Intrusion detection method and device for container, medium, equipment and program product

By building a container behavior baseline model and abnormal detection of container system behavior events is solved, the problem of high intrusion detection resource occupancy in the existing technology is solved, efficiency and performance are improved, and costs are reduced.

CN118940259APending Publication Date: 2024-11-12BEIJING VOLCANO ENGINE TECH CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411155979.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

When existing container intrusion detection methods increase rules, the resource occupancy rate increases linearly, resulting in a decrease in detection performance and efficiency, and it is impossible to effectively deal with massive real-time behavioral data.

Method used

By obtaining the target image of container system behavior events, a behavior baseline model based on the target image is built, abnormal detection is performed, and only intrusion detection is performed on abnormal events that deviate from behavior is performed, reducing the amount of data and improving efficiency.

Benefits of technology

It reduces the resource occupancy rate of the intrusion detection process, improves the intrusion detection efficiency and performance in the container environment, reduces the cost and complexity of security analysis, and realizes rapid abnormality detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940259A_ABST
    Figure CN118940259A_ABST
Patent Text Reader

Abstract

The invention relates to an intrusion detection method and device for a container, a medium, equipment and a program product. The method comprises the steps of determining a target mirror image corresponding to a first container to which a first container system behavior event belongs; determining whether the first container system behavior event is an abnormal event or not by using a first behavior baseline model corresponding to the target mirror image; and if the first container system behavior event is an abnormal event, performing intrusion detection on the first container system behavior event. Therefore, the first behavior baseline model can be constructed by using the information of the container system behavior events of one or more containers created based on the target mirror image by utilizing the characteristic that the container created by a single mirror image has single and stable business behavior, and then the first behavior baseline model based on the mirror image dimension is obtained; according to the method, anomaly detection is carried out on massive real-time behavior data, then intrusion detection is carried out only for abnormal events with a small proportion, and the data volume of intrusion detection is reduced, so that the resource occupancy rate of intrusion detection is reduced, and the intrusion detection efficiency and performance are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of computer and network communication technology, and in particular, to an intrusion detection method, device, medium, equipment and program product for a container. Background Art

[0002] In terminal security products, intrusion detection is a key core technology. It is also important for intrusion detection in container environments. It can detect and prevent malicious network intrusions in a timely manner and ensure the security of the container environment. At present, rule matching is usually used to perform intrusion detection on the continuously generated event streams. However, as the number of rules increases, the resource usage of the intrusion detection process increases linearly, reducing the performance and efficiency of intrusion detection. Summary of the invention

[0003] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0004] In a first aspect, the present disclosure provides an intrusion detection method for a container, comprising: Acquire a first container system behavior event, and determine a target image corresponding to a first container to which the first container system behavior event belongs, wherein the first container is created based on the target image; Determine whether the first container system behavior event is an abnormal event by using a first behavior baseline model corresponding to the target image; wherein the first behavior baseline model is obtained by training based on a target data set, and the target data set includes information on container system behavior events of one or more containers created based on the target image; If the first container system behavior event is an abnormal event, the first container system behavior event is detected to determine whether the first container system behavior event is an intrusion event.

[0005] In a second aspect, the present disclosure provides an intrusion detection device for a container, comprising: A first determination module is configured to obtain a first container system behavior event and determine a target image corresponding to a first container to which the first container system behavior event belongs, wherein the first container is created based on the target image; A second determination module is configured to determine whether the first container system behavior event is an abnormal event by using a first behavior baseline model corresponding to the target image; wherein the first behavior baseline model is obtained by training based on a target data set, and the target data set includes information on container system behavior events of one or more containers created based on the target image; An intrusion detection module is used to perform intrusion detection on the first container system behavior event if the first container system behavior event is an abnormal event, so as to determine whether the first container system behavior event is an intrusion event.

[0006] In a third aspect, the present disclosure provides a computer-readable medium having a computer program stored thereon, which, when executed by a processing device, implements the steps of the intrusion detection method for a container provided in the first aspect of the present disclosure.

[0007] In a fourth aspect, the present disclosure provides an electronic device, including: a storage device having a computer program stored thereon; A processing device is used to execute the computer program in the storage device to implement the steps of the intrusion detection method for a container provided in the first aspect of the present disclosure.

[0008] In a fifth aspect, the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the intrusion detection method for a container provided in the first aspect of the present disclosure.

