Business exception processing method and device based on DPU and storage medium

Abstract

The invention provides a service exception processing method and device based on a DPU (Data Processing Unit) and a storage medium. The method is applied to a DPU, the DPU is connected with a host, a service grid is deployed in the DPU, the service grid is used for performing proxy processing on service flow transmitted and received by the host, and the method comprises the following steps: acquiring abnormal log data of the service grid; based on the abnormal log data and a log database, determining an abnormal reason corresponding to the abnormal log data; determining an influence factor of the abnormal log data on the service grid based on the abnormal reason; and updating the service grid based on the influence factor to eliminate abnormal log data. In this way, the abnormal problem existing in proxy processing of the service grid on the service flow can be solved. In addition, the whole anomaly analysis process is performed on the DPU, so that the occupation of system resources of the host is further reduced.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of data processing technology, and in particular to a DPU-based service exception processing method, device, and storage medium. Background Art

[0002] At present, on the host in the computer cluster network, the service grid proxy needs to consume certain system resources when processing business traffic, thereby reducing the system resources available for the service. In order to solve this problem, the proxy processing of business traffic by the service grid can be offloaded to the data processing unit (DPU) to save system resources.

[0003] However, since the DPU is not in the computer cluster network, if there is an abnormality in the proxy processing of business traffic by the service grid, it may not be possible to use the operation and maintenance means in the computer cluster network to resolve the abnormality. Summary of the invention

[0004] In view of this, the present application provides a DPU-based service exception processing method, device and storage medium, which can solve the exception problem existing in the proxy processing of service traffic by the service grid on the DPU.

[0005] In a first aspect, the present application provides a DPU-based business exception handling method, which is applied to a data processing unit DPU, the DPU is connected to a host, a service grid is deployed in the DPU, and the service grid is used to perform proxy processing on business traffic sent and received by the host. The method may include: obtaining abnormal log data of the service grid, the abnormal log data characterizing that there is an abnormality in the proxy processing of the business traffic by the service grid; based on the abnormal log data and a log database, determining the abnormal cause corresponding to the abnormal log data, the log database including the corresponding relationship between different abnormal log data and the abnormal cause; based on the abnormal cause, determining the impact factor of the abnormal log data on the service grid; updating the service grid based on the impact factor to eliminate the abnormal log data.

[0006] In the second aspect, the present application provides a DPU-based business exception processing device, which is applied to a data processing unit DPU, the DPU is connected to a host, and a service grid is deployed in the DPU. The service grid is used to perform proxy processing on the business traffic sent and received by the host. The device may include: an acquisition module, which is used to obtain abnormal log data of the service grid, and the abnormal log data indicates that there is an abnormality in the proxy processing of the business traffic by the service grid; a determination module, which is used to determine the abnormal cause corresponding to the abnormal log data based on the abnormal log data and a log database, and the log database includes the corresponding relationship between different abnormal log data and the abnormal cause; the determination module is also used to determine the impact factor of the abnormal log data on the service grid based on the abnormal cause; an update module, which is used to update the service grid based on the impact factor to eliminate the abnormal log data.

[0007] In a third aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by at least one processor, it implements the method described in the first aspect.

[0008] In a fourth aspect, the present application provides a computer program product, comprising computer program instructions, which enable a computer to execute the method described in the first aspect.

[0009] In a fifth aspect, the present application provides a computer program, which, when executed on a computer, enables the computer to execute the method described in the first aspect above.

[0010] The present application provides a service exception processing method, device and storage medium based on DPU. The method may include: if there is an exception in the proxy processing of service traffic by the service grid on the DPU, the impact factor of the abnormal log data on the service grid can be determined by the abnormal cause corresponding to the abnormal log data, and the service grid is updated based on the impact factor to remove the abnormal log data. In this way, the abnormal problem of the proxy processing of service traffic by the service grid can be solved. In addition, the entire abnormal analysis process is performed on the DPU, which further reduces the occupation of system resources of the host. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0012] Figure 1 It is a schematic diagram of a DPU-based business processing scenario of the present application;

[0013] Figure 2 This is a flow diagram of a DPU-based service exception handling method provided in an embodiment of the present application. Figure 1 ;

[0014] Figure 3 This is a flow diagram of a DPU-based service exception handling method provided in an embodiment of the present application. Figure 2 ;

[0015] Figure 4 It is a schematic diagram of the composition structure of a DPU-based service exception processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0016] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and in detail below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the embodiments of the present application should fall within the scope of protection of the embodiments of the present application.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0018] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0019] It can be understood that the term "and / or" in the embodiments of the present application is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0020] To facilitate understanding of the technical solutions of the embodiments of the present application, the relevant technologies of the embodiments of the present application are described below. The following related technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all belong to the protection scope of the embodiments of the present application.

