Fault reason positioning method and device, equipment, storage medium and program product

Abstract

The invention provides a fault cause positioning method and device, equipment, a storage medium and a program product, and the method comprises the steps: determining target event association information of a target fault, the target event association information comprises a plurality of historical events and a target association relationship of the plurality of historical events, the occurrence moment of the historical event is before the occurrence moment of the target fault; determining at least one target historical event in the plurality of historical events according to the target event association information; and determining a fault reason of the target fault according to the at least one target historical event. And the accuracy of determining the fault reason is improved.

Description

Technical Field

[0001] This application relates to the field of computers, and in particular, to a method, apparatus, device, storage medium, and program product for fault cause location. Background Art

[0002] During the operation of the container orchestration and management platform (Kubernetes), faults may occur. For each fault, it is necessary to determine the corresponding fault cause to resolve the fault.

[0003] In the related art, multiple historical event troubleshooting documents corresponding to multiple historical events can be manually entered in advance in an electronic device. When a target fault occurs, the model can be used to determine the target fault troubleshooting document corresponding to the target fault in the multiple historical event troubleshooting documents, and then one-by-one troubleshooting can be performed according to the target fault troubleshooting document to determine the fault cause of the target fault. However, in the above method, the multiple historical event troubleshooting documents are formulated based on human experience, resulting in low accuracy of troubleshooting according to the target fault troubleshooting document, and thus low accuracy of determining the fault cause. Summary of the Invention

[0004] Multiple aspects of this application provide a method, apparatus, device, storage medium, and program product for fault cause location to improve the accuracy of determining the fault cause.

[0005] In a first aspect, an embodiment of this application provides a method for fault cause location, including:

[0006] Determine the target event association information of the target fault, where the target event association information includes multiple historical events and the target association relationships of the multiple historical events, and the occurrence time of the historical event is before the occurrence time of the target fault;

[0007] Determine at least one target historical event from the multiple historical events according to the target event association information;

[0008] Determine the fault cause of the target fault according to the at least one target historical event.

[0009] In a possible implementation manner, determining the target event association information of the target fault includes:

[0010] Obtain reference information, where the reference information includes event association information corresponding to multiple fault types, and / or multiple first historical events that occurred within a preset time period before the discovery time of the target fault;

[0011] Determine the target event association information according to the reference information.

[0012] In a possible implementation manner, determining the target event association information according to the reference information includes:

[0013] Determining first event association information according to the multiple first historical events;

[0014] Determining second event association information from the event association information corresponding to the multiple fault types;

[0015] Determining the target event association information according to the first event association information and / or the second event association information.

[0016] In a possible implementation manner, determining second event association information from the event association information corresponding to the multiple fault types includes:

[0017] Determining the target fault type of the target fault;

[0018] Determining, as the second event association information, the event association information corresponding to the target fault type among the multiple fault types.

[0019] In a possible implementation manner, determining first event association information according to the multiple first historical events includes:

[0020] Determining the multiple components corresponding to the multiple first historical events and the dependency relationships between the multiple components;

[0021] Determining the association relationships between the multiple first historical events according to the occurrence times of each first historical event and the dependency relationships, where the association relationships include time sequence relationships and the dependency relationships;

[0022] Determining the influence degree of each first historical event on the target fault according to the association relationships;

[0023] Generating the first event association information according to the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault.

[0024] In a possible implementation manner, generating the first event association information according to the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault includes:

[0025] Generating initial event association information according to the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault;

[0026] Performing pruning processing on the initial event association relationships according to preset rules to obtain the first event association information.

[0027] In a possible implementation manner, determining the target event association information according to the first event association information and / or the second event association information includes:

[0028] Determining the first event association information as the target event association information; or,

[0029] Determining the second event association information as the target event association information; or,

[0030] Performing an aggregation process on the first event association information and the second event association information to obtain the target event association information.

[0031] In a possible implementation manner, the first event association information includes a plurality of first historical events, a first association relationship of the plurality of first historical events, and an influence degree of each first historical event on the target fault; the second event association information includes a plurality of second historical events, a second association relationship of the plurality of second historical events, and an influence degree of each second historical event on the target fault; the target associated event further includes a target influence degree of each historical event on the target fault;

[0032] Performing an aggregation process on the first event association information and the second event association information to obtain the target event association information includes:

[0033] Determining the plurality of historical events according to the plurality of first historical events and the plurality of second historical events, where the plurality of historical events includes the plurality of first historical events and the plurality of second historical events;

[0034] Determining a target association relationship of the plurality of historical events according to the first association relationship and the second association relationship;

[0035] Performing an aggregation process according to the first event association information and the second event association information to determine a target influence degree of each historical event on the target fault;

[0036] Generating the target event association information according to the plurality of historical events, the target association relationship of the plurality of historical events, and the target influence degree of each historical event on the target fault.

[0037] In a possible implementation manner, for any one historical event; determining a target influence degree of the historical event on the target fault according to the first event association information and the second event association information includes:

[0038] If the historical event is included in the first event association information and the second event association information respectively, determine a first influence degree of the historical event on the target fault in the first event association information, and determine a second influence degree of the historical event on the target fault in the second event association information, and determine a target influence degree of the historical event on the target fault according to the first influence degree, the second influence degree, and the target association relationship;

[0039] If the historical event is included in one of the first event association information and the second event association information, determine a third influence degree of the historical event on the target fault in the one event association information, and determine a target influence degree of the historical event on the target fault according to the third influence degree and the target association relationship.

[0040] In a possible implementation manner, determining a target influence degree of the historical event on the target fault according to the first influence degree, the second influence degree, and the target association relationship includes:

[0041] Determine the maximum influence degree among the first influence degree and the second influence degree;

[0042] Determine a first adjustment amount according to the target association relationship;

[0043] Determine the sum of the maximum influence degree and the first adjustment amount as the target influence degree of the historical event on the target fault.

[0044] In a possible implementation manner, determining a fault cause of the target fault according to the at least one target historical event includes:

[0045] Process the target historical event through a preset language model to obtain the fault cause of the target fault.

[0046] In a second aspect, an embodiment of the present application provides a fault cause location device, including: a first determination module, a second determination module, and a third determination module, where

[0047] The first determination module is configured to determine target event association information of a target fault, where the target event association information includes a plurality of historical events and a target association relationship of the plurality of historical events, and the occurrence time of the historical event is before the occurrence time of the target fault;

[0048] The second determination module is configured to determine at least one target historical event from the plurality of historical events according to the target event association information;

[0049] The third determination module is configured to determine the cause of the target fault according to the at least one target historical event.

[0050] In a possible implementation manner, the first determination module is specifically configured to:

[0051] Obtain reference information, where the reference information includes event association information corresponding to multiple fault types, and / or multiple first historical events that occurred within a preset time period before the discovery time of the target fault;

[0052] Determine the target event association information according to the event association information corresponding to the multiple fault types, and / or the multiple first historical events.

[0053] In a possible implementation manner, the first determination module is specifically configured to:

[0054] Determine first event association information according to the multiple first historical events;

[0055] Determine second event association information from the event association information corresponding to the multiple fault types;

[0056] Determine the target event association information according to the first event association information and / or the second event association information.