[0009] In the above technical solution, for the first container system behavior event, first determine the target image corresponding to the first container to which the first container system behavior event belongs; then, use the first behavior baseline model corresponding to the target image to determine whether the first container system behavior event is an abnormal event, wherein the first behavior baseline model is trained based on the information of the container system behavior event of one or more containers created by the target image; when the first container system behavior event is an abnormal event, perform intrusion detection on the first container system behavior event to determine whether the first container system behavior event is an intrusion event. In this way, the container created by a single image can be used to have the characteristics of single and stable business behavior, and the first behavior baseline model can be constructed using the information of the container system behavior event of one or more containers created based on the target image, and then the first behavior baseline model of the image dimension can be used to perform anomaly detection on massive real-time behavior data to obtain abnormal events that deviate from the behavior baseline, and then only perform intrusion detection on these abnormal events with a small proportion, thereby reducing the amount of data for intrusion detection, thereby reducing the resource occupancy rate of the intrusion detection process, improving the efficiency and performance of intrusion detection in the container environment, and effectively reducing the cost and complexity of subsequent security analysis. In addition, all containers created based on the same target image can use the same first behavior baseline model for anomaly detection, without having to build a corresponding first behavior baseline model for each container separately, avoiding the cost of developing and maintaining a first behavior baseline model for each container separately, and reducing development and operation and maintenance costs. In addition, by using a unified first behavior baseline model, real-time anomaly detection can be quickly performed on all behavioral events in containers created based on the same target image, improving the speed of anomaly detection.

[0010] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings: Figure 1 The figure is a diagram showing an implementation architecture of an intrusion detection system for a container according to an exemplary embodiment.

[0012] Figure 2 The present invention is a flowchart of a method for intrusion detection of a container according to an exemplary embodiment.

[0013] Figure 3 The diagram is a schematic diagram showing an intrusion detection process for a container according to an exemplary embodiment.

[0014] Figure 4 It is a schematic diagram of a core processing flow of a feature extractor according to an exemplary embodiment.

[0015] Figure 5 The figure is a flowchart of a method for constructing a first behavior baseline model according to an exemplary embodiment.

[0016] Figure 6 The diagram is a schematic diagram showing a first-line baseline model building process according to an exemplary embodiment.

[0017] Fig. 7A It is a schematic diagram of traceability analysis in related technologies.

[0018] Figure 7B It is a schematic diagram of a source tracing analysis according to an exemplary embodiment.

[0019] Figure 8 The invention is a block diagram showing an intrusion detection device for a container according to an exemplary embodiment.

[0020] Fig. 9 It is a block diagram of a first behavior baseline model building device according to an exemplary embodiment.

[0021] Fig.10 The diagram is a schematic structural diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0022] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0023] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0024] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0025] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0026] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0027] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0028] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0029] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0030] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0031] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0032] At the same time, it is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0033] Before introducing the specific embodiments of the present disclosure, the relevant terms and core ideas of the present disclosure are first explained.

[0034] An image is a read-only file system template that contains all the code, runtime libraries, environment variables, and configuration files needed to run a container.

[0035] A container is a running instance created based on an image using operating system layer virtualization technology. It has an independent file system, network, and process space, allowing applications to run in an isolated environment.

[0036] A container environment is a lightweight, portable, self-sufficient software runtime environment that allows developers to package applications and their dependencies together to run seamlessly on any platform that supports container technology.

[0037] The immutability of an image means that its content cannot be changed after it is created.

[0038] The traditional intrusion detection engine is the intrusion detection engine commonly used at this stage. This intrusion detection engine can use rule matching to perform intrusion detection on the continuously generated event stream, or it can use a pre-built intrusion detection model (for example, convolutional neural network, long short-term memory network, etc.) to perform intrusion detection on the event stream.

[0039] The host is a variable asset, and generally multiple application services are deployed. The impact of multiple applications on the host may interfere with each other, which leads to highly complex and changeable system behavior, which brings many troubles to intrusion detection. In the container environment, the image is immutable, and the container created based on a single image has the characteristics of single and stable business behavior, that is, singleness and stability. Therefore, in the container environment, the present disclosure uses the immutable asset characteristics of the container image and the singleness and stability characteristics of the container to propose an efficient intrusion detection method. The core idea of ​​the method is: on the basis of traditional intrusion detection technology, it is further proposed to combine the behavior events of the container with the container information and the image information, and to perform behavior modeling on the behavior events of the container to form the first behavior baseline model of the container, wherein the first behavior baseline model can reflect the behavior characteristics of the container under normal circumstances; in the real-time detection stage, the behavior events in most scenarios are normal events. Through the first behavior baseline model detection, only a small number of abnormal events that deviate from the model are detected for intrusion detection, thereby greatly reducing the amount of data for intrusion detection and improving the efficiency of intrusion detection. At the same time, only tracing the threat source of abnormal events is also more conducive to the analysis efficiency of threat tracing.

[0040] Specifically, if Figure 1 As shown, the implementation process mainly includes the model building stage and the real-time detection stage.

[0041] like Figure 1As shown, in the model building phase, the historical behavior events of one or more containers (called homologous containers) created based on the target image are first detected by the traditional intrusion detection engine to obtain historical normal events, and then passed to the feature extractor, where the homologous container includes the container created in the host based on the target image; the feature extractor designs the features of different behavior events based on expert experience to form a feature set, and converts the behavior events into feature vectors through a specific algorithm, and then these feature vectors are learned and trained by the image model learner, and finally the first behavior baseline model of the homologous container is obtained, that is, the first behavior baseline model corresponding to the target image.