[0021] Figure 1 This is a schematic diagram of a DPU-based business processing scenario of the present application.

[0022] It should be understood that DPU is a new generation of computing chip that is data-centric, input / output (I / O) intensive, and uses software-defined technology to support infrastructure resource layer virtualization. It has the ability to improve computing system efficiency, reduce the total cost of ownership of the overall system, improve data processing performance, and reduce performance loss of other computing chips.

[0023] like Figure 1 As shown, the DPU can be plugged into the host through a high-speed serial computer expansion bus standard (Peripheral Component Interconnect Express, PCIE) slot to achieve the connection between the DPU and the host. The host can be a computing node in a computer cluster, and the computer cluster can be a container orchestration management cluster (Kubernetes, K8s). The computer cluster may include a management node and multiple computing nodes. Computing nodes, also known as workload nodes, can run containerized business applications, i.e., services, in such nodes; management nodes can be responsible for managing and controlling containerized business applications running in computing nodes, for example, deploying containerized business applications to computing nodes. Both computing nodes and management nodes can be servers.

[0024] As the complexity of business applications increases, traditional business applications are gradually replaced by microservice architecture. Microservice architecture refers to splitting a large and complex business application into a series of small, independent microservices, each of which can be deployed and expanded independently. However, when using microservice architecture, there will be complex communication problems between microservices. For this reason, service mesh technology came into being. Service mesh manages the communication between microservices by inserting a proxy layer between microservices, thereby reducing the burden of microservices themselves in handling communication.

[0025] A service mesh is a platform for managing communication between microservices in a microservice architecture. It usually consists of two components: a control plane and a data plane. The control plane is responsible for tasks such as service discovery, load balancing, and traffic management, while the data plane is responsible for actual request forwarding and processing.

[0026] As you can understand, Envoy is an open source, high-performance proxy that is usually used as the data plane of the service grid, responsible for actual request forwarding and processing. The Envoy architecture can be divided into the data plane, management plane, and control plane. The control plane is used to manage policies and configurations related to traffic routing and forwarding; the management plane obtains the latest traffic configuration information through the standard application programming interface (Application Programming Interface, API); the data plane forwards traffic based on the configuration rules issued by the control plane.

[0027] At present, in computer cluster networks, the traditional sidecar service grid is deployed in sidecar mode. When starting each service, a sidecar container needs to be started. Therefore, when there are many services, more sidecar containers need to be started, which causes the service grid proxy to consume more system resources (such as central processing unit (CPU), memory, etc.) when processing business traffic, reducing the system resources available for the service. In order to solve this problem, the service grid can offload the proxy processing of business traffic to the DPU to save system resources, increase the system resources available for the service, and thus improve the efficiency of the service.

[0028] It should be understood that offloading the proxy processing of business traffic by the service grid to the DPU can be understood as offloading the data plane of the service grid to the DPU so that the data plane can implement proxy processing of business traffic on the DPU.

[0029] In a computer cluster network, in the process of service grid proxy processing of business traffic, common operation and maintenance methods can be three types of releases: blue-green release, grayscale release and rolling release. These three types of releases are mainly used for application upgrades and switching between old and new businesses, bringing the software from the final stage of testing to the production environment.

[0030] However, on the one hand, since the DPU is not in the computer cluster network, if there is an abnormality in the service grid's proxy processing of business traffic on the DPU, it may not be possible to use the operation and maintenance means in the computer cluster network to resolve the abnormality; on the other hand, the above three releases are mainly used for upgrade scenarios and do not involve solutions to abnormalities in the service grid's proxy processing of business traffic.

[0031] Based on this, the embodiment of the present application provides a method for handling business exceptions. If there is an exception in the proxy processing of business traffic by the service grid on the DPU, the impact factor of the abnormal log data on the service grid can be determined by the abnormal cause corresponding to the abnormal log data, and the service grid can be updated based on the impact factor to remove the abnormal log data. In this way, the abnormal problem of the proxy processing of business traffic by the service grid can be solved. In addition, the entire abnormal analysis process is performed on the DPU, which further reduces the occupation of system resources of the host.