[0057] In a possible implementation manner, the first determination module is specifically configured to:

[0058] Determine the target fault type of the target fault;

[0059] Determine the event association information corresponding to the target fault type among the multiple fault types as the second event association information.

[0060] In a possible implementation manner, the first determination module is specifically configured to:

[0061] Determine multiple components corresponding to the multiple first historical events and the dependency relationships between the multiple components;

[0062] According to the occurrence time of each first historical event and the dependency relationship, determine the association relationship between the multiple first historical events, where the association relationship includes a time sequence relationship and the dependency relationship;

[0063] According to the association relationship, determine the influence degree of each first historical event on the target fault;

[0064] Generate the first event association information according to the multiple first historical events, the association relationship, and the influence degree of each first historical event on the target fault.

[0065] In a possible implementation manner, the first determination module is specifically configured to:

[0066] Generate initial event association information according to the multiple first historical events, the association relationship, and the influence degree of each first historical event on the target fault;

[0067] Perform pruning processing on the initial event association relationship according to preset rules to obtain the first event association information.

[0068] In a possible implementation manner, the first determination module is specifically configured to:

[0069] Determine the first event association information as the target event association information; or,

[0070] Determine the second event association information as the target event association information; or,

[0071] Perform aggregation processing on the first event association information and the second event association information to obtain the target event association information.

[0072] In a possible implementation manner, the first event association information includes multiple first historical events, the first association relationship of the multiple first historical events, and the influence degree of each first historical event on the target fault; the second event association information includes multiple second historical events, the second association relationship of the multiple second historical events, and the influence degree of each second historical event on the target fault; the target associated event further includes the target influence degree of each historical event on the target fault;

[0073] The first determination module is specifically configured to:

[0074] Determine the multiple historical events according to the multiple first historical events and the multiple second historical events, where the multiple historical events include the multiple first historical events and the multiple second historical events;

[0075] Perform aggregation processing on the first association relationship and the second association relationship to determine the target association relationship of the multiple historical events;

[0076] Determine the target influence degree of each historical event on the target fault according to the first event association information and the second event association information;

[0077] Generate the target event association information according to the multiple historical events, the target association relationship of the multiple historical events, and the target influence degree of each historical event on the target fault.

[0078] In a possible implementation, for any historical event, the first determination module is specifically configured to:

[0079] If the historical event is included in the first event association information and the second event association information respectively, determine a first influence degree of the historical event on the target fault in the first event association information, and determine a second influence degree of the historical event on the target fault in the second event association information, and determine a target influence degree of the historical event on the target fault according to the first influence degree, the second influence degree, and the target association relationship;

[0080] If the historical event is included in one of the first event association information and the second event association information, determine a third influence degree of the historical event on the target fault in the one event association information, and determine a target influence degree of the historical event on the target fault according to the third influence degree and the target association relationship.

[0081] In a possible implementation, the first determination module is specifically configured to:

[0082] Determine the maximum influence degree between the first influence degree and the second influence degree;

[0083] Determine a first adjustment amount according to the target association relationship;

[0084] Determine the sum of the maximum influence degree and the first adjustment amount as the target influence degree of the historical event on the target fault.

[0085] In a possible implementation, the third determination module is specifically configured to:

[0086] Process the target historical event through a preset language model to obtain the cause of the target fault.

[0087] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory and a processor;

[0088] The memory stores computer execution instructions;

[0089] The processor executes the computer execution instructions stored in the memory, so that the processor executes the method according to any one of the first aspects.

[0090] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer execution instructions are stored, and when the computer execution instructions are executed by a processor, they are used to implement the method according to any one of the first aspects.

[0091] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program which, when executed by a processor, implements the method shown in any one of the first aspect.

[0092] The embodiments of the present application provide a method, apparatus, device, storage medium and program product for fault cause location. An electronic device can determine target event association information of a target fault, and determine at least one target historical event from multiple historical events according to the target event association information, and then can determine the fault cause of the target fault according to the at least one target historical event. On the one hand, since the electronic device can determine the target event association information of the target fault, the target event association information may include multiple historical events and the internal connection between the multiple historical events, that is, the target association relationship, and there is no need to rely on the historical troubleshooting documents formulated by expert experience for one-by-one troubleshooting, so the accuracy of determining the fault cause is improved; and on the other hand, since there is no need for manual input of multiple historical troubleshooting documents one by one in the electronic device, the labor cost is greatly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] The drawings described herein are used to provide a further understanding of the present application, and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation to the present application. In the drawings:

[0094] Figure 1 is a schematic diagram of a scenario provided for an exemplary embodiment of the present application;

[0095] Figure 2 is a schematic flowchart of a method for fault cause location provided for an exemplary embodiment of the present application;

[0096] Figure 3 is a schematic diagram of target event association information provided for an exemplary embodiment of the present application;

[0097] Figure 4 is a schematic flowchart of another method for fault cause location provided for an exemplary embodiment of the present application;

[0098] Figure 5 is a schematic diagram of first event association information provided for an exemplary embodiment of the present application;

[0099] Figure 6 is a schematic diagram of second event association information provided for an exemplary embodiment of the present application;

[0100] Figure 7 is a schematic diagram of the process of a method for fault cause location provided for an exemplary embodiment of the present application;

[0101] Figure 8Structural schematic diagram of a fault cause location device provided by an embodiment of the present application;

[0102] Figure 9 Structural schematic diagram of an electronic device provided by an exemplary embodiment of the present application. Detailed implementation manners

[0103] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties. Moreover, the collection, use, and processing of relevant data need to comply with relevant laws, regulations, and standards, and corresponding operation entrances are provided for users to choose to authorize or refuse.

[0104] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0105] Figure 1 Scenario schematic diagram provided by an exemplary embodiment of the present application. Please refer to Figure 1 , a Kubernetes cluster may include multiple components, and the multiple components may be component 1, component 2, component 3,..., component n respectively. For example, component 1 may be container group (pod) 1, component 2 may be management controller (ReplicaSet) 1, component 3 may be scheduler (Deployment) 1, etc.

[0106] If the target fault is that component 2 fails, then for this target fault, the target event association information corresponding to the target fault can be determined, and then the fault cause of the target fault can be determined according to the target event association information. Among them, the target event association information may include multiple historical events, the target association relationships between the multiple historical events, and the target influence degree of each historical event on the target fault.

[0107] For example, the target event correlation information may include 4 historical events, namely historical event 1, historical event 2, historical event 3, and historical event 4, and the target correlation relationship between the multiple historical events is as follows: historical event 2 and historical event 3 depend on historical event 1, and historical event 4 depends on historical event 2. That is, the occurrence of historical event 1 will cause the occurrence of historical event 2 and historical event 3, and the occurrence of historical event 2 will cause the occurrence of historical event 4. Then, the cause of the target fault can be determined according to the target event correlation information.

[0108] In the related art, historical event troubleshooting documents corresponding to multiple historical events can be manually entered in advance in an electronic device. When a target fault occurs, the model can be used to determine the target fault troubleshooting document corresponding to the target fault among the multiple historical event troubleshooting documents. Then, the target fault can be troubleshot one by one according to the target fault troubleshooting document to determine the cause of the target fault. However, in the above method, the multiple historical event troubleshooting documents are formulated based on human experience, resulting in low accuracy in troubleshooting according to the target fault troubleshooting document, and thus low accuracy in determining the cause of the fault.