[0042] In the real-time detection stage, Figure 1 As shown, intrusion detection can be performed on the first container system behavior event of the target container (i.e., the real-time container system behavior event) in the following manner, wherein the target container is any container of the above homologous containers: First, the feature extractor extracts features of the first container system behavior event of the target container to obtain a feature vector of the event; Then, the detection accelerator will determine whether the first container system behavior event is an abnormal event based on the previously constructed first behavior baseline model corresponding to the target image and the feature vector of the first container system behavior event. In this process, all normal events will be filtered out, and only abnormal events that are determined to deviate from the behavior baseline will enter the subsequent detection stage; Next, the abnormal event will be transmitted to the abnormal flow collector and the traditional intrusion detection engine. The abnormal flow collector is responsible for storing abnormal events and providing data support for subsequent threat tracing; the traditional intrusion detection engine uses traditional logic to perform intrusion detection on the first container system behavior event. If the first container system behavior event is detected as an intrusion event, an intrusion alarm will be issued.

[0043] After detecting an intrusion alarm, the traceability enhancer will combine the abnormal events in the abnormal traffic collector (i.e. historical abnormal events) to conduct a comprehensive analysis and traceability work on the corresponding assets.

[0044] Among them, the detection accelerator acts in the real-time detection stage. It receives the feature flow of the first container system behavior events from the feature extractor, and then makes full use of the first behavior baseline model that has been built to accurately judge the first container system behavior events, clean up normal events, and only retain a small proportion of abnormal events that deviate from the baseline, and enter the subsequent traditional intrusion detection engine, which greatly reduces the amount of detection data and is the core of the entire efficiency improvement work.

[0045] Figure 2 FIG. 1 is a flow chart of an intrusion detection method for a container according to an exemplary embodiment. Figure 2As shown, the intrusion detection method for a container may include S101 to S103.

[0046] In S101, a first container system behavior event is obtained, and a target image corresponding to a first container to which the first container system behavior event belongs is determined.

[0047] In the present disclosure, the target image used to create the first container can be determined by parsing the container information of the first container to which the first container system behavior event belongs, that is, the first container is created based on the target image. The container information includes the image information used to create the current container.

[0048] In S102, a first behavior baseline model corresponding to the target image is used to determine whether the first container system behavior event is an abnormal event.

[0049] In the present disclosure, after identifying the above-mentioned target image, the detection accelerator can determine whether there is a first behavior baseline model corresponding to the target image. Since in a container environment, the image is immutable, and the container created based on a single image has the characteristics of single and stable business behavior, the first behavior baseline model can be obtained based on the target data set training, wherein the target data set includes information on container system behavior events of one or more containers created based on the target image. Specifically, the information on container system behavior events of one or more containers created based on the target image can include reference feature information of historical normal events of the corresponding containers to construct a first behavior baseline model, wherein the one or more containers created based on the target image belong to the host of the first container system behavior event, that is, the homologous container includes the container created based on the target image in the host of the first container system behavior event.

[0050] like Figure 3 As shown, if there is a first behavior baseline model corresponding to the target image, the detection accelerator uses the first behavior baseline model corresponding to the target image through the model judger to detect whether the first container system behavior event deviates from the behavior baseline to determine whether the first container system behavior event is an abnormal event; if there is no first behavior baseline model corresponding to the target image, the first container system behavior event is directly determined as an abnormal event. Among them, events that deviate from the behavior baseline are abnormal events, and events that do not deviate from the behavior baseline are normal events.

[0051] In S103, if the first container system behavior event is an abnormal event, an intrusion detection is performed on the first container system behavior event to determine whether the first container system behavior event is an intrusion event.

[0052] like Figure 3As shown, if the first container system behavior event is an abnormal event, detection continues. At this time, traditional intrusion detection can be performed on the first container system behavior event to determine whether the first container system behavior event is an intrusion event. At the same time, the abnormal event is stored in the abnormal traffic collector; if the first container system behavior event is a normal event, subsequent detection is stopped.

[0053] In the above technical solution, for the first container system behavior event, first determine the target image corresponding to the first container to which the first container system behavior event belongs; then, use the first behavior baseline model corresponding to the target image to determine whether the first container system behavior event is an abnormal event, wherein the first behavior baseline model is trained based on the information of the container system behavior event of one or more containers created by the target image; when the first container system behavior event is an abnormal event, perform intrusion detection on the first container system behavior event to determine whether the first container system behavior event is an intrusion event. In this way, the container created by a single image can be used to have the characteristics of single and stable business behavior, and the first behavior baseline model can be constructed using the information of the container system behavior event of one or more containers created based on the target image, and then the first behavior baseline model of the image dimension can be used to perform anomaly detection on massive real-time behavior data to obtain abnormal events that deviate from the behavior baseline, and then only perform intrusion detection on these abnormal events with a small proportion, thereby reducing the amount of data for intrusion detection, thereby reducing the resource occupancy rate of the intrusion detection process, improving the efficiency and performance of intrusion detection in the container environment, and effectively reducing the cost and complexity of subsequent security analysis. In addition, all containers created based on the same target image can use the same first behavior baseline model for anomaly detection, without having to build a corresponding first behavior baseline model for each container separately, avoiding the cost of developing and maintaining a first behavior baseline model for each container separately, and reducing development and operation and maintenance costs. In addition, by using a unified first behavior baseline model, real-time anomaly detection can be quickly performed on all behavioral events in containers created based on the same target image, improving the speed of anomaly detection.

