it can be determined what type of alarm the first alarm message is, what the alarm information expressed by the first alarm message is, which manufacturer the IP device corresponding to the first alarm message belongs to, etc. By expanding the information of the alarm message, more abundant information about the alarm message can be obtained to better handle the alarm.

[0041] As an optional embodiment, the first warning message is input into a classifier for classification to obtain classification information of the first warning message, wherein the classifier includes: a word embedding module, a bidirectional long and short word memory Bi-LSTM module, and a conditional random field CRF module. The first warning message can be input into the classifier, the matrix parameters corresponding to the first warning message can be obtained through the word embedding module in the classifier, the feature parameters corresponding to the first warning message can be obtained through the bidirectional long and short word memory Bi-LSTM module in the classifier, and the conditional random field CRF module, and then the classification information of the first warning message can be obtained more effectively and accurately based on the matrix parameters and the feature parameters.

[0042] As an optional embodiment, when the parameter information of the first warning message includes classification information of the first warning message, the classification of the first warning message and the number of warning messages included in each classification can be determined based on the classification information of the first warning message; and the cause of the fault can be determined based on the classification and the number of warning messages included in each classification, so that the fault processing can be performed more purposefully.

[0043] Based on the above embodiments and optional embodiments, an optional implementation is provided, which is described in detail below.

[0044] The alarm message monitoring solution in the related technology is difficult to solve the problem of a large number of events appearing on the monitoring management platform page at the same time. Once this problem occurs, the monitoring page may be paralyzed and cannot be closed and maintained manually in time, which will hinder the normal alarm messages of other devices and cause serious security incidents.

[0045] In view of this, an optional implementation of the present invention provides an event blocking method. The purpose is to block and process event storms when they occur, and locate the source of the fault so that equipment problems can be repaired in a timely manner and the normal operation of the alarm message monitoring system can be quickly restored. It should be noted that event storms can be divided into the following types: (1) CPU utilization is too high; (2) memory utilization is too high; (3) port traffic is high; (4) power failure; (5) board failure. When encountering the above faults, a large number of events will appear on the monitoring management platform page at the same time.

[0046] Figure 2 is a flow chart of an event blocking method provided by an optional embodiment of the present invention, such as Figure 2 As shown, the following is a detailed description of the implementation mode of the present invention:

[0047] Step S1, alerting the alarm message collection module.

[0048] The alarm messages of the equipment are sent to the designated server through the network protocol. The alarm message collection module is responsible for collecting these alarm messages into a predetermined file for storage in advance so as to be used in subsequent fault research. The predetermined file can be set independently.

[0049] Step S2, alerting the message backup module.

[0050] The alarm messages collected in step S1 are stored in a specified database in real time. The database is secondary developed based on the component service filebeat, and can realize the filtering and screening functions of the alarm messages.

[0051] Step S3, determining whether the event is an event storm, if it is an event storm, entering the event storm blocking module, and blocking all warning messages corresponding to the event storm.

[0052] Step S4: If it is determined that the event is not an event storm, or the event storm has been recovered, then the alarm message enrichment module is entered.

[0053] For example, the device's warning message is used for word meaning recognition, the device is identified as belonging to a certain manufacturer, and the warning message is classified. In order to facilitate the subsequent analysis of the source of the warning message, this module uses a classifier to classify the warning message and store the classified warning message in a specified database. The classifier includes a word embedding module, a bidirectional long and short word memory Bi-LSTM module, and a conditional random field CRF module to extract and classify warning message features, classify and normalize different types of warning messages, and store the processed warning messages in a specified database in real time.

[0054] This is to solve the problem in related technologies that the alarm messages for the same type of fault from devices of different manufacturers are not quite the same in content structure, but have similarities. Manual classification is too time-consuming and time-consuming, and the characteristics of different types of alarm messages such as switches, routers, and firewalls are different, resulting in unclear classification.

[0055] The classifier in this module can use the TextCNN model to classify warning messages, and can use neural networks to learn the features of different types of warning messages, thereby classifying the warning messages. It should be noted that the TextCNN model used in this unit is different from the basic TextCNN model. In order to improve the efficiency of feature processing, the optional implementation method of the present invention optimizes the convolution layer in TextCNN to a depth-separable convolution layer, and combines it with the MobileNet model to form a network model with fewer parameters and more lightweight. This improved model can improve the accuracy of warning message information classification to a higher degree than the basic TextCNN model. It can more accurately extract the keywords and central ideas of the warning message.