[0109] In the embodiment of the present application, the electronic device can determine the target event correlation information of the target fault. The target event correlation information includes multiple historical events and the target correlation relationship between the multiple historical events. Then, at least one target historical event can be determined among the multiple historical events according to the target event correlation information, and the cause of the target fault can be determined according to the at least one target historical event. Since the electronic device can determine the target event correlation information of the target fault, and the target event correlation information can include multiple historical events and the internal connection between the multiple historical events, that is, the target correlation relationship, there is no need to rely on the historical troubleshooting documents formulated by expert experience for one-by-one troubleshooting, so the accuracy of determining the cause of the fault is improved.

[0110] Next, the technical solution shown in the present application will be described in detail through specific embodiments. It should be noted that the following several embodiments can exist independently or be combined with each other. For the same or similar content, it will not be repeated in different embodiments.

[0111] Figure 2 It is a schematic flowchart of a fault cause location method provided for an exemplary embodiment of the present application. Please refer to Figure 2 , the method may include:

[0112] S201. Determine the target event correlation information of the target fault.

[0113] The execution subject of the embodiments of this application can be an electronic device or a fault cause location device provided in the electronic device. The fault cause location device can be implemented by software or by a combination of software and hardware. The fault cause location device can be a processor in the electronic device. For ease of understanding, in the following, the execution subject being an electronic device is taken as an example for description.

[0114] The target event association information may include multiple historical events and the target association relationships of the multiple historical events. Optionally, the target event association information may further include the target influence degree of each historical event on the target fault.

[0115] The occurrence time of the historical event is before the occurrence time of the target fault.

[0116] A historical event refers to an event that occurs to a certain component in the Kubernetes cluster. Therefore, for any historical event, there is a corresponding component for the historical event. For example, if historical event 1 is that pod-1 fails to pull the image file, then the component corresponding to historical event 1 is pod-1.

[0117] The multiple historical events can be events that occur to a certain component or events that occur to multiple components.

[0118] The target influence degree can be determined based on the historical event and the target association relationship. The target influence degree can be represented by a percentage.

[0119] Optionally, the target event association information can be represented by an ordered tree structure. Next, in combination with Figure 3 , the target event association information will be described.

[0120] Figure 3 This is a schematic diagram of the target event association information provided for the exemplary embodiments of this application. Please refer to Figure 3, the target event correlation information may include multiple historical events, namely historical event 1, historical event 2, …, historical event 7. The target correlation relationships among these 7 historical events may include: historical event 3, historical event 4, and historical event 5 all depend on historical event 1. That is, when historical event 1 occurs, it will cause historical event 3, historical event 4, and historical event 5 to occur, and historical event 3, historical event 4, and historical event 5 are in a parallel relationship; historical event 6 and historical event 7 both depend on historical event 2. That is, when historical event 2 occurs, it will cause historical event 6 and historical event 7 to occur, and historical event 6 and historical event 7 are in a parallel relationship. The target event correlation information may also include the target influence degree of each historical event on the target fault. For example, the target influence degree of historical event 1 on the target fault may be 89%; the target influence degree of historical event 2 on the target fault may be 80%; …; the target influence degree of historical event 7 on the target fault may be 60%.

[0121] In an optional embodiment, the target event correlation information of the target fault may be determined in the following manner: obtain reference information; determine the target event correlation information according to the reference information.

[0122] The reference information may include the event correlation information corresponding to multiple fault types, and / or multiple first historical events that occurred within a preset time period before the discovery time of the target fault.

[0123] Optionally, for any one fault type, the fault type may have corresponding event correlation information. The event correlation information is automatically generated in advance without human intervention and is similar to the target event correlation information shown in Figure 3 and will not be elaborated here.

[0124] The first historical event may be a historical event that occurred within a preset time period before the discovery time of the target fault.

[0125] The discovery time of the target fault may be the feedback time when the user discovers the target fault and gives feedback on the target fault. For example, if the occurrence time of the target fault is 2023 / 10 / 20 13:05, the discovery time of the target fault may be 2023 / 10 / 20 13:15. That is, 10 minutes after the target fault occurs, the user may discover the target fault and give feedback on the target fault.

[0126] The preset time period may be adjusted manually or automatically. For example, the preset time period may be 15 minutes.

[0127] Since the first historical event is a historical event that occurred within a preset time period between the discovery times of the target fault, and the target fault had occurred before the discovery time of the target fault, the first historical event includes the target fault event. For example, if the discovery time of the target fault is 2023 / 10 / 20 13:15 and the preset time period is 15 minutes, the first historical event may include 8 historical events that occurred between 2023 / 10 / 20 13:00 and 2023 / 10 / 20 13:15, and the target fault event that occurred at 2023 / 10 / 20 13:05 is included in the multiple historical events.

[0128] For example, for the target fault, the electronic device can obtain reference information, which may include event association information corresponding to 10 fault types, and / or 8 first historical events. Then, according to the reference information, the target event association information can be determined, and the target event association information can be as Figure 3 shown.

[0129] S202. Determine at least one target historical event from the multiple historical events according to the target event association information.

[0130] Optionally, since the target event association information includes the target influence degree of each historical event on the target fault, at least one target historical event can be determined according to the target influence degree of each historical event on the target fault.

[0131] In an optional embodiment, at least one target historical event can be determined according to the target influence degree of each historical event on the target fault in the following manner: determine a preset threshold; in the multiple historical events, determine the historical events whose target influence degree is greater than or equal to the preset threshold as the target historical events.

[0132] The preset threshold can be preset manually. For example, the preset threshold can be 80%.

[0133] For example, if the target event association information of the target fault is as Figure 3 shown, and the preset threshold is 80%, since the target influence degrees of historical event 1, historical event 2, historical event 3, and historical event 4 on the target fault are greater than the preset threshold, these 4 historical events can be determined as 4 target historical events.

[0134] S203. Determine the cause of the target fault according to at least one target historical event.

[0135] In an optional embodiment, the cause of the target fault can be determined according to at least one target historical event in the following manner: process the target historical event through a preset language model to obtain the cause of the target fault.

[0136] Optionally, the preset language model may be a large language model (LLM).

[0137] Optionally, the electronic device may store Kubernetes general knowledge. The Kubernetes general knowledge may include basic knowledge such as the component names, functions of each component in the Kubernetes cluster, and the meanings corresponding to each failure code when each component fails.

[0138] The electronic device may perform natural language conversion processing on each historical event according to the Kubernetes general knowledge through the LLM model to obtain the cause of the target failure. The cause of the failure may include the processing results of natural language conversion processing on multiple historical events.

[0139] For example, if there are 4 target historical events, the LLM model may process the 4 historical events to obtain 4 processing results. For example, if historical event 1 is "Node-1error", the LLM model may process "Node-1error" according to the Kubernetes general knowledge to obtain processing result 1 as "Node 1 failure"; similarly, if historical event 2 is "pod-1ErrImagePull", and if the Kubernetes general knowledge includes that ErrImagePull means failure to pull the image, the LLM model may process "pod-1ErrImagePull" to obtain processing result 2 as "Container group 1 fails to pull the image";...; The historical event 4 may be processed to obtain processing result 4, and the cause of the target failure may include the 4 processing results.