[0054] The following is a detailed description of the specific implementation method of determining whether the first container system behavior event is an abnormal event by using the first behavior baseline model corresponding to the target image in the above S102. Specifically, it can be achieved by the following steps (a1) to (a4): Step (a1): extracting features of the first container system behavior event to obtain target feature information.

[0055] Step (a2): using the first line corresponding to the target image as a baseline model, determining whether the target feature information deviates from the reference feature information.

[0056] Step (a3): If the target characteristic information deviates from the reference characteristic information, the first container system behavior event is determined to be an abnormal event.

[0057] Step (a4): If the target characteristic information does not deviate from the reference characteristic information, the first container system behavior event is determined to be a normal event.

[0058] The following describes in detail the specific implementation method of extracting features from the first container system behavior event in the above step (a1) to obtain target feature information.

[0059] Specifically, it can be implemented through various implementations. In one implementation, a pre-built feature extractor can be used to extract target feature information of the first container system behavior event.

[0060] In another embodiment, in order to improve the accuracy of feature extraction, different feature extractors can be used for different types of behavior events to perform feature extraction. Specifically, the target event type of the first container system behavior event can be determined first; then, the target feature information of the first container system behavior event can be extracted using the feature extractor corresponding to the target event type.

[0061] In the present disclosure, the feature extractor can design features of different types of behavioral events based on expert experience, convert each event information into a feature vector through a machine learning feature engineering algorithm, and provide feature extraction services for subsequent model construction and real-time detection.

[0062] like Figure 4 As shown, the types of behavior events may include process, network, and file, that is, the target event type is any one of process, network, and file, and the behavior events may be divided into process behavior events, network behavior events, and file behavior events. Among them, the process behavior event corresponds to a network data extractor, the process behavior event corresponds to a process data extractor, and the file behavior event corresponds to a file data extractor.

[0063] like Figure 4 As shown, after obtaining the first container system behavior event, the behavior event can be classified first, that is, the target event type of the first container system behavior event can be determined, and then the features of the first container system behavior event can be extracted by using a feature extractor corresponding to the target event type. Specifically, the attribute information of the first container system behavior event can be obtained first; then, the feature extraction of the attribute information can be performed, and finally, the extracted features can be encoded using feature engineering algorithms such as label encoding and unique hot encoding to obtain the corresponding feature vector.

[0064] For example, the first container system behavior event is a process behavior event, that is, the target event type is a process, wherein the process behavior event generally includes the following attribute information: process path, parent process path, command parameters, input and output types, associated container information, command line information, etc. Figure 4 As shown, the characteristics of process behavior events usually include: process path, parent process path, number of parameters, number of numeric parameters, number of special characters in parameters, character length of parameters, whether parameters exist in files, whether parameters exist in scripts, whether parameters exist in IP addresses, whether parameters exist in domain names, whether the input source is the network (i.e., whether the process behavior event is triggered through the network), whether the output source is the network, and other multi-dimensional characteristics.

[0065] Among them, when the target event type is process, the target feature information of the first container system behavior event may include one or more of the following: process path, parent process path, number of parameters, number of numeric parameters, number of special characters of parameters, character length of parameters, whether the parameter exists in a file, whether the parameter exists in a script, whether the parameter exists in an IP, whether the parameter exists in a domain name, whether the input source is a network, and whether the output source is a network.

[0066] The following is a detailed description of the specific implementation method of using the first behavior baseline model corresponding to the target image in the above step (a2) to determine whether the target feature information deviates from the reference feature information. Specifically, since the first behavior baseline model is used to determine whether the target feature information of the first container system behavior event deviates from the reference feature information of the historical normal event, the first behavior baseline model can be an outlier judgment model. At this time, the first behavior baseline model corresponding to the target image can be used to determine whether the target feature information is an outlier relative to the reference feature information; if the target feature information is an outlier relative to the reference feature information, it is determined that the target feature information deviates from the reference feature information; if the target feature information is not an outlier relative to the reference feature information, it is determined that the target feature information does not deviate from the reference feature information. Among them, the above-mentioned outlier judgment model can be, for example, an isolation forest model, a K-means clustering model, etc.

[0067] Optionally, the first behavior baseline model may be an isolation forest model, and the target feature information may be a target feature matrix. In this case, the step of using the first behavior baseline model corresponding to the target image to determine whether the target feature information is an outlier relative to the reference feature information may include: matching the target feature matrix with the isolation forest model to determine whether the target feature matrix is ​​an outlier.

[0068] In the present disclosure, the isolation forest model uses a binary tree to segment the data. The depth of the data point in the binary tree reflects the degree of "alienation" of the data, wherein the shallower the depth, the more likely it is an outlier. In this embodiment, the first line is a binary tree structure constructed by the baseline model including the feature matrix of historical normal events. The target feature matrix is ​​matched with the isolation forest model to determine the depth of the target feature matrix in the binary tree, thereby determining whether the target feature matrix is ​​an outlier.