[0032] To facilitate understanding of the technical solutions of the embodiments of the present application, the technical solutions of the present application are described in detail below through specific embodiments. The above related technologies can be combined arbitrarily with the technical solutions of the embodiments of the present application as optional solutions, and they all belong to the protection scope of the embodiments of the present application. The embodiments of the present application include at least part of the following contents.

[0033] Figure 2This is a flow diagram of a DPU-based service exception handling method provided in an embodiment of the present application. Figure 1 ,This method is applied to DPU, which is connected to the host. A service grid is deployed in the DPU, and the service grid is used to proxy the service traffic sent and received by the host. Figure 2 As shown, the method may include the following steps.

[0034] S210: Acquire abnormal log data of the service grid, where the abnormal log data indicates that there is an abnormality in the proxy processing of business traffic by the service grid.

[0035] It should be noted that, in the embodiment of the present application, the DPU-based service exception processing method can be applied to a DPU-based service exception processing device.

[0036] Exemplarily, the operation log data of the service grid may be obtained, and then the abnormal log data may be filtered out from the operation log data.

[0037] In some embodiments, obtaining abnormal log data of a service grid may include: obtaining operation log data of the service grid based on preset acquisition rules; the preset acquisition rules include a preset acquisition frequency or an acquisition instruction sent by the execution host; and determining the abnormal log data based on preset abnormal parameters and the operation log data.

[0038] Exemplarily, a timing module may be provided in the DPU, and a preset acquisition frequency may be set through the timing module, and the operation log data may be acquired through the preset acquisition frequency. For example, the preset acquisition frequency may be 24 hours, 36 hours, or a week.

[0039] Exemplarily, a user may send an acquisition instruction to the DPU through the host, and the DPU acquires the operation log data based on the acquisition instruction.

[0040] Exemplarily, based on the preset acquisition rules, obtaining the operation log data of the service grid may include: obtaining the operation log data based on the preset log level. For example, the operation log data may be graded in advance based on the preset log level to obtain operation log data of different levels; then, the operation log data corresponding to the target level may be screened out from the operation log data of different levels, and the operation log data corresponding to the target level may be used to obtain abnormal log data, thereby reducing the amount of data to be processed subsequently and improving data processing efficiency.

[0041] Among them, the preset log level can be set manually or in other ways, and the embodiments of the present application are not limited to this.

[0042] The target level can be obtained based on the log detection level in the configuration file of the DPU.

[0043] Furthermore, based on the preset abnormal parameter, the abnormal log data can be determined from the operation log data. For example, assuming that the preset abnormal parameter is a delay threshold, and the operation log data includes session delay data, then when the session delay data indicates that the session delay exceeds the delay threshold, the session delay data is abnormal log data.

[0044] It should be understood that the preset abnormal parameters can be flexibly set by those skilled in the art based on actual conditions, or can be set in other ways, and the embodiments of the present application are not limited to this.

[0045] Through this method, operation log data can be acquired based on a preset acquisition frequency or an acquisition instruction sent by the execution host to obtain abnormal log data, thereby allowing users to control the frequency and timing of log analysis according to needs, with high flexibility.

[0046] It should be noted that before obtaining the exception log data of the service grid, you can first configure the configuration file for the DPU in the business exception handling process.

[0047] Furthermore, a configuration file can be statically created on any host, and the configuration file can be quickly deployed on all DPUs through the high-speed transmission channel between the host and the DPU. The high-speed transmission channel is a high-speed network port used for internal communication between the host and the DPU, which can be transmitted based on the DPU hardware direct memory access (DMA), and can achieve a cluster-like effect. The high-speed transmission channel can improve the efficiency of data interaction between the host and the DPU.

[0048] Exemplarily, first, the user can customize the configuration parameters in the configuration file based on the user template; secondly, the user template is rendered and parsed to obtain the parsed configuration file; then, the configuration required on the DPU side is written into a pre-template; finally, the parsed configuration file is used to fill the pre-template to obtain the configuration file that needs to be deployed on the DPU.

[0049] It should be understood that rendering and parsing the user template can be understood as converting a configuration file recognizable by a user into a configuration file recognizable by a machine.

[0050] Furthermore, by rendering and parsing the user template, it can flexibly support users to customize the configuration parameters in various configuration files, which can take effect quickly without modifying the code, thereby improving usability.

[0051] It should be understood that the configuration parameters in the configuration file may include business exception response events, log detection levels, log alarm levels, etc., which are not limited in the embodiments of the present application.

[0052] Exemplarily, a business exception response event can be understood as, when the service grid performs proxy processing on the business traffic sent and received by the host, it is necessary to pay attention to whether there is any abnormality in the entire proxy processing.