[0140] In the embodiments of the present application, the electronic device may determine the target event association information of the target failure, and determine at least one target historical event among multiple historical events according to the target event association information, and then may determine the cause of the target failure according to the at least one target historical event. On the one hand, since the electronic device can automatically determine the target event association information of the target failure, and the target event association information may include multiple historical events, the internal connection between multiple historical events, that is, the target association relationship, and the target impact degree of each historical event on the target failure, there is no need to rely on the historical troubleshooting documents formulated by expert experience for one-by-one troubleshooting, so the accuracy of determining the cause of the failure is improved; and on the other hand, since there is no need to manually input multiple historical troubleshooting documents one by one in the electronic device, the labor cost is greatly reduced.

[0141] Next, based on the Figure 2 illustrated embodiments, in combination with Figure 4, a detailed description of the above fault cause location method will be given.

[0142] Figure 4 It is a flowchart of another fault cause location method provided by an exemplary embodiment of this application. Please refer to Figure 4 , the method may include:

[0143] S401. Obtain event association information corresponding to multiple fault types, and / or multiple first historical events that occurred within a preset time period before the discovery time of the target fault.

[0144] Optionally, multiple fault types and event association information corresponding to each fault type may be stored in the electronic device.

[0145] The electronic device may determine multiple fault types and obtain event association information corresponding to the multiple fault types. For example, the electronic device may obtain 10 fault types and event association information corresponding to the 10 fault types respectively.

[0146] It should be noted that for any fault type, the event association information of the fault type is obtained through multiple optimizations.

[0147] Optionally, the electronic device may obtain multiple first historical events within a preset time period before the discovery time of the target fault according to the discovery time of the target fault, and obtain multiple first historical events.

[0148] For example, if the discovery time of the target fault is 2023 / 10 / 20 13:15 and the preset time period is 15 minutes, the electronic device may obtain multiple first historical events that occurred between 2023 / 10 / 20 13:00 and 2023 / 10 / 20 13:15. Suppose 10 first historical events can be obtained.

[0149] S402. Determine first event association information according to the multiple first historical events.

[0150] Since the multiple first historical events occurred within a preset time period before the discovery time of the target fault, all the multiple first historical events may be the reasons for the occurrence of the target fault. Therefore, first event association information corresponding to the target fault may be determined according to the multiple first historical events.

[0151] In an optional embodiment, the first event association information can be determined based on multiple first historical events in the following manner: determining multiple components corresponding to the multiple first historical events and the dependency relationships between the multiple components; determining the association relationships between the multiple first historical events according to the occurrence times of each first historical event and the dependency relationships; determining the influence degree of each first historical event on the target fault according to the association relationships; and generating the first event association information based on the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault.

[0152] Optionally, the dependency relationships between multiple components can be pre-stored in the electronic device, or the dependency relationships between multiple components can be automatically determined according to the architecture of the Kubernetes cluster.

[0153] Optionally, the electronic device can determine the association relationships between the multiple first historical events through an automated association model according to the occurrence times of each first historical event and the dependency relationships between the multiple components. The association relationships can include chronological relationships and dependency relationships.

[0154] After determining the association relationships between the multiple first historical events, the influence degree of each first historical event on the target fault can be determined through the LLM model according to the association relationships. Specifically, it can be determined through the LLM model whether a first historical event is a normal event. If so, it can be determined that the influence degree of this first historical event on the target fault is small; if not, it can be determined that the influence degree of this first historical event on the target fault is large. Optionally, it can also be determined through the LLM model whether a first historical event is a parent node in the association relationship. If the first historical event is a parent node in the association relationship, it can be determined that the influence degree of this first historical event on the target fault is large; if the first historical event is a child node in the association relationship, it can be determined that the influence degree of this first historical event on the target fault is small.

[0155] Optionally, the first event association information can be generated in the following manner based on the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault: generating initial event association information based on the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault; and pruning the initial event association relationship according to preset rules to obtain the first event association information.

[0156] The initial event association relationship includes multiple first historical events, the association relationships between the multiple first historical events, and the influence degree of each first historical event on the target fault.

[0157] Optionally, the size of the initial event association relationship can be represented by the number of historical events in the initial event association relationship. For example, if there are 10 historical events in the initial event association relationship, the size of the initial event association relationship is 10.

[0158] Optionally, the preset rule can be set manually according to the actual application scenario.

[0159] Optionally, the size of the target association relationship in the first event association information can be preset manually. For example, the size of the fault association relationship in the first event association information can be a first threshold. If the size of the initial event association relationship is greater than the first threshold, pruning processing can be performed on the initial event association relationship according to the preset rule; if the size of the initial event association relationship is less than or equal to the first threshold, pruning processing on the initial event association relationship is not required according to the preset rule.

[0160] The first event association information may include multiple first historical events, the first association relationships of the multiple first historical events, and the influence degree of each first historical event on the target fault.

[0161] Next, in conjunction with Figure 5 , the first event association information will be described.

[0162] Figure 5 FIG. is a schematic diagram of the first event association information provided by an exemplary embodiment of the present application. Please refer to Figure 5 , for example, if there are 10 first historical events, the components corresponding to the 10 first historical events can be determined respectively. Assume that the 10 first historical events, the occurrence time of each first historical event, and the corresponding components are shown in Table 1:

[0163] Table 1

[0164]

[0165]

[0166] And if the dependency relationships among multiple components are as follows: Component 2 depends on Component 1, and Component 3 depends on Component 2, then the occurrence time of historical event 14 is earlier than that of historical event 15, and the components corresponding to both historical event 14 and historical event 15 are Component 1, indicating that after historical event 14 occurred in Component 1, it might have caused historical event 15 to occur. Therefore, it is possible to determine through the automated association model that historical event 15 depends on historical event 14; similarly, since the occurrence time of historical event 1 is earlier than those of historical event 3, historical event 4, and historical event 5, and Component 2 depends on Component 1, it shows that after historical event 1 occurred, it might have caused historical event 3, historical event 4, and historical event 5 to occur in sequence. Therefore, it is possible to determine through the automated association model that historical event 3, historical event 4, and historical event 5 all depend on historical event 1; and historical event 3, historical event 4, and historical event 5 are in a parallel relationship; similarly, it is possible to determine through the automated association model that historical event 6 depends on historical event 2 based on the occurrence times of historical event 2, historical event 6, and historical event 7; it is possible to determine through the automated association model that historical event 13 depends on historical event 12 based on the occurrence times of historical event 12 and historical event 13, and the fact that Component 3 depends on Component 2. Then the association relationship as shown in Figure 5 can be obtained.

[0167] Optionally, after determining the association relationships among the 10 first historical events, the LLM model can be used to determine the impact degree of each first historical event on the target fault based on the association relationships.