[0069] Figure 5 FIG. 1 is a flowchart of a method for constructing a first line baseline model according to an exemplary embodiment. Figure 5 As shown, the first line baseline model corresponding to the target image can be constructed by following the steps S201 to S203: In S201, historical normal events of one or more containers created based on the target image are obtained.

[0070] In S202, feature extraction is performed on historical normal events to obtain reference feature information.

[0071] In the present disclosure, a method similar to the above step (a1) may be used to extract features from historical normal events to obtain reference feature information.

[0072] In S203, the initial model is trained using the reference feature information to obtain a first line as a baseline model.

[0073] In the present disclosure, when training the first behavior baseline model, only reference feature information of normal historical events is used, while abnormal samples are excluded. This can prevent the behavior features learned by the first behavior baseline model from including features of abnormal events, thereby ensuring the abnormal event detection accuracy of the first behavior baseline model.

[0074] In addition, the first behavior baseline model can be built through the image model learner. Generally speaking, containers are microservice instances, and containers created by the same image are highly similar in behavior. For example, although the MySQL cluster built by the MySQL image has multiple MySQL containers, the system behaviors of these containers are extremely similar and only provide MySQL business services, with a strong behavioral singleness. Therefore, the image model learner can consider building the first behavior baseline model for the container cluster instance from the image dimension.

[0075] like Figure 6 As shown, after the feature extractor extracts features of historical normal events, reference feature information can be obtained. After that, the mirror model learner uses the reference feature information to train the initial model to generate a model, and the first line is obtained as the baseline model.

[0076] For example, the initial model is an isolation forest model. At this time, unsupervised learning training can be performed using reference feature information to obtain the first line as a baseline model.

[0077] The following is a detailed description of the specific implementation method of obtaining the historical normal events of one or more containers created based on the target image in the above S201. Specifically, it can be achieved through the following steps (b1) to (b3): Step (b1): obtain multiple historical behavior events of one or more containers created based on the target image.

[0078] In the present disclosure, the first container system behavior events of all containers on the host can be collected in real time by an event collection device (e.g., an agent collector) deployed on the host of the first container system behavior event. In this way, the historical behavior events of each container built based on the target image collected in the recent period (e.g., the last month) can be obtained from the event collection device as the historical behavior events of one or more containers created based on the target image.

[0079] Step (b2): Perform intrusion detection on multiple historical behavior events to determine whether each historical behavior event is an intrusion event.

[0080] Step (b3): ​​remove historical behavior events that are intrusion events from multiple historical behavior events, and obtain historical normal events of one or more containers created based on the target image.

[0081] like Figure 1 As shown, a traditional intrusion detection engine can be used to perform traditional intrusion detection on multiple historical behavior events to exclude abnormal behavior events that belong to intrusion events and only retain normal behavior events for training the first behavior baseline model.

[0082] like Fig. 7A As shown, in the threat tracing stage, in the traditional mode, threat tracing analysis is usually performed on all events. In this way, there may be a large number of interferences from risk-free tracing nodes, which affects the accuracy and efficiency of threat tracing analysis. To this end, the present disclosure can only perform threat tracing analysis on abnormal events. Specifically, the above method can also include the following steps: If the first container system behavior event is an intrusion event, a threat tracing analysis is performed on historical abnormal events.

[0083] In this disclosure, Figure 1As shown, the source tracing enhancer acts in the real-time detection stage and is triggered by intrusion alarms. It organizes historical abnormal events in the abnormal traffic collector and performs overall behavior analysis to obtain a source tracing analysis report. The source tracing analysis report can include the source of the attack, the attack path, and the specific details of the attack, thereby providing defenders with more targeted protection and countermeasures.

[0084] like Figure 7B As shown in the figure, since the traceability enhancer only performs traceability analysis on behaviors that deviate from the baseline (i.e., abnormal events), it can avoid interference from a large number of risk-free traceability nodes (i.e., normal events), effectively enhance the anti-interference and readability of the traceability, and improve the accuracy and efficiency of threat traceability analysis.

[0085] Figure 8 FIG. 1 is a block diagram of an intrusion detection device for a container according to an exemplary embodiment. Figure 8 As shown, the intrusion detection device 300 for a container includes: A first determination module 301 is used to obtain a first container system behavior event and determine a target image corresponding to a first container to which the first container system behavior event belongs, wherein the first container is created based on the target image; A second determination module 302 is used to determine whether the first container system behavior event is an abnormal event by using a first behavior baseline model corresponding to the target image; wherein the first behavior baseline model is obtained by training based on a target data set, and the target data set includes information on container system behavior events of one or more containers created based on the target image; The intrusion detection module 303 is used to perform intrusion detection on the first container system behavior event if the first container system behavior event is an abnormal event, so as to determine whether the first container system behavior event is an intrusion event.