[0056] The general process of implementing this model is as follows: In the first step, the event storm warning message is converted into a two-dimensional matrix as the model input through the word embedding operation; in the second step, the meaning of the warning message and the relevance of the context can be taken into account through the ordinary standard convolution operation; in the third step, the deep feature extraction is performed through the stacked depth-separable convolution operation, and the activation function after each layer of convolution is ReLU; in the fourth step, the pooling operation uses k-max pooling to only retain the features with the top-k scores in the feature values ​​after the convolution operation, so that more feature information can be retained; finally, a fully connected layer is used, followed by a softmax activation function. The softmax activation function can obtain the probability that the warning message belongs to a certain category, and the category with the largest probability is used as the classification result of this warning message.

[0057] It should be noted that the above classifier can also be used to process the warning messages corresponding to the event storm. When used for the warning messages corresponding to the event storm, the warning messages corresponding to the event storm can be stored first.

[0058] Step S5, alert the module to send the message.

[0059] Upload normal alarm messages. Before uploading, analyze the classified alarm messages in the database to find out how many types of alarm messages there are and the number of alarm messages of each type. Find out the most likely cause of equipment failure, how many types of alarm messages there are, and the number of alarm messages of each type. Then, analyze the cause of the failure more accurately.

[0060] It should be noted that the above analysis of how many types of warning messages there are and the number of warning messages of each type can also be used for the warning messages corresponding to event storms. When used for the warning messages corresponding to event storms, the warning messages corresponding to event storms can be stored first.

[0061] Figure 3 is a flow chart of an event storm blocking module provided by an optional embodiment of the present invention, such as Figure 3 As shown, the following is a detailed introduction to the event storm blocking module:

[0062] The implementation process of the event storm blocking device S103 in the present invention is as follows: Figure 3 As shown:

[0063] Step S3.1, event storm identification unit.

[0064] The backup data in the database within these 10 seconds is traversed every 10 seconds. If the number of warning messages sent by a device IP exceeds the set threshold, the warning messages sent by this device are considered to be event storms.

[0065] Step S3.2, event storm blocking unit.

[0066] This unit is designed with front-end and back-end control. When a warning message sent by a device is identified as an event storm, the IP of this device is added to the filter of the event storm blocking module. The blocking module will delete all warning message information sent by this IP device from the database and ignore all warning message information sent by this IP device. All warning messages sent by this device will no longer be enriched or sent up. This unit can also send the IP to a predetermined terminal so that the person using the predetermined terminal can handle the IP device to solve the fault.

[0067] When the event storm is resolved, that is, the IP device fault is handled, the back-end controller will delete the device IP from the filter, and the alarm message of this IP can be stored in the database normally through the alarm message data backup module, and the scheduled terminal will also synchronously display that the event storm has been resolved.

[0068] Through the above optional implementation, at least the following beneficial effects can be achieved:

[0069] (1) Respond promptly and clear event storms in the alarm message monitoring system to prevent memory overload and reduce the risk of system paralysis.

[0070] (2) Ensure that normal warning messages are sent to the monitoring platform and processed promptly.

[0071] (3) Alarm messages are classified and stored to facilitate operation and maintenance personnel to find problems in a timely manner, improve operation and maintenance efficiency, and reduce operation and maintenance costs.

[0072] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0073] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of various embodiments of the present invention.

[0074] Example 2

[0075] According to an embodiment of the present invention, a device for implementing the above-mentioned warning message processing method is also provided. Figure 4 is a structural block diagram of an alarm message processing device according to an embodiment of the present invention. Figure 4 As shown, the device includes: a receiving module 402, a first determining module 404, a second determining module 406, an ignoring module 408 and a sending module 410. The device is described in detail below.

[0076] A receiving module 402 is used to receive a first alarm message; a first determining module 404 is connected to the above-mentioned receiving module 402, and is used to determine a first Internet Protocol IP address of the first alarm message; a second determining module 406 is connected to the above-mentioned first determining module 404, and is used to determine the number of second alarm messages collected within a predetermined time, wherein the second alarm message is an alarm message with the same first IP address as the first alarm message; an ignoring module 408 is connected to the above-mentioned second determining module 406, and is used to ignore the alarm message sent by the first terminal corresponding to the second IP address when the number of second alarm messages collected exceeds a predetermined threshold, wherein the second IP address is the IP address of the second alarm message; a sending module 410 is connected to the above-mentioned ignoring module 408, and is used to send the second IP address to the second terminal to perform fault handling on the first terminal corresponding to the second IP address.

[0077] It should be noted here that the above-mentioned receiving module 402, first determination module 404, second determination module 406, ignoring module 408 and sending module 410 correspond to steps S102 to S110 in the above-mentioned warning message processing method, and the instances and application scenarios implemented by multiple modules and corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned embodiment 1.

[0078] Example 3

[0079] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor, wherein the processor is configured to execute the instructions to implement any of the above-mentioned alarm message processing methods.

[0080] Example 4

[0081] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute any of the above-mentioned alarm message processing methods.