[0053] Exemplarily, the log detection level gives a target level when filtering the operation log data.

[0054] Exemplarily, the log alarm level gives the alarm level corresponding to different abnormal causes. The greater the impact of the abnormal cause, the higher the corresponding alarm level. Different alarm levels can correspond to different alarm methods, which can include different color distinctions, different voice broadcasts, etc., which are not limited in the embodiments of the present application.

[0055] S220: Determine the abnormal cause corresponding to the abnormal log data based on the abnormal log data and the log database, wherein the log database includes a correspondence between different abnormal log data and the abnormal cause.

[0056] It should be noted that the abnormal log data and the abnormal causes in the log database can be obtained based on historical records, based on reasoning, or in other ways, and the embodiments of the present application are not limited to this.

[0057] It should also be noted that the abnormal problem reflected by the abnormal log data can be used as a matching object to match the corresponding abnormal cause from the log database.

[0058] It can be understood that one set of abnormal log data may correspond to one or more abnormal reasons. For example, when the abnormal problem reflected by the abnormal log data is high session delay, the corresponding abnormal reason may be a communication network problem or a system memory problem of the DPU, such as excessive memory usage, etc.

[0059] S230. Based on the cause of the exception, determine the impact factor of the abnormal log data on the service grid.

[0060] It should be noted that abnormal log data will have an impact on the service grid itself. The impact factor can also be called the influencing factor, the degree of impact, the scope of impact, etc. The impact factor is used to characterize the type of impact of abnormal log data on the service grid, that is, what aspects of the impact of abnormal log data on the service grid.

[0061] In some embodiments, the impact factor may include one or more of the following:

[0062] The current version of the service mesh is abnormal;

[0063] The current configuration of the service mesh is abnormal;

[0064] An abnormal number of containers deploying multiple versions of the service mesh.

[0065] Exemplarily, when the cause of the exception is a communication network problem, the influencing factor may be that the current version of the service grid is abnormal, and / or that the current configuration of the service grid is abnormal.

[0066] Exemplarily, when the cause of the exception is a system memory problem of the DPU, the influencing factor may be an abnormal number of containers deploying multiple versions of the service grid.

[0067] It is understandable that the DPU system memory problem indicates that the number of containers in the service grid may be too large, thus occupying too much DPU system memory.

[0068] Through this method, the impact factor of abnormal log data on the service grid can be determined based on the cause of the abnormality, so that the abnormal log data can be eliminated according to the impact factor to solve the abnormal problem of the service grid's proxy processing of business traffic.

[0069] S240. Update the service grid based on the impact factor to remove abnormal log data.

[0070] In some embodiments, updating the service grid based on the impact factor may include:

[0071] If the current version of the service grid is abnormal, roll back the version of the service grid from the current version to the historical version; and / or,

[0072] If the current configuration of the service grid is abnormal, roll back the configuration of the service grid from the current configuration to the historical configuration; and / or,

[0073] If the number of containers for deploying multiple versions of the service mesh is abnormal, reduce the number of containers.

[0074] It should be noted that if the current configuration of the service grid is abnormal, a configuration rollback is required (i.e., the configuration of the service grid is rolled back from the current configuration to the historical configuration). For example, a backup record of the current configuration can be made, and the cause of the current problem can be marked, and then a relatively stable configuration version in the historical record can be hot-loaded, thus completing the configuration rollback.

[0075] It should be understood that the DPU can record all historical configuration information of the service grid, including the health status of all historical configurations, etc. When a configuration rollback is required, a fully functional and relatively stable historical configuration can be selected.

[0076] It should also be understood that the historical configuration may be pre-stored in the DPU, and the pre-stored historical configuration may be directly used when performing a configuration rollback; alternatively, the historical configuration may be generated based on the historical configuration information when a configuration rollback is required, and this is not limited to this in the embodiments of the present application.

[0077] It should be noted that if the current version of the service grid is abnormal, a version rollback is required (i.e., the version of the service grid is rolled back from the current version to the historical version).

[0078] It should be understood that the DPU can record all historical version information of the service grid, including the health status of all historical versions, etc. When a version rollback is required, a fully functional and relatively stable historical version can be selected.

[0079] It should also be understood that historical versions can be pre-deployed in the container of the DPU, and the pre-deployed historical versions can be directly used when performing version rollback; alternatively, historical versions can also be generated based on historical version information when version rollback is required, and the embodiments of the present application are not limited to this.