[0086] In the above technical solution, for the first container system behavior event, first determine the target image corresponding to the first container to which the first container system behavior event belongs; then, use the first behavior baseline model corresponding to the target image to determine whether the first container system behavior event is an abnormal event, wherein the first behavior baseline model is trained based on the information of the container system behavior event of one or more containers created by the target image; when the first container system behavior event is an abnormal event, perform intrusion detection on the first container system behavior event to determine whether the first container system behavior event is an intrusion event. In this way, the container created by a single image can be used to have the characteristics of single and stable business behavior, and the first behavior baseline model can be constructed using the information of the container system behavior event of one or more containers created based on the target image, and then the first behavior baseline model of the image dimension can be used to perform anomaly detection on massive real-time behavior data to obtain abnormal events that deviate from the behavior baseline, and then only perform intrusion detection on these abnormal events with a small proportion, thereby reducing the amount of data for intrusion detection, thereby reducing the resource occupancy rate of the intrusion detection process, improving the efficiency and performance of intrusion detection in the container environment, and effectively reducing the cost and complexity of subsequent security analysis. In addition, all containers created based on the same target image can use the same first behavior baseline model for anomaly detection, without having to build a corresponding first behavior baseline model for each container separately, avoiding the cost of developing and maintaining a first behavior baseline model for each container separately, and reducing development and operation and maintenance costs. In addition, by using a unified first behavior baseline model, real-time anomaly detection can be quickly performed on all behavioral events in containers created based on the same target image, improving the speed of anomaly detection.

[0087] Optionally, the information includes reference feature information of historical normal events of the corresponding container, and one or more containers created based on the target image belong to the host of the first container system behavior event.

[0088] Optionally, the second determining module 302 includes: A first feature extraction submodule is used to extract features of the first container system behavior event to obtain target feature information; A judgment submodule, configured to judge whether the target feature information deviates from the reference feature information by using a first behavior corresponding to the target image as a baseline model; The first determination submodule is configured to determine that the first container system behavior event is an abnormal event if the target feature information deviates from the reference feature information.

[0089] Optionally, the first feature extraction submodule includes: A second determination submodule is used to determine a target event type of the first container system behavior event; The second feature extraction submodule is used to extract target feature information of the first container system behavior event by using a feature extractor corresponding to the target event type.

[0090] Optionally, the target event type is any one of process, network, and file.

[0091] Optionally, the target event type is a process, and the target feature information includes one or more of the following: Process path, parent process path, number of parameters, number of numeric parameters, number of special characters in parameters, character length of parameters, whether the parameter exists in a file, whether the parameter exists in a script, whether the parameter exists in an IP, whether the parameter exists in a domain name, whether the input source is a network, and whether the output source is a network.

[0092] Optionally, the first behavior baseline model is an outlier judgment model; The judgment submodule includes: A third determination submodule, configured to determine whether the target feature information is an outlier relative to the reference feature information by using a first behavior baseline model corresponding to the target image; The fourth determination submodule is configured to determine that the target feature information deviates from the reference feature information if the target feature information is an outlier relative to the reference feature information.

[0093] Optionally, the first behavior baseline model is an isolation forest model, and the target feature information is a target feature matrix; The third determination submodule is used to match the target feature matrix with the isolation forest model to determine whether the target feature matrix is ​​an outlier.

[0094] Optionally, the first behavior baseline model is constructed by a mirror model learner, such as Fig. 9 As shown, the mirror model learner 400 includes: An acquisition module 401 is used to acquire historical normal events of one or more containers created based on the target image; A feature extraction module 402 is used to extract features of the historical normal events to obtain the reference feature information; The training module 403 is used to train the initial model using the reference feature information to obtain the first behavior as a baseline model.

[0095] Optionally, the acquisition module 401 includes: An acquisition submodule, used to acquire multiple historical behavior events of one or more containers created based on the target image; A fifth determination submodule, configured to perform intrusion detection on the plurality of historical behavior events to determine whether each of the historical behavior events is an intrusion event; The elimination submodule is used to eliminate the historical behavior events belonging to intrusion events from the multiple historical behavior events to obtain the historical normal events.

[0096] Optionally, the intrusion detection device 300 for a container further includes: The source tracing analysis module is used to perform threat source tracing analysis on historical abnormal events if the first container system behavior event is an intrusion event.

[0097] Reference below Fig.10 , which shows a schematic diagram of the structure of an electronic device (such as a terminal device or a server) 600 suitable for implementing the embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig.10 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0098] like Fig.10 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0099] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Fig.10 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0100] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 609, or installed from a storage device 608, or installed from a ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0101] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0102] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0103] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0104] The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: obtains a first container system behavior event, determines a target image corresponding to a first container to which the first container system behavior event belongs, wherein the first container is created based on the target image; uses a first behavior baseline model corresponding to the target image to determine whether the first container system behavior event is an abnormal event; wherein the first behavior baseline model is obtained based on a target data set training, and the target data set includes information on container system behavior events of one or more containers created based on the target image; if the first container system behavior event is an abnormal event, the first container system behavior event is detected to determine whether the first container system behavior event is an intrusion event.

[0105] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0106] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0107] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the acquisition module may also be described as "a module for acquiring historical normal events of one or more containers created based on the target image".