[0168] For example, if historical event 14 is that Node 1 is running normally, historical event 15 is that pod-1 is running normally, and historical event 1 is that the pod has a running fault, since historical event 14 and historical event 15 are normal events and historical event 1 is an abnormal event, then the LLM model can be used to determine that the impact degree of historical event 1 on the target fault is greater than the impact degrees of historical event 14 and historical event 15 on the target fault respectively. Assume that the impact degrees of historical event 14, historical event 15, and historical event 1 on the target fault can be determined to be 35%, 30%, and 84% respectively. Also, since historical event 1 is the parent node, assume that it can be further determined that the impact degree of historical event 1 on the target fault is 84%. Assume that the LLM model can be used to determine the impact degrees of multiple first historical events on the target fault respectively, as shown in Figure 5 Then, based on the multiple first historical events, the association relationships, and the impact degree of each first historical event on the target fault, initial event association information can be generated, as shown in Figure 5 can be obtained.

[0169] If the preset rule is to sort the influence degrees in ascending order and remove the first X historical events from the sorting result, where X is obtained by subtracting the first threshold from the size of the initial event association information, and X is an integer greater than or equal to 1. For example, if the initial event association information is as shown in Figure 5 and there are 10 first historical events in the initial event association relationship, that is, the size of the initial event association information is 10. If the first threshold is 6, and if the influence degrees of these 10 first historical events on the target fault are as shown in Figure 5 , then according to the preset rule, the first 4 first historical events with smaller influence degrees can be removed from the initial event association relationship, that is, historical event 14, historical event 15, historical event 12, and historical event 13, so as to perform pruning processing on the initial event association relationship and obtain the first event association information, as shown in Figure 5 . That is, the first event association information includes 6 first historical events, the first association relationships of 6 first historical events, and the influence degree of each first historical event on the target fault.

[0170] S403. Determine the second event association information among the event association information corresponding to multiple fault types.

[0171] Since the type of the target fault is the target fault type, if the target fault type is included in multiple fault types, it means that a fault of the same type has occurred before. The event association information corresponding to the target fault type can be used as a reference for determining the fault cause of the target fault. Therefore, the event association information of the target fault type can be determined as the second event association information corresponding to the target fault.

[0172] In an alternative embodiment, the second event association information can be determined among the event association information corresponding to multiple fault types in the following manner: Determine the target fault type of the target fault; and determine the event association information corresponding to the target fault type among the multiple fault types as the second event association information.

[0173] For example, if the target fault is that pod-1 fails to pull the image file, the target fault type can be determined as pulling the image file fails. If there are 10 fault types and fault type 2 is pulling the image file fails, then the event association information corresponding to fault type 2 can be determined as the second event association information.

[0174] The second event association information may include multiple second historical events, the second association relationships of multiple second historical events, and the influence degree of each second historical event on the target fault.

[0175] Next, the second event association information will be described in conjunction with Figure 6 this.

[0176] Figure 6 Schematic diagram of the second event association information provided by an exemplary embodiment of the present application. Please refer to Figure 6 , the second event association information may include 4 second historical events, namely historical event 1, historical event 2, historical event 3, historical event 4, and historical event 7. The second association relationships between the multiple historical events may include: historical event 3 and historical event 4 depend on historical event 1, and historical event 7 depends on historical event 2. The influence degree of each historical event on the target fault may be as shown in Figure 6 .

[0177] S404. Determine the target event association information according to the first event association information and / or the second event association information.

[0178] In an alternative embodiment, the target event association information may be determined according to the first event association information and / or the second event association information in the following manner: determine the first event association information as the target event association information; or, determine the second event association information as the target event association information; or, perform an aggregation process on the first event association information and the second event association information to obtain the target event association information.

[0179] For example, if the electronic device can determine that the first event association information is as shown in Figure 5 , then the first event association information may be determined as the target event association information; or, if the electronic device can determine that the second event association information is as shown in Figure 6 , then the second event association information may be determined as the target event association information.

[0180] When the electronic device determines that there is the first event association information and the second event association information, since the first event association information is determined according to multiple first historical events within a preset time period before the discovery time of the target fault, and the second event association information is determined according to the event association information corresponding to the same fault type, the multiple first historical events included in the first event association information and the multiple second historical events included in the second event association information are all possible causes that may lead to the occurrence of the target fault. Therefore, an aggregation process may be performed on the first event association information and the second event association information to obtain the target event association information of the target fault.

[0181] In an alternative embodiment, the first event association information and the second event association information may be aggregated to obtain target event association information in the following manner: determining a plurality of historical events according to a plurality of first historical events and a plurality of second historical events, where the plurality of historical events includes the plurality of first historical events and the plurality of second historical events; aggregating the first association relationship and the second association relationship to determine the target association relationship of the plurality of historical events; determining the target impact degree of each historical event on the target fault according to the first event association information and the second event association information; generating the target event association information according to the plurality of historical events, the target association relationship of the plurality of historical events, and the target impact degree of each historical event on the target fault.

[0182] For example, if the first event association information is as Figure 5 shown, and the second event association information is as Figure 6 shown. Since the first event association information includes historical event 1, historical event 2, historical event 3, historical event 4, historical event 5, and historical event 6, and the second event association information includes historical event 1, historical event 2, historical event 3, historical event 4, and historical event 7, it can be determined that the plurality of historical events includes historical event 1, historical event 2, historical event 3, historical event 4, historical event 5, historical event 6, and historical event 7.

[0183] Since the first association relationship included in the first event association information is as Figure 5 shown: historical event 3, historical event 4, and historical event 5 all depend on historical event 1, and historical event 6 depends on historical event 2; and the second association relationship included in the second event association information is as Figure 6 shown: historical event 3 and historical event 4 both depend on historical event 1, and historical event 7 depends on historical event 2. Then, the first association relationship and the second association relationship can be aggregated to determine the target association relationship of the plurality of historical events. The target association relationship is as Figure 3 shown: historical event 3, historical event 4, and historical event 5 all depend on historical event 1, and historical event 6 and historical event 7 both depend on historical event 2.

[0184] Optionally, since each historical event in the first event association information and the second event association information has an impact degree on the target fault, for any historical event, the target impact degree of the historical event on the target fault can be determined according to the first event association information and the second event association information in the following manner: If the historical event is included in both the first event association information and the second event association information, determine the first impact degree of the historical event on the target fault in the first event association information, and determine the second impact degree of the historical event on the target fault in the second event association information, and determine the target impact degree of the historical event on the target fault according to the first impact degree, the second impact degree, and the target association relationship; If only one of the first event association information and the second event association information includes the historical event, determine the third impact degree of the historical event on the target fault in one of the event association information, and determine the target impact degree of the historical event on the target fault according to the third impact degree and the target association relationship.

[0185] In an alternative embodiment, the target impact degree of the historical event on the target fault can be determined according to the first impact degree, the second impact degree, and the target association relationship in the following manner: Determine the maximum impact degree among the first impact degree and the second impact degree; Determine the first adjustment amount according to the target association relationship; Determine the sum of the maximum impact degree and the first adjustment amount as the target impact degree of the historical event on the target fault.