[0080] It should be noted that if the number of containers for deploying multiple versions of the service grid is abnormal, the system memory of the DPU can be released by reducing the number of containers.

[0081] It is understandable that a container usually deploys one version of the service mesh. An abnormal number of containers indicates that too many historical versions of the deployed service mesh are deployed. In this case, you can reduce the number of containers and the historical versions of the service mesh to free up the system memory of the DPU.

[0082] Through this method, the service grid can be updated based on the impact factor to remove abnormal log data. In this way, the abnormal problem of the service grid's proxy processing of business traffic can be solved.

[0083] In some embodiments, after rolling back the version of the service grid from the current version to the historical version, the method may further include: performing one or more of the following operations on the historical version of the service grid:

[0084] Synchronize the configuration of the historical version of the service mesh with the current version of the service mesh;

[0085] Based on the diversion strategy, the business traffic flowing through the current version of the service mesh is diverted to the historical version of the service mesh.

[0086] Exemplarily, synchronously configuring a historical version of a service mesh with a current version of a service mesh may include: synchronously configuring a shared memory address parameter or a shared configuration file to achieve configuration synchronization between the historical version of the service mesh and the current version of the service mesh.

[0087] It should be noted that the traffic diversion strategy is a special high-performance traffic diversion strategy provided by DPU. The traffic diversion strategy can realize the switching of business traffic between different versions of service networks, so that when the current version of the service grid is abnormal, business traffic can be switched more efficiently, thereby improving the efficiency of traffic switching.

[0088] Through this method, by synchronously configuring the historical version of the service mesh and the current version of the service mesh, and / or, based on the diversion strategy, diverting the business traffic flowing through the current version of the service mesh to the historical version of the service mesh, smooth rolling from the current version of the service mesh to the historical version of the service mesh can be achieved.

[0089] In some embodiments, the method may also include: if there is an abnormality in the service grid's proxy processing of business traffic, and the business traffic is greater than a preset threshold, reducing the business traffic to less than the preset threshold, so that the service grid performs proxy processing on the business traffic less than the preset threshold.

[0090] It should be noted that the preset threshold may be set manually or by other means, and the embodiments of the present application are not limited to this.

[0091] Through this method, the business flow of the service grid with abnormalities can be reduced, and the workload of the service grid when proxy processing the business flow can be reduced, thereby reducing the frequency of abnormal problems.

[0092] The embodiment of the present application provides a method for handling business exceptions. If there is an exception in the proxy processing of business traffic by the service grid on the DPU, the impact factor of the abnormal log data on the service grid can be determined by the abnormal cause corresponding to the abnormal log data, and the service grid is updated based on the impact factor to remove the abnormal log data. In this way, the abnormal problem of the proxy processing of business traffic by the service grid can be solved. In addition, the entire abnormal analysis process is performed on the DPU, which further reduces the occupation of system resources of the host.

[0093] In the embodiment of the present application, the abnormal cause corresponding to the abnormal log data can be determined based on the abnormal log data and the log database. Figure 3 The technical solution for determining the cause of the abnormality is explained.

[0094] Figure 3 This is a flow diagram of a DPU-based service exception handling method provided in an embodiment of the present application. Figure 2 .like Figure 3 As shown, the method may include the following steps.

[0095] S310, determining one or more initial abnormal causes corresponding to the abnormal log data based on the abnormal log data and the log database;

[0096] S320, when one or more initial abnormal reasons include a system abnormality of the DPU, obtaining system log data of the DPU;

[0097] S330, determining the system operation status of the DPU based on the system log data;

[0098] S340: Based on the system operation status, select an initial abnormality cause from one or more initial abnormality causes as the abnormality cause.

[0099] It should be noted that, in the case where one or more initial abnormal causes include a system abnormality of the DPU, it can be determined through the system log data of the DPU whether the abnormal problem reflected in the abnormal log data is caused by the system abnormality of the DPU.

[0100] For example, when the abnormal problem reflected by the abnormal log data is a high session delay, the corresponding abnormal cause may be a communication network problem or a system abnormality problem of the DPU. At this time, the system log data of the DPU can be read by the dot counting method, and the system memory occupancy of the DPU can be determined based on the system log data. If the system memory occupancy of the DPU is greater than the threshold, it means that the abnormal problem reflected by the abnormal log data is caused by the system abnormality of the DPU; if the system memory occupancy of the DPU is less than the threshold, it means that the abnormal problem reflected by the abnormal log data is caused by the communication network.