[0108] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0109] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0110] According to one or more embodiments of the present disclosure, Example 1 provides an intrusion detection method for containers, including: obtaining a first container system behavior event, determining a target image corresponding to a first container to which the first container system behavior event belongs, wherein the first container is created based on the target image; using a first behavior baseline model corresponding to the target image, determining whether the first container system behavior event is an abnormal event; wherein the first behavior baseline model is obtained based on target data set training, and the target data set includes information on container system behavior events of one or more containers created based on the target image; if the first container system behavior event is an abnormal event, detecting the first container system behavior event to determine whether the first container system behavior event is an intrusion event.

[0111] According to one or more embodiments of the present disclosure, Example 2 provides the method of Example 1, wherein the information includes reference feature information of historical normal events of the corresponding container, and one or more containers created based on the target image belong to the host of the first container system behavior event.

[0112] According to one or more embodiments of the present disclosure, Example 3 provides the method of Example 2, wherein the method of using a first behavior baseline model corresponding to the target image to determine whether the first container system behavior event is an abnormal event includes: performing feature extraction on the first container system behavior event to obtain target feature information; using the first behavior baseline model corresponding to the target image to determine whether the target feature information deviates from the reference feature information; if the target feature information deviates from the reference feature information, determining that the first container system behavior event is an abnormal event.

[0113] According to one or more embodiments of the present disclosure, Example 4 provides the method of Example 3, wherein feature extraction is performed on the first container system behavior event to obtain target feature information, including: determining the target event type of the first container system behavior event; and extracting the target feature information of the first container system behavior event using a feature extractor corresponding to the target event type.

[0114] According to one or more embodiments of the present disclosure, Example 5 provides the method of Example 4, wherein the target event type is any one of a process, a network, and a file.

[0115] According to one or more embodiments of the present disclosure, Example 6 provides the method of Example 4, wherein the target event type is process, and the target feature information includes one or more of the following: process path, parent process path, number of parameters, number of numeric parameters, number of special characters in parameters, character length of parameters, whether the parameter exists in a file, whether the parameter exists in a script, whether the parameter exists in an IP, whether the parameter exists in a domain name, whether the input source is a network, and whether the output source is a network.

[0116] According to one or more embodiments of the present disclosure, Example 7 provides the method of Example 3, wherein the first behavior baseline model is an outlier judgment model; the use of the first behavior baseline model corresponding to the target image to judge whether the target feature information deviates from the reference feature information includes: using the first behavior baseline model corresponding to the target image to determine whether the target feature information is an outlier relative to the reference feature information; if the target feature information is an outlier relative to the reference feature information, determining that the target feature information deviates from the reference feature information.

[0117] According to one or more embodiments of the present disclosure, Example 8 provides the method of Example 7, wherein the first behavior baseline model is an isolation forest model, and the target feature information is a target feature matrix; the use of the first behavior baseline model corresponding to the target mirror to determine whether the target feature information is an outlier relative to the reference feature information includes: matching the target feature matrix with the isolation forest model to determine whether the target feature matrix is ​​an outlier.

[0118] According to one or more embodiments of the present disclosure, Example 9 provides the method of Example 2, wherein the first behavior baseline model is constructed in the following manner: obtaining historical normal events of one or more containers created based on the target image; performing feature extraction on the historical normal events to obtain the reference feature information; and using the reference feature information to train the initial model to obtain the first behavior baseline model.

[0119] According to one or more embodiments of the present disclosure, Example 10 provides the method of Example 9, wherein obtaining historical normal events of one or more containers created based on the target image includes: obtaining multiple historical behavior events of one or more containers created based on the target image; performing intrusion detection on the multiple historical behavior events to determine whether each of the historical behavior events is an intrusion event; and eliminating historical behavior events that are intrusion events from the multiple historical behavior events to obtain the historical normal events.

[0120] According to one or more embodiments of the present disclosure, Example 11 provides the method of Example 1, and the method further includes: if the first container system behavior event is an intrusion event, performing a threat tracing analysis on historical abnormal events.

[0121] According to one or more embodiments of the present disclosure, Example 12 provides an intrusion detection device for a container, comprising: a first determination module, used to obtain a first container system behavior event, and determine a target image corresponding to a first container to which the first container system behavior event belongs, wherein the first container is created based on the target image; a second determination module, used to determine whether the first container system behavior event is an abnormal event using a first behavior baseline model corresponding to the target image; wherein the first behavior baseline model is obtained based on training of a target data set, and the target data set includes information on container system behavior events of one or more containers created based on the target image; an intrusion detection module, used to perform intrusion detection on the first container system behavior event if the first container system behavior event is an abnormal event, so as to determine whether the first container system behavior event is an intrusion event.

[0122] According to one or more embodiments of the present disclosure, Example 13 provides a computer-readable medium having a computer program stored thereon, which implements the steps of any of the methods described in Examples 1-11 when executed by a processing device.

[0123] According to one or more embodiments of the present disclosure, Example 14 provides an electronic device, comprising: a storage device on which a computer program is stored; and a processing device for executing the computer program in the storage device to implement the steps of any one of the methods described in Examples 1-11.