[0186] For example, if the first event association information is as Figure 5 shown, and the second event association information is as Figure 6 shown, since both the first event association information and the second event association information include historical event 1, historical event 2, historical event 3, and historical event 4, the first impact degree and the second impact degree of these 4 historical events can be determined in the first event association information as shown in Table 2:

[0187] Table 2

[0188]

[0189]

[0190] Then, for historical event 1, the maximum impact degree can be determined to be 84%, and assuming that the first adjustment amount corresponding to historical event 1 can be determined to be 5% according to the target association relationship, the target impact degree of historical event 1 on the target fault can be determined to be 89%; Similarly, for historical event 2, historical event 3, and historical event 4, the maximum impact degree and the first adjustment amount corresponding to these 3 historical events can be determined respectively, and then the sum of the maximum impact degree and the first adjustment amount is determined as the target impact degree of the historical event, as shown in Table 2.

[0191] For historical event 5 and historical event 6, since historical event 5 and historical event 6 exist in the first event association information but do not exist in the second event association information, the first influence degree of historical event 5, 70%, and the first influence degree of historical event 6, 75%, are determined as the third influence degree of historical event 5, 70%, and the third influence degree of historical event 6, 75%, in the first event association information. Furthermore, based on the third influence degree and the target association relationship, the target influence degrees of historical event 5 and historical event 6 on the target fault can be determined. Suppose the target influence degree of historical event 5 can be determined as 69% and the target influence degree of historical event 6 can be determined as 74%. Similarly, for historical event 7, since historical event 7 does not exist in the first event association information but exists in the second event association information, the first influence degree of historical event 7, 60%, can be determined as the third influence degree of historical event 7 in the second event association information. Furthermore, based on the third influence degree and the target association relationship, the target influence degree of historical event 7 on the target fault can be determined. Suppose the target influence degree of historical event 7 can be determined as 60%.

[0192] After determining the target association relationship among the 7 historical events and the influence degree of each historical event on the target fault, the target event association information can be generated based on the 7 historical events, the target association relationship of the 7 historical events, and the influence degree of each historical event on the target fault, as Figure 3 shown below.

[0193] It should be noted that the method of determining the target influence degree of each historical event on the target fault in the target event association information by the LLM model can be set according to the actual application scenario. In any embodiment of the present application, the specific numerical value of the influence degree is only an example.

[0194] S405. Determine at least one target historical event among multiple historical events according to the target event association information.

[0195] S406. Determine the cause of the target fault according to at least one target historical event.

[0196] It should be noted that the specific execution process of steps S405 - S406 can refer to steps S202 - S203, and will not be elaborated here.

[0197] In an embodiment of the present application, the electronic device can obtain event association information corresponding to multiple fault types, and / or multiple first historical events that occurred within a preset time period before the discovery time of the target fault. The electronic device can determine the first event association information based on the multiple first historical events; and can determine the second event association information from the event association information corresponding to the multiple fault types. The electronic device can determine the target event association information based on the first event association information and / or the second event association information, and determine at least one target historical event from the multiple historical events according to the target event association information, and then determine the cause of the target fault based on the at least one target historical event. On the one hand, since the electronic device can determine the target event association information of the target fault, and the target event association information can include multiple historical events, the internal relationship between the multiple historical events, that is, the target association relationship, and the target impact degree of each historical event on the target fault, there is no need to rely on the historical troubleshooting documents formulated by expert experience for one-by-one troubleshooting, so the accuracy of determining the cause of the fault is improved; and on the other hand, since there is no need to manually input multiple historical troubleshooting documents into the electronic device one by one, the labor cost is greatly reduced.

[0198] Next, on the basis of any of the above embodiments, in combination with Figure 7 ,the fault cause location method will be further described.

[0199] Figure 7 It is a schematic diagram of the process of a fault cause location method provided by an exemplary embodiment of the present application. Please refer to Figure 7 ,including steps ①②③④⑤⑥⑦.

[0200] The Kubernetes cluster can include multiple components and multiple historical events. The multiple components can be component 1, component 2,..., component n respectively, and the multiple historical events can be historical event 1, historical event 2,..., historical event m respectively.

[0201] In step ①, the electronic device can query multiple historical events in the Kubernetes cluster and the component corresponding to each historical event, and generate event association information corresponding to multiple fault types through an automated association model and an LLM model. For example, event association information 1-10 corresponding to fault type 1,..., event association information K-3 corresponding to fault type K can be generated.

[0202] It should be noted that for any type of fault, the event association information corresponding to the fault type is continuously optimized based on historical event association information. For example, if historical faults 1-1, 1-2, …, 1-10 all belong to fault type 1, when historical fault 1-1 occurs, the event association information 1-1 corresponding to historical fault 1 can be determined through the automated association model and the LLM model, and then the event association information 1-1 can be determined as the event association information corresponding to fault type 1; when historical fault 1-2 occurs, on the basis of event association information 1-1, the LLM model can be used to optimize event association information 1-1 to obtain the event association information 1-2 corresponding to historical fault 2, and the event association information 1-2 can be determined as the event association information corresponding to fault type 1; …; when historical fault 1-10 occurs, on the basis of event association information 1-9, the LLM model can be used to optimize event association information 1-9 to obtain the event association information 1-10 corresponding to historical fault 10, and the event association information 1-10 can be determined as the event association information corresponding to fault type 1.

[0203] The event association information corresponding to the K fault types has been generated before the target fault occurs.

[0204] In step ②, the electronic device can determine the target fault type of the target fault. Suppose it can be determined that the target fault type is fault type 1.

[0205] In step ③, the electronic device can determine to use the event association information corresponding to the target fault type as the second event association information. For example, if the target fault type is fault type 1, the event association information 1-10 corresponding to fault type 1 can be determined as the second event association information.

[0206] In step ④, the electronic device can determine multiple first historical events within a preset time period before the discovery time of the target fault according to the discovery time of the target fault. Suppose it can be determined that the 3 first historical events are historical event 1, historical event 2, and historical event 6 respectively.

[0207] In step ⑤, the electronic device can generate first event association information based on multiple first historical events. For example, the electronic device can determine multiple components corresponding to historical event 1, historical event 2, and historical event 6 respectively, as well as the dependency relationships between the multiple components. Furthermore, based on the occurrence times of the three first historical events and the dependency relationships, the electronic device can determine the association relationship between the three first historical events as follows: historical event 2 and historical event 6 depend on historical event 1. After determining the association relationship, the electronic device can determine the impact degrees of the target faults corresponding to the three first historical events according to the association relationship. Suppose it can be determined that the impact degrees of historical event 1, historical event 2, and historical event 6 on the target fault are 84%, 75%, and 70% respectively.

[0208] In step ⑥, the electronic device can determine target event association information based on the first event association information and / or the second event association information. Optionally, the electronic device can perform an aggregation process on the first event association information and the second event association information to obtain the target event association information. The process of determining the target event association information can refer to step S404 and will not be elaborated here. Suppose the target event association information is as shown in Figure 7 The target event association information includes historical event 1, historical event 2, historical event 3, historical event 4, historical event 5, and historical event 6, and the target association relationship between the six historical events is as follows: historical event 2, historical event 3, historical event 4, historical event 5, and historical event 6 all depend on historical event 1, and the impact degrees of historical event 1, historical event 2, historical event 3, historical event 4, historical event 5, and historical event 6 on the target fault are 89%, 80%, 80%, 76%, 70%, and 73% respectively.