[0101] It should be understood that reading the system log data of the DPU by the dot counting method can be understood as dividing the system log data of the DPU into multiple parts, each part is recorded as a point, and the system log data is read in sequence according to the points to check the system operation status of the DPU.

[0102] In some embodiments, the method may further include: receiving update information for the log database sent by the host, the update information including newly added abnormal log data and corresponding abnormal reasons; and updating the log database based on the update information.

[0103] Exemplarily, the log database can be updated periodically. For example, an update information template can be provided to the user periodically through the host, so that the user can fill in the newly added exception log data and the corresponding exception cause into the update information template, and send the update information template to the DPU through the host. The DPU can update the log database based on the update information in the update information template.

[0104] Through this method, by updating the log database, the data comprehensiveness of the log database can be improved, thereby making the results obtained by log analysis more accurate.

[0105] In some embodiments, the method may also include: obtaining address information of the DPU; generating an exception analysis report for the service grid to perform proxy processing on business traffic based on exception log data, exception causes and address information; and sending the exception analysis report to the host to display the exception analysis report to the user through the host.

[0106] It can be understood that in addition to the abnormal analysis report, the log alarm level corresponding to the abnormal cause can also be displayed to the user through the host, so that the user can see the abnormal problem more clearly and intuitively based on the log alarm level.

[0107] By using this method, by adding the address information of the DPU to the abnormal analysis report, when there are multiple DPUs, the abnormal traffic processing of each DPU can be quickly and accurately distinguished. In addition, by showing the abnormal analysis report to the user, the user (such as the operation and maintenance personnel) can accurately locate the abnormal DPU and the cause of the abnormality after receiving the abnormal analysis report, so that the user can troubleshoot and solve the abnormal problem as soon as possible, saving human resources.

[0108] In some embodiments, the method may also include: if the exception cause corresponding to the exception log data is not matched from the log database, then based on the exception log data and the address information of the DPU, generating an exception analysis report for the service grid to perform proxy processing on the business traffic; sending the exception analysis report to the host to display the exception analysis report to the user through the host.

[0109] Through this method, if there is abnormal log data obtained and the corresponding abnormal cause is not matched from the log database, it indicates that the abnormal log data may be a new abnormal situation. At this time, the abnormal log data and the address information of the DPU can be used to generate an abnormal analysis report, and then the abnormal analysis report can be sent to the host, so that the abnormal analysis report can be displayed to the user through the host, so that the user can manually troubleshoot the abnormal cause.

[0110] The embodiment of the present application provides a method for handling business exceptions, which can determine the abnormal cause corresponding to the abnormal log data based on the abnormal log data and the log database, so that the abnormal problem of the proxy processing of business traffic by the service grid can be solved based on the abnormal cause. In addition, the entire abnormal analysis process is performed on the DPU, which further reduces the occupation of the system resources of the host.

[0111] The preferred embodiments of the present application are described in detail above in conjunction with the accompanying drawings. However, the present application is not limited to the specific details in the above embodiments. Within the technical concept of the present application, the technical solution of the present application can be subjected to a variety of simple modifications, and these simple modifications all belong to the protection scope of the present application. For example, the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present application will not further explain various possible combinations. For another example, the various different embodiments of the present application can also be arbitrarily combined, as long as they do not violate the idea of ​​the present application, they should also be regarded as the contents disclosed in the present application. For another example, under the premise of no conflict, the various embodiments and / or the technical features in the various embodiments described in the present application can be arbitrarily combined with the prior art, and the technical solution obtained after the combination should also fall within the protection scope of the present application.

[0112] It should also be understood that in the various method embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0113] Based on the same inventive concept as the above embodiments, Figure 4 1 is a schematic diagram of the structure of a DPU-based service exception processing device provided in an embodiment of the present application. The device is applied to the DPU, the DPU is connected to the host, and a service grid is deployed in the DPU. The service grid is used to perform proxy processing on the service traffic sent and received by the host. Figure 4 As shown, the DPU-based service exception processing device 400 may include an acquisition module 410, a determination module 420 and an update module 430, wherein:

[0114] The acquisition module 410 is used to acquire abnormal log data of the service grid, where the abnormal log data indicates that there is an abnormality in the proxy processing of the service traffic by the service grid;

[0115] A determination module 420, configured to determine the abnormal cause corresponding to the abnormal log data based on the abnormal log data and a log database, wherein the log database includes a correspondence between different abnormal log data and the abnormal cause;

[0116] The determination module 420 is further used to determine the impact factor of the abnormal log data on the service grid based on the abnormal cause;

[0117] The updating module 430 is used to update the service grid based on the impact factor to remove abnormal log data.