[0124] According to one or more embodiments of the present disclosure, Example 15 provides a computer program product, including a computer program, which implements the steps of any one of the methods of Examples 1-11 when executed by a processor.

[0125] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0126] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0127] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims. Regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Claims

1. A method for intrusion detection of a container, characterized in that: include: Acquire a first container system behavior event, and determine a target image corresponding to a first container to which the first container system behavior event belongs, wherein the first container is created based on the target image; Determine whether the first container system behavior event is an abnormal event by using a first behavior baseline model corresponding to the target image; wherein the first behavior baseline model is obtained by training based on a target data set, and the target data set includes information on container system behavior events of one or more containers created based on the target image; If the first container system behavior event is an abnormal event, the first container system behavior event is detected to determine whether the first container system behavior event is an intrusion event.

2. The method according to claim 1, characterized in that: The information includes reference feature information of historical normal events of the corresponding container, and one or more containers created based on the target image belong to the host of the first container system behavior event.

3. The method according to claim 2, characterized in that The determining whether the first container system behavior event is an abnormal event by using the first behavior baseline model corresponding to the target image includes: Extracting features of the first container system behavior event to obtain target feature information; Using a first behavior corresponding to the target image as a baseline model, determining whether the target feature information deviates from the reference feature information; If the target characteristic information deviates from the reference characteristic information, the first container system behavior event is determined to be an abnormal event.

4. The method according to claim 3, characterized in that The extracting features of the first container system behavior event to obtain target feature information includes: Determine a target event type of the first container system behavior event; Utilizing a feature extractor corresponding to the target event type, target feature information of the first container system behavior event is extracted.

5. The method according to claim 4, characterized in that The target event type is any one of process, network, and file.

6. The method according to claim 4, characterized in that The target event type is a process, and the target feature information includes one or more of the following: Process path, parent process path, number of parameters, number of numeric parameters, number of special characters in parameters, character length of parameters, whether the parameter exists in a file, whether the parameter exists in a script, whether the parameter exists in an IP, whether the parameter exists in a domain name, whether the input source is a network, and whether the output source is a network.

7. The method according to claim 3, characterized in that The first behavior baseline model is an outlier judgment model; The using the first behavior corresponding to the target image as a baseline model to determine whether the target feature information deviates from the reference feature information includes: Using a first behavior corresponding to the target image as a baseline model, determining whether the target feature information is an outlier relative to the reference feature information; If the target feature information is an outlier relative to the reference feature information, it is determined that the target feature information deviates from the reference feature information.

8. The method according to claim 7, characterized in that The first behavior baseline model is an isolation forest model, and the target feature information is a target feature matrix; The determining, using the first behavior corresponding to the target image as a baseline model, whether the target feature information is an outlier relative to the reference feature information comprises: The target feature matrix is ​​matched with the isolation forest model to determine whether the target feature matrix is ​​an outlier.

9. The method according to claim 2, characterized in that: The first line baseline model is constructed in the following way: Obtain historical normal events of one or more containers created based on the target image; Extracting features of the historical normal events to obtain the reference feature information; The initial model is trained using the reference feature information to obtain the first behavior as a baseline model.

10. The method according to claim 9, characterized in that The obtaining of historical normal events of one or more containers created based on the target image includes: Acquire multiple historical behavior events of one or more containers created based on the target image; Performing intrusion detection on the multiple historical behavior events to determine whether each of the historical behavior events is an intrusion event; The historical behavior events belonging to intrusion events are eliminated from the multiple historical behavior events to obtain the historical normal events.

11. The method according to claim 1, characterized in that: The method further comprises: If the first container system behavior event is an intrusion event, a threat tracing analysis is performed on historical abnormal events.

12. An intrusion detection device for a container, characterized in that: include: A first determination module is configured to obtain a first container system behavior event and determine a target image corresponding to a first container to which the first container system behavior event belongs, wherein the first container is created based on the target image; A second determination module is configured to determine whether the first container system behavior event is an abnormal event by using a first behavior baseline model corresponding to the target image; wherein the first behavior baseline model is obtained by training based on a target data set, and the target data set includes information on container system behavior events of one or more containers created based on the target image; An intrusion detection module is used to perform intrusion detection on the first container system behavior event if the first container system behavior event is an abnormal event, so as to determine whether the first container system behavior event is an intrusion event.

13. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processing device, the steps of the method according to any one of claims 1 to 11 are implemented.

14. An electronic device, characterized in that: include: a storage device having a computer program stored thereon; A processing device, configured to execute the computer program in the storage device to implement the steps of the method according to any one of claims 1 to 11.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 11 are implemented.

Citation Information

Patent Citations

  • Container environment anomaly detection method and device, medium and computer equipment

    CN111813497A

  • Container mirror image data processing method and device, equipment and medium

    CN114706658A

  • Container environment intrusion detection method and device, and storage medium

    CN116821898A

  • Attack link restoration method and device, computer equipment and storage medium

    CN117692243A

  • Method, electronic device, and computer program product for data protection

    US20220405176A1