[0209] Optionally, after determining the target event association information of the target fault, the electronic device can use the target association information as the event association information 1-11 corresponding to fault type 1, so that when a target fault of target fault type 1 occurs again, the event association information corresponding to fault type 1 can be determined as event association information 1-11.

[0210] In step ⑦, the electronic device can use the LLM model to determine at least one target historical event among multiple historical events based on the target event association information, and determine the cause of the target fault according to the at least one target historical event.

[0211] For example, if the target event association information of the target fault is as shown in Figure 7As shown, if the preset threshold is 80%, since the target impact degrees of historical event 1, historical event 2, and historical event 3 on the target fault are greater than the preset threshold, these 3 historical events can be determined as 3 target historical events. Furthermore, through the LLM model, based on the general knowledge of Kubernetes, the target historical events can be processed to obtain the cause of the target fault. For example, if historical event 1 is "Node-1error", through the LLM model, based on the general knowledge of Kubernetes, "Node-1error" can be processed to obtain that the cause of the fault 1 is "Node 1 failure", and the target impact degree of the cause of the fault 1 on the target fault is 89%; similarly, if historical event 2 is "pod-1ErrImagePull", if the general knowledge of Kubernetes includes that ErrImagePull indicates that the image pull fails, then "pod-1ErrImagePull" can be processed through the LLM to obtain that the cause of the fault 2 is "The image pull of container group 1 fails", and the target impact degree of the cause of the fault 2 on the target fault is 80%; if historical event 3 is "Network PluginNot Ready", then through the LLM model, based on the general knowledge of Kubernetes, "Network Plugin NotReady" can be processed to obtain that the cause of the fault 3 is "The network plugin is not started", and the target impact degree of the cause of the fault 3 on the target fault is 80%.

[0212] Through the solution of this application, event association information corresponding to multiple fault types can be generated through the automated association model and the LLM model, without the need to actively input the troubleshooting documents formulated according to expert experience, reducing the labor cost; and when a target fault occurs, not only can the first event association information be immediately generated through the automated association model; but also through the LLM model, based on the event association information corresponding to multiple fault types, the second event association information can be determined. Furthermore, through the LLM model, based on the first event association information and / or the second event association information, the target event association information can be determined. Through the LLM model, a good diagnostic effect on problems in the Kubernetes cloud native field can be achieved. Even for new faults occurring in the Kubernetes field, the cause of the fault can be inferred through the internal association relationships of multiple components in the Kubernetes field.

[0213] In an embodiment of the present application, an electronic device may obtain event association information corresponding to multiple fault types and / or multiple first historical events that occurred within a preset period before the discovery time of the target fault. The electronic device may determine first event association information based on the multiple first historical events, and may determine second event association information from the event association information corresponding to the multiple fault types. The electronic device may determine target event association information based on the first event association information and / or the second event association information, and determine at least one target historical event from the multiple historical events according to the target event association information. Furthermore, the root cause of the target fault may be determined based on the at least one target historical event. On the one hand, since the electronic device can determine the target event association information of the target fault, and the target event association information may include multiple historical events, the inherent connection between the multiple historical events, i.e., the target association relationship, and the target influence degree of each historical event on the target fault, there is no need to rely on the historical troubleshooting documents formulated based on expert experience for one-by-one troubleshooting, thus improving the accuracy of determining the root cause of the fault. On the other hand, since there is no need for manual input of multiple historical troubleshooting documents one by one in the electronic device, the labor cost is greatly reduced.

[0214] Figure 8 FIG. is a schematic structural diagram of a root cause location device provided by an embodiment of the present application. Please refer to Figure 8 The root cause location device 10 includes: a first determination module 11, a second determination module 12, and a third determination module 13, where

[0215] The first determination module 11 is configured to determine target event association information of a target fault, where the target event association information includes multiple historical events and the target association relationship of the multiple historical events, and the occurrence time of the historical event is before the occurrence time of the target fault;

[0216] The second determination module 12 is configured to determine at least one target historical event from the multiple historical events according to the target event association information;

[0217] The third determination module 13 is configured to determine the root cause of the target fault according to the at least one target historical event.

[0218] The root cause location device provided by the embodiment of the present application may execute the technical solutions shown in the above method embodiments, and the implementation principles and beneficial effects are similar, which will not be elaborated here.

[0219] In a possible implementation manner, the first determination module 11 is specifically configured to:

[0220] Obtain reference information, where the reference information includes event association information corresponding to multiple fault types, and / or, multiple first historical events that occurred within a preset time period before the discovery time of the target fault;

[0221] Determine the target event association information according to the reference information.

[0222] In a possible implementation manner, the first determination module 11 is specifically configured to:

[0223] Determine first event association information according to the multiple first historical events;

[0224] Determine second event association information from the event association information corresponding to the multiple fault types;

[0225] Determine the target event association information according to the first event association information and / or the second event association information.

[0226] In a possible implementation manner, the first determination module 11 is specifically configured to:

[0227] Determine the target fault type of the target fault;

[0228] Determine the event association information corresponding to the target fault type among the multiple fault types as the second event association information.

[0229] In a possible implementation manner, the first determination module 11 is specifically configured to:

[0230] Determine the multiple components corresponding to the multiple first historical events and the dependency relationships between the multiple components;

[0231] Determine the association relationships between the multiple first historical events according to the occurrence time of each first historical event and the dependency relationships, where the association relationships include time sequence relationships and the dependency relationships;

[0232] Determine the influence degree of each first historical event on the target fault according to the association relationships;

[0233] Generate the first event association information according to the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault.

[0234] In a possible implementation manner, the first determination module 11 is specifically configured to:

[0235] Generate initial event association information according to the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault;

[0236] Prune the initial event association relationship according to preset rules to obtain the first event association information.

[0237] In a possible implementation manner, the first determination module 11 is specifically configured to:

[0238] Determine the first event association information as the target event association information; or,

[0239] Determine the second event association information as the target event association information; or,

[0240] Perform an aggregation process on the first event association information and the second event association information to obtain the target event association information.

[0241] In a possible implementation manner, the first event association information includes multiple first historical events, the first association relationship of the multiple first historical events, and the influence degree of each first historical event on the target fault; the second event association information includes multiple second historical events, the second association relationship of the multiple second historical events, and the influence degree of each second historical event on the target fault; the target associated event also includes the target influence degree of each historical event on the target fault;

[0242] The first determination module 11 is specifically configured to:

[0243] Determine the multiple historical events according to the multiple first historical events and the multiple second historical events, where the multiple historical events include the multiple first historical events and the multiple second historical events;

[0244] Perform an aggregation process on the first association relationship and the second association relationship to determine the target association relationship of the multiple historical events;

[0245] Determine the target influence degree of each historical event on the target fault according to the first event association information and the second event association information;

[0246] Generate the target event association information according to the multiple historical events, the target association relationship of the multiple historical events, and the target influence degree of each historical event on the target fault.