[0118] In some embodiments, the impact factor includes one or more of the following:

[0119] The current version of the service mesh is abnormal;

[0120] The current configuration of the service mesh is abnormal;

[0121] An abnormal number of containers deploying multiple versions of the service mesh.

[0122] In some embodiments, the DPU-based service exception processing device 400 may further include a rollback module 440 and a processing module 450, wherein:

[0123] A rollback module 440 is used to roll back the version of the service grid from the current version to the historical version if the current version of the service grid is abnormal; and / or, if the current configuration of the service grid is abnormal, roll back the configuration of the service grid from the current configuration to the historical configuration; and / or,

[0124] The processing module 450 is configured to reduce the number of containers if the number of containers for deploying multiple versions of the service grid is abnormal.

[0125] In some embodiments, after rolling back the version of the service grid from the current version to the historical version, the processing module 450 is further configured to perform one or more of the following operations on the historical version of the service grid:

[0126] Synchronize the configuration of the historical version of the service mesh with the current version of the service mesh;

[0127] Based on the diversion strategy, the business traffic flowing through the current version of the service mesh is diverted to the historical version of the service mesh.

[0128] In some embodiments, the processing module 450 is also used to reduce the business traffic to less than the preset threshold if there is an abnormality in the service grid's proxy processing of the business traffic and the business traffic is greater than a preset threshold, so that the service grid performs proxy processing on the business traffic less than the preset threshold.

[0129] In some embodiments, the DPU-based service exception processing device 400 may further include a selection module 460, wherein:

[0130] The determination module 420 is further configured to determine one or more initial abnormal causes corresponding to the abnormal log data based on the abnormal log data and the log database;

[0131] The acquisition module 410 is further configured to acquire system log data of the DPU when one or more initial abnormality causes include a system abnormality of the DPU;

[0132] The determination module 420 is further used to determine the system operation status of the DPU based on the system log data;

[0133] The selection module 460 is used to select an initial abnormality cause from one or more initial abnormality causes as the abnormality cause based on the system operation status.

[0134] In some embodiments, the acquisition module 410 is also used to acquire the operation log data of the service grid based on preset acquisition rules; the preset acquisition rules include a preset acquisition frequency or an acquisition instruction sent by the execution host; the determination module 420 is also used to determine the abnormal log data based on preset abnormal parameters and the operation log data.

[0135] In some embodiments, the DPU-based service exception processing device 400 may further include a generation module 470 and a communication module 480, wherein:

[0136] The acquisition module 410 is also used to acquire the address information of the DPU;

[0137] A generating module 470, for generating an abnormal analysis report for proxy processing of business traffic by the service grid based on abnormal log data, abnormal cause and address information;

[0138] The communication module 480 is used to send the abnormality analysis report to the host so as to display the abnormality analysis report to the user through the host.

[0139] Those skilled in the art should understand that the relevant description of the above-mentioned sound pressure level calibration device in the embodiment of the present application can be understood by referring to the relevant description of the sound pressure level calibration method in the embodiment of the present application.

[0140] The embodiment of the present application provides a service exception processing device. If there is an exception in the proxy processing of service traffic by the service grid on the DPU, the impact factor of the abnormal log data on the service grid can be determined by the abnormal cause corresponding to the abnormal log data, and the service grid is updated based on the impact factor to remove the abnormal log data. In this way, the abnormal problem of the proxy processing of service traffic by the service grid can be solved. In addition, the entire abnormal analysis process is performed on the DPU, which further reduces the occupation of system resources of the host.

[0141] An embodiment of the present application also provides a computer-readable storage medium for storing a computer program.

[0142] In some embodiments, the computer-readable storage medium can be applied to the DPU in the embodiments of the present application, and when the computer program is executed by at least one processor, it implements the corresponding processes implemented by the DPU in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0143] An embodiment of the present application also provides a computer program product, including computer program instructions.

[0144] In some embodiments, the computer program product can be applied to the DPU in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the DPU in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0145] The embodiment of the present application also provides a computer program.

[0146] In some embodiments, the computer program can be applied to the DPU in the embodiments of the present application. When the computer program runs on a computer, the computer executes the corresponding processes implemented by the DPU in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0147] Those of ordinary skill in the art will appreciate that the units and method steps of each example described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the embodiments of this application.