[0082] Example 5

[0083] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for processing alarm messages.

[0084] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0085] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0086] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0087] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0088] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0089] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0090] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for processing an alarm message, characterized in that: include: Receiving the first warning message; Determining a first Internet Protocol (IP) address of the first warning message; Determine the number of second alarm messages collected within a predetermined time, wherein the second alarm message is an alarm message in the first alarm message having the same IP address as the first alarm message; When the number of collected second alarm messages exceeds a predetermined threshold, ignoring the alarm message sent by the first terminal corresponding to the second IP address, wherein the second IP address is the IP address of the second alarm message; sending the second IP address to the second terminal to perform fault processing on the first terminal corresponding to the second IP address; Deleting a third alarm message, wherein the third alarm message is an alarm message in which the first IP address is the second IP address in the stored first alarm message; The first warning message is input into a classifier for classification. In the classifier, the following steps are performed: in the first step, the first warning message is converted into a two-dimensional matrix through a word embedding module word embedding operation; in the second step, a standard convolution operation is performed on the two-dimensional matrix to consider the meaning of the first warning message and the relevance of the context; in the third step, a depth-separable convolution operation is performed to extract features and obtain feature values; in the fourth step, a pooling operation is performed to retain the features with top-k scores in the feature values ​​after the convolution operation; in the fifth step, a fully connected layer and a softmax activation function are used for processing to obtain the probability that the warning message belongs to a certain category, and the category with the largest probability is used as the classification result of the warning message, wherein the activation function after each layer of convolution is ReLU, and the pooling operation is an operation of using k-max pooling to only retain the features with top-k scores in the feature values ​​after the convolution operation.

2. The method according to claim 1, further comprising: The received first warning message is stored, and parameter information of the first warning message is determined, wherein the parameter information includes at least one of the following: classification information of the first warning message, semantic information of the first warning message, and manufacturer information of the first warning message.

3. The method according to claim 2, comprising: The first warning message is input into a classifier for classification to obtain classification information of the first warning message, wherein the classifier includes: a word embedding module, a bidirectional long short word memory Bi-LSTM module, and a conditional random field CRF module.

4. The method according to claim 3, wherein the step of inputting the first warning message into a classifier for classification to obtain classification information of the first warning message comprises: Inputting the first warning message into a classifier to obtain a matrix parameter corresponding to the first warning message and a feature parameter corresponding to the first warning message; According to the matrix parameters and the characteristic parameters, classification information of the first warning message is obtained.

5. The method according to claim 2, wherein, when the parameter information of the first warning message includes classification information of the first warning message, the method further comprises: Determining, according to the classification information of the first warning message, the classification of the first warning message and the number of warning messages included in each classification; The cause of the fault is determined based on the categories and the number of warning messages included in each category.

6. The method according to claim 2, ignoring the warning message sent by the first terminal corresponding to the second IP address, comprises: Ignore the warning message sent by the first terminal corresponding to the second IP address; A third alarm message is deleted, wherein the third alarm message is a stored alarm message in which the first IP address is the second IP address.

7. A warning message processing device, characterized in that: include: A receiving module, used for receiving a first warning message; A first determining module, configured to determine a first Internet Protocol IP address of the first warning message; A second determination module is used to determine the number of second alarm messages collected within a predetermined time, wherein the second alarm message is an alarm message in the first alarm message having the same IP address as the first alarm message; an ignoring module, configured to ignore the alarm message sent by the first terminal corresponding to the second IP address when the number of collected second alarm messages exceeds a predetermined threshold, wherein the second IP address is the IP address of the second alarm message; A sending module, used for sending the second IP address to the second terminal, so as to perform fault processing on the first terminal corresponding to the second IP address; The sending module is further configured to delete a third warning message, wherein the third warning message is a warning message in which the first IP address is the second IP address in the stored first warning message; The sending module is also used to input the first warning message into a classifier for classification, and in the classifier, the following steps are performed: the first step is to convert the first warning message into a two-dimensional matrix through a word embedding module wordembedding operation; the second step is to perform a standard convolution operation on the two-dimensional matrix to consider the meaning of the first warning message and the relevance of the context; the third step is to perform a deep separable convolution operation to extract features and obtain feature values; the fourth step is to perform a pooling operation to retain the features with top-k scores in the feature values ​​after the convolution operation; the fifth step is to use a fully connected layer and a softmax activation function to obtain the probability that the warning message belongs to a certain category, and the category with the largest probability is used as the classification result of the warning message, wherein the activation function after each layer of convolution is ReLU, and the pooling operation is an operation that uses k-maxpooling to only retain the features with top-k scores in the feature values ​​after the convolution operation.

8. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the warning message processing method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the alarm message processing method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for processing an alarm message according to any one of claims 1 to 6 is implemented.