[0247] In a possible implementation manner, for any one historical event; the first determination module 11 is specifically configured to:

[0248] If the historical event is included in the first event association information and the second event association information respectively, determine the first influence degree of the historical event on the target fault in the first event association information, and determine the second influence degree of the historical event on the target fault in the second event association information, and determine the target influence degree of the historical event on the target fault according to the first influence degree, the second influence degree and the target association relationship;

[0249] If the historical event is included in one of the first event association information and the second event association information, determine the third influence degree of the historical event on the target fault in the one event association information, and determine the target influence degree of the historical event on the target fault according to the third influence degree and the target association relationship.

[0250] In a possible implementation manner, the first determination module 11 is specifically configured to:

[0251] Determine the maximum influence degree among the first influence degree and the second influence degree;

[0252] Determine a first adjustment amount according to the target association relationship;

[0253] Determine the sum of the maximum influence degree and the first adjustment amount as the target influence degree of the historical event on the target fault.

[0254] In a possible implementation manner, the third determination module 13 is specifically configured to:

[0255] Process the target historical event through a preset language model to obtain the cause of the target fault.

[0256] The fault cause location device provided by the embodiments of the present application can execute the technical solutions shown in the above method embodiments, and its implementation principles and beneficial effects are similar, and will not be described in detail here.

[0257] An exemplary embodiment of the present application provides a schematic structural diagram of an electronic device. Please refer to Figure 9 , the electronic device 20 may include a processor 21 and a memory 22. Exemplarily, the processor 21 and the memory 22 are interconnected with each other through a bus 23.

[0258] The memory 22 stores computer execution instructions;

[0259] The processor 21 executes the computer execution instructions stored in the memory 22, so that the processor 21 executes the method as shown in the above method embodiments.

[0260] Figure 9 The electronic device shown in [description] can be a server in a Kubernetes cluster.

[0261] Accordingly, an embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the method described in the above method embodiment.

[0262] Accordingly, an embodiment of the present application can also provide a computer program product including a computer program that, when executed by a processor, can implement the method shown in the above method embodiment.

[0263] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0264] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified function in [a] Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0265] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the specified function in [a] Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0266] These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing in [a]Figure 1 one or more processes and / or blocks Figure 1 steps of the functions specified in one or more blocks

[0267] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0268] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

[0269] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology for information storage. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0270] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0271] The above description is only for the embodiments of the present application and is not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for locating the cause of a fault, characterized in that, Including: Determine the target event association information of the target fault, where the target event association information includes multiple historical events and the target association relationships of the multiple historical events, and the occurrence time of the historical events is before the occurrence time of the target fault; Determine at least one target historical event from the multiple historical events according to the target event association information; Determine the cause of the target fault according to the at least one target historical event.

2. The method according to claim 1, wherein Determine the target event association information of the target fault, including: Obtain reference information, where the reference information includes event association information corresponding to multiple fault types, and / or multiple first historical events that occurred within a preset time period before the discovery time of the target fault; Determine the target event association information according to the reference information.

3. The method according to claim 2, wherein Determine the target event association information according to the reference information, including: Determine first event association information according to the multiple first historical events; Determine second event association information from the event association information corresponding to the multiple fault types; Determine the target event association information according to the first event association information and / or the second event association information.

4. The method according to claim 3, wherein Determine second event association information from the event association information corresponding to the multiple fault types, including: Determine the target fault type of the target fault; Determine the event association information corresponding to the target fault type among the multiple fault types as the second event association information.

5. The method according to claim 3, wherein Determine first event association information according to the multiple first historical events, including: Determine the multiple components corresponding to the multiple first historical events and the dependency relationships between the multiple components; Determine the association relationships between the multiple first historical events according to the occurrence time of each first historical event and the dependency relationships, where the association relationships include time sequence relationships and the dependency relationships; Determine the influence degree of each first historical event on the target fault according to the association relationships; Generate the first event association information according to the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault.

6. The method according to claim 5, characterized in that, Generate the first event association information according to the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault, including: Generate initial event association information according to the multiple first historical events, the association relationships, and the influence degree of each first historical event on the target fault; Perform pruning processing on the initial event association relationship according to preset rules to obtain the first event association information.

7. The method according to any one of claims 3-6, characterized in that Determine the target event association information according to the first event association information and / or the second event association information, including: Determine the first event association information as the target event association information; or, Determine the second event association information as the target event association information; or, Perform aggregation processing on the first event association information and the second event association information to obtain the target event association information.

8. The method according to claim 7, characterized in that, The first event association information includes multiple first historical events, the first association relationships of the multiple first historical events, and the influence degree of each first historical event on the target fault; the second event association information includes multiple second historical events, the second association relationships of the multiple second historical events, and the influence degree of each second historical event on the target fault; The target associated event further includes the target influence degree of each historical event on the target fault; Performing an aggregation process on the first event association information and the second event association information to obtain the target event association information, including: Determining the multiple historical events according to the multiple first historical events and the multiple second historical events, where the multiple historical events include the multiple first historical events and the multiple second historical events; Performing an aggregation process on the first association relationship and the second association relationship to determine the target association relationship of the multiple historical events; Determining the target influence degree of each historical event on the target fault according to the first event association information and the second event association information; Generating the target event association information according to the multiple historical events, the target association relationship of the multiple historical events, and the target influence degree of each historical event on the target fault.

9. The method according to claim 8, wherein For any one historical event; determining the target influence degree of the historical event on the target fault according to the first event association information and the second event association information, including: If the historical event is respectively included in the first event association information and the second event association information, determining the first influence degree of the historical event on the target fault in the first event association information, and determining the second influence degree of the historical event on the target fault in the second event association information, and determining the target influence degree of the historical event on the target fault according to the first influence degree, the second influence degree, and the target association relationship; If the historical event is included in one of the first event association information and the second event association information, determining the third influence degree of the historical event on the target fault in the one event association information, and determining the target influence degree of the historical event on the target fault according to the third influence degree and the target association relationship.

10. The method according to claim 9, wherein Determining the target influence degree of the historical event on the target fault according to the first influence degree, the second influence degree, and the target association relationship, including: Determining the maximum influence degree between the first influence degree and the second influence degree; Determining a first adjustment amount according to the target association relationship; Taking the sum of the maximum influence degree and the first adjustment amount as the target influence degree of the historical event on the target fault.

11. The method according to any one of claims 1 to 10, characterized in that, Determining the fault cause of the target fault according to the at least one target historical event, including: Processing the target historical event through a preset language model to obtain the fault cause of the target fault.

12. A fault cause location device, characterized in that, Including: A first determination module, a second determination module, and a third determination module, wherein the first determination module is configured to determine target event association information of a target fault, the target event association information includes a plurality of historical events and target association relationships of the plurality of historical events, and occurrence times of the historical events are before the occurrence time of the target fault; the second determination module is configured to determine at least one target historical event from the plurality of historical events according to the target event association information; the third determination module is configured to determine a fault cause of the target fault according to the at least one target historical event.

13. An electronic device, characterized in that, Comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to cause the electronic device to execute the method according to any one of claims 1-11.

14. A computer-readable storage medium, characterized in that, Computer-executable instructions are stored in the computer-readable storage medium, and when the processor executes the computer-executable instructions, the method according to any one of claims 1-11 is implemented.

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

Images