[0148] It should be noted that, in this application, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0149] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0150] The methods disclosed in several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0151] The features disclosed in several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0152] The features disclosed in several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0153] The above implementation methods are only used to illustrate the embodiments of the present application, and are not limitations on the embodiments of the present application. Ordinary technicians in the relevant technical field can make various changes and modifications without departing from the spirit and scope of the embodiments of the present application. Therefore, all equivalent technical solutions also belong to the scope of the embodiments of the present application. The scope of patent protection of the embodiments of the present application should be limited by the claims.

Claims

1. A service exception processing method based on DPU, characterized in that: Applied to a data processing unit DPU, the DPU is connected to a host, a service grid is deployed in the DPU, and the service grid is used to perform proxy processing on the service traffic sent and received by the host. The method includes: Acquire abnormal log data of the service grid, where the abnormal log data indicates that there is an abnormality in the proxy processing of the service traffic by the service grid; Determining the abnormal cause corresponding to the abnormal log data based on the abnormal log data and a log database, wherein the log database includes a correspondence between different abnormal log data and the abnormal cause; Based on the cause of the exception, determining an impact factor of the abnormal log data on the service grid; The service grid is updated based on the impact factor to remove the abnormal log data.

2. The method according to claim 1, characterized in that The impact factors include one or more of the following: The current version of the service grid is abnormal; The current configuration of the service grid is abnormal; An abnormal number of containers deploying multiple versions of the service mesh.

3. The method according to claim 1 or 2, characterized in that: The updating of the service grid based on the impact factor includes: If the current version of the service grid is abnormal, rolling back the version of the service grid from the current version to a historical version; and / or, If the current configuration of the service grid is abnormal, rolling back the configuration of the service grid from the current configuration to the historical configuration; and / or, If the number of containers of the service grids deployed with multiple versions is abnormal, the number of containers is reduced.

4. The method according to claim 3, characterized in that: After rolling back the version of the service grid from the current version to the historical version, the method further includes: Perform one or more of the following operations on the historical version of the service mesh: Synchronously configuring the historical version of the service grid and the current version of the service grid; Based on the diversion strategy, the business traffic flowing through the current version of the service grid is diverted to the service grid of the historical version.

5. The method according to claim 1 or 2, characterized in that: The method further comprises: If there is an abnormality in the proxy processing of the business traffic by the service grid, and the business traffic is greater than a preset threshold, the business traffic is reduced to less than the preset threshold so that the service grid performs proxy processing on the business traffic less than the preset threshold.

6. The method according to claim 1 or 2, characterized in that: The determining, based on the abnormal log data and the log database, the abnormal cause corresponding to the abnormal log data includes: Determine one or more initial abnormal causes corresponding to the abnormal log data based on the abnormal log data and the log database; and obtain the system log data of the DPU when the one or more initial abnormal causes include a system abnormality of the DPU; Determining a system operating status of the DPU based on the system log data; Based on the system operation state, an initial abnormality cause is selected from the one or more initial abnormality causes as the abnormality cause.

7. The method according to claim 1 or 2, characterized in that: The obtaining of abnormal log data of the service grid includes: Based on a preset acquisition rule, acquiring the operation log data of the service grid; the preset acquisition rule includes a preset acquisition frequency or executing an acquisition instruction sent by the host; Based on the preset abnormal parameters and the operation log data, the abnormal log data is determined.

8. The method according to claim 1 or 2, characterized in that: The method further comprises: Obtaining address information of the DPU; Based on the abnormal log data, the abnormal cause and the address information, generating an abnormal analysis report of the service grid performing proxy processing on the business traffic; The abnormality analysis report is sent to the host so as to display the abnormality analysis report to the user through the host.

9. A service exception processing device based on DPU, characterized in that: Applied to a data processing unit DPU, the DPU is connected to a host, a service grid is deployed in the DPU, the service grid is used to perform proxy processing on the service traffic sent and received by the host, and the device includes: An acquisition module, used to acquire abnormal log data of the service grid, wherein the abnormal log data indicates that there is an abnormality in the proxy processing of the service traffic by the service grid; A determination module, configured to determine the abnormal cause corresponding to the abnormal log data based on the abnormal log data and a log database, wherein the log database includes a correspondence between different abnormal log data and the abnormal cause; The determination module is further configured to determine, based on the cause of the exception, an impact factor of the abnormal log data on the service grid; An updating module is used to update the service grid based on the impact factor to remove the abnormal log data.

10. A computer-readable storage medium storing a computer program, wherein the computer program implements the method according to any one of claims 1 to 8 when executed by at least one processor.

Images