A cloud continuous integration alarm processing method, server and storage medium

Abstract

The embodiment of the present invention provides a cloud continuous integration alarm processing method, server and storage medium. The server automatically identifies the immediate alarm corresponding to the inspected code according to the detection execution log of the inspected code on the cloud integration platform, and then determines the target response event corresponding to the immediate alarm by querying the alarm-response mapping table, and automatically triggers the target response event corresponding to the immediate alarm, thereby responding to the immediate alarm and automatically resolving cloud CI related faults, avoiding dependence on manual processing by operation and maintenance personnel, reducing the requirements for operation and maintenance personnel and the burden on the operation and maintenance personnel themselves. Moreover, the response plan provided by the server follows the alarm-response mapping table, and will not be affected by the knowledge, experience and subjectivity of the operation and maintenance personnel themselves like the manual response plan, which is conducive to ensuring the effectiveness of the response plan. At the same time, because of the automatic processing of the server, the processing efficiency is high, which is conducive to reducing the impact on users of the cloud integration platform and maintaining the user experience.

Description

Technical Field

[0001] The present invention relates to the field of communications, and in particular to a cloud continuous integration alarm processing method, a server and a storage medium. Background Art

[0002] In recent years, the cloudification of CI (Continuous Integration) microservices has effectively solved the problems in traditional CI. However, as the industry's requirements for code quality and security continue to increase, it has also brought new challenges to the operation and maintenance monitoring of cloud CI: in order to meet the requirements for code quality and security, more and more detection components and detection services are added to the CI process, resulting in the need for operation and maintenance personnel to manually check the anomaly and start the solution once a cloud integration platform fails. This not only requires operation and maintenance personnel to be familiar with the entire process of cloud CI and understand CI infrastructure and operations, resulting in a high threshold for operation and maintenance personnel, but also requires operation and maintenance personnel to be prepared to deal with failures at any time and be on standby at any time. Moreover, when there are a large number of detection components and detection services in the entire cloud CI process, the efficiency of manual processing is low, resulting in long-term unresponsiveness to alarms, affecting the user experience of the cloud integration platform. Summary of the invention

[0003] The cloud continuous integration alarm processing method, server and storage medium provided in the embodiments of the present invention mainly solve the technical problem that in the related technology, operation and maintenance personnel manually identify and respond to cloud CI alarms, the processing efficiency is low, and the operation and maintenance personnel are burdened with a large amount of money.

[0004] In order to solve the above technical problems, an embodiment of the present invention provides a cloud continuous integration alarm processing method, including:

[0005] Identify the instant alerts corresponding to the inspected code based on the inspection execution log of the inspected code on the cloud integration platform;

[0006] Query the alarm-response mapping table to determine the target response event corresponding to the immediate alarm, where the alarm-response mapping table contains the corresponding relationship between each alarm and each response event;

[0007] The target response event corresponding to the immediate alarm is triggered.

[0008] An embodiment of the present invention further provides a server, the server comprising a processor, a memory and a communication bus;

[0009] The communication bus is used to realize the connection and communication between the processor and the memory;

[0010] The processor is used to execute one or more programs stored in the memory to implement the steps of the above-mentioned cloud continuous integration alarm processing method.

[0011] An embodiment of the present invention also provides a storage medium, characterized in that the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the above-mentioned cloud continuous integration alarm processing method.

[0012] The beneficial effects of the present invention are:

[0013] The cloud continuous integration alarm processing method, server and storage medium provided by the embodiment of the present invention, firstly, the server of the cloud integration platform automatically identifies the immediate alarm corresponding to the inspected code according to the detection execution log of the inspected code in the cloud integration platform, and then by querying the alarm-response mapping table, the corresponding relationship between each alarm and each response event is contained in the alarm-response mapping table, and the target response event corresponding to the immediate alarm can be determined by querying, and then the server can automatically trigger the target response event corresponding to the immediate alarm, thereby responding to the immediate alarm, automatically solving cloud CI related faults, avoiding dependence on manual processing of operation and maintenance personnel, reducing the requirements for operation and maintenance personnel and the burden of the operation and maintenance personnel themselves. Moreover, the response scheme provided by the server follows the alarm-response mapping table, and will not be affected by the knowledge, experience and subjectivity of the operation and maintenance personnel themselves like the manual response scheme, which is conducive to ensuring the effectiveness of the response scheme. At the same time, because of the automatic processing of the server, the immediate alarm can be identified and the corresponding response can be made as soon as the detection execution log of the inspected code is obtained, and the processing efficiency is high, which is conducive to reducing the impact on the users of the cloud integration platform and maintaining the user experience.

[0014] Other features and corresponding beneficial effects of the present invention are described in the latter part of the specification, and it should be understood that at least part of the beneficial effects become obvious from the description in the specification of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of the cloud continuous integration alarm processing method provided in the first embodiment of the present invention;

[0016] Figure 2 A schematic diagram of a pipeline detection link shown in the first embodiment of the present invention;

[0017] Figure 3 A schematic diagram of a typical cloud CI framework provided in the second embodiment of the present invention;

[0018] Figure 4 A schematic diagram of a cloud continuous integration alarm processing system provided in Embodiment 2 of the present invention;

[0019] Figure 5 A schematic diagram of the hardware structure of the server provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the following is a further detailed description of the embodiments of the present invention through specific implementation methods combined with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0021] Embodiment 1:

[0022] In order to solve the problem that in the related art, the abnormality of the cloud CI process can only be manually identified and then the response event is initiated, the processing efficiency is low, and the burden on operation and maintenance personnel is heavy, this embodiment provides a cloud continuous integration alarm processing method, which can be executed by the server of the cloud integration platform. Figure 1 The flowchart shown:

[0023] S102: The server identifies an immediate alarm corresponding to the inspected code according to the inspection execution log of the inspected code on the cloud integration platform.

[0024] When a code submitter submits a code to the cloud integration platform, the code needs to be subjected to various detection scans of the cloud integration platform as the inspected code, such as check in pipeline, Jenkins pipeline, UT (Unit Testing), code static inspection, compilation build, etc. The quality and security of the inspected code can be determined through the detection scan of the cloud integration platform. The cloud integration platform detects each inspected code in a pipeline-style manner, and multiple detection items are set in the detection link, such as Figure 2 As shown, development detection project 21, configuration detection project 22, integration detection project 23, build detection project 24, test detection project 25, deployment detection project 26 and container cloud detection project 27 are arranged in sequence on the detection link. When a code submitter submits a code to the cloud integration platform, the code will be scanned by the development detection project 21, configuration detection project 22...container cloud detection project 27 in sequence as the inspected code.

[0025] In the process of each detection project on the cloud integration platform performing detection scanning on the inspected code, the cloud integration platform will generate a detection execution log according to the process of executing the detection scan. In the execution log corresponding to a certain detection project, basic information such as the detection start time, detection end time, status information and keywords of the inspected code under the detection project are usually recorded. It should be understood by those skilled in the art that the basic information in the detection execution log can also include other contents in addition to the above-mentioned ones, for example, including the time when the inspected code enters the detection project queue, etc. When a piece of inspected code passes the detection scan of all detection projects, the cloud integration platform will obtain a full-link execution log for the inspected code, and the full-link execution log contains the basic information of the inspected code being detected in each detection project.

[0026] The server can identify the immediate alarm corresponding to the inspected code based on the detection execution log of the inspected code on the cloud integration platform. It should be understood that the server identifies the alarm corresponding to the inspected code, which is actually identifying the abnormality of the inspected code during the detection and scanning process. The so-called "immediate alarm" is actually the abnormal alarm found during this detection and scanning process. In addition to the "immediate alarm", there may be some "potential alarms" that have not yet been discovered but may already exist.

[0027] In some examples of this embodiment, the server can query the detection execution log of the detected code to determine the status information of the execution of the detected code under each detection item. Generally, the status information is divided into two types, one is execution success and the other is execution failure.

[0028] For each detection item in the detection link, if the status information in the detection execution log of the detected code under the detection item indicates that the detected code is successfully executed under the detection item, the server can determine the execution index of the detected code under the detection item based on the detection execution log. The so-called execution index may include the execution time of the detected code under the detection item, etc. After determining the execution index, the server can compare the execution index with the corresponding execution index threshold, so as to identify whether the detected code has an immediate alarm in the detection item and what immediate alarms exist.

[0029] If the status information in the detection execution log of the detected code under the detection project indicates that the detected code failed to execute under the detection project, the server can extract keywords in the detection execution log and then use a pre-set keyword-alarm mapping table to determine the immediate alarm corresponding to the extracted keyword.

[0030] In some examples of this embodiment, after the server queries the detection execution log of the inspected code and determines the execution status information of the inspected code under each detection item, the server can also display the queried status information, so that operation and maintenance personnel can understand the detection scanning status of the inspected code through display screens when necessary.

[0031] S104: The server queries the alarm-response mapping table to determine the target response event corresponding to the immediate alarm.

[0032] After determining the alarm, the server can query the alarm-response mapping table, which stores the correspondence between each alarm and each response event. Therefore, the server can query the response event corresponding to the immediate alarm in the alarm-response mapping table by using the immediate alarm as the key value. It can be understood that in the alarm-response mapping table, an alarm should correspond to at least one response event. In some cases, an alarm can also correspond to multiple response events at the same time. However, different alarms can correspond to the same response event. For example, the processing solutions for alarm a and alarm c are the same, so in the alarm-response mapping table, the response events corresponding to alarm a and alarm c are also the same.

[0033] A response event is actually a solution for solving the abnormality corresponding to the alarm. Usually, multiple operations can be included in a solution, so multiple operations can also be included in a response event. In some examples of this embodiment, a response event can be composed of N sub-response events, where N is greater than or equal to 1. Each sub-response event is the minimum operation performed by the server, and the operation cannot be further decomposed and refined. For example, a sub-response event can be "restarting the cloud CI server", "enabling the backup gerrit mirror", "email notification code submitter", etc. There is no doubt that different combinations of sub-response events can constitute different response events. The relationship between sub-response events and response events is similar to the relationship between microscopic atoms and macroscopic matter.

[0034] It is obvious that if an immediate alarm corresponds to only one response event in the alarm-response mapping table, then this response event is the target response event of the immediate alarm. However, when an immediate alarm corresponds to more than one response event in the alarm-response mapping table, because the relationship between the response events corresponding to the same alarm in the alarm-response mapping table is a parallel optional relationship, the server does not need to use all the response events corresponding to the immediate alarm as the target response event, but only needs to select one from these response events.

[0035] In some examples, if there are at least two response events corresponding to the immediate alarm in the alarm-response mapping table, the server can determine which response event to select as the target response event of the immediate alarm based on the cost required to execute each response event respectively. Of course, the server will select the one with a relatively low cost as the target response event. When evaluating the cost of each response event, the server can consider the cost of executing the response event in at least one of the aspects of human cost, event cost and resource cost. In some examples of this embodiment, the server can make a choice based on the above costs.

[0036] Because the cost assessments made by the server before actually triggering a response event are only estimates, in this embodiment, the costs of each response event corresponding to the immediate alarm determined before the server selects the target response event are called "estimated costs." Optionally, when separately determining the estimated costs of each response event corresponding to the immediate alarm, the server can first separately determine the key cost indicator X corresponding to each response event. For example, the key cost indicator can be at least one of the code pull time, verifyCI time (verify CI), mergeCI time, product library upload time, labor cost, resource cost, etc. Then, the server determines the estimated cost corresponding to each response event according to the following formula:

[0037]

[0038] Among them, Y is the estimated cost corresponding to a response event, n is the total number of key cost indicators X corresponding to the response event, i is the i-th key cost indicator corresponding to the response event, and X i represents the value of the i-th key cost indicator, W i Indicates the weight of the i-th key cost indicator. For example, assuming that the key cost indicators of a response event include code pull time, verifyCI time, and mergeCI time, then the value of n is 3, X1 represents the value of code pull time, W1 represents the weight corresponding to the key cost indicator of code pull time...X3 represents the value of mergeCI time, and W3 represents the weight corresponding to the key cost indicator of mergeCI time.

[0039] In some embodiments, after the server determines the estimated cost of each response event corresponding to the immediate alert, it selects a response event with the lowest estimated cost as the target response event of the immediate alert. Of course, in some other examples, the server does not only consider the size of the estimated cost when selecting the target response event. For example, the server may further consider factors such as the thoroughness or effectiveness of each response event in handling the exception.

[0040] S106: The server triggers a target response event corresponding to the immediate alarm.

[0041] After the server determines the target response event corresponding to the immediate alarm, it can trigger the target response event. Specifically, the server triggers and executes each sub-response event in the target response event.

[0042] It can be understood that after the server triggers and executes the target response event, it can determine the "actual cost" corresponding to the target response event. Using the actual cost of the target response event, the server can use a logistic regression algorithm to determine the weight W corresponding to each key cost indicator. Therefore, in this embodiment, the weight value corresponding to each updated key cost indicator is not fixed in the long term.

[0043] In some examples of the present embodiment, after the server determines the execution index of the inspected code under each detection project according to the detection execution log, it can also generate a code submission record for the current submission of the inspected code. For example, the server generates a code submission record for the current submission of the inspected code according to the order in which the inspected code undergoes each detection project. The code submission record may include but is not limited to the following information: commit ID (submission identification) of the inspected code, submitter identification information, cloud continuous integration initial teaming time, KW (klocwork) scan details, SM (Source monitor, resource monitoring) scan details, verifyCI duration, mergeCI duration. In some other examples of the present embodiment, if the inspected code has an immediate alarm in the detection link, the code submission record may further include an alarm code corresponding to the immediate alarm and a corresponding target response event. After the server generates the code submission record, the code submission record can be stored in a database, which can be set in a storage control local to the server or on other storage devices.

[0044] The code submission records stored in the database can be used by the server for overall analysis, so as to analyze the "potential alarms" that have not yet been identified but may already have problems. In one embodiment of the present embodiment, the server can perform execution indicator fitting based on the code submission records stored in the database, and then determine the potential alarms for the cloud integration platform based on the fitting results. For example, the server can calculate the average verifyCI duration based on the verifyCI duration in each code submission record, and calculate the average mergeCI duration based on the mergeCI duration in each code submission record. Then, determine whether there is an abnormality in the verifyCI detection item based on the verifyCI average duration, and determine whether there is an abnormality in the mergeCI detection item based on the mergeCI average duration. Identifying potential alarms helps to make early predictions and interventions in cloud CI scenarios to prevent deep deterioration of indicators.

[0045] On the other hand, the server can also regress and statistically analyze historical information based on the code submission records to obtain the verifyCI success rate, mergeCI success rate, average verifyCI time, average mergeCI time, etc., and then use these statistical values ​​to update the relevant execution indicator thresholds.

[0046] The cloud continuous integration alarm processing method provided in this embodiment is that the server executes the cloud continuous integration alarm processing process, automatically identifies the immediate alarm corresponding to the inspected code based on the detection execution log of the inspected code on the cloud integration platform, and then determines the target response event that can solve the immediate alarm based on the correspondence between the preset alarm and the response event, and automatically triggers the target response event, so that the abnormality can be quickly located and eliminated, avoiding the problem of low efficiency of manual processing; and the determination of the target response event is not affected by the knowledge, personal experience, temporary adaptability, etc. of the operation and maintenance personnel, ensuring the effectiveness and reliability of the response plan. At the same time, the cloud continuous integration alarm processing method can reduce the requirements for operation and maintenance personnel and reduce the pressure on operation and maintenance personnel.

[0047] Embodiment 2:

[0048] In order to make those skilled in the art more aware of the advantages and details of the cloud continuous integration alarm processing method, this embodiment will continue to explain the solution with specific examples:

[0049] Figure 3The diagram shows a typical cloud CI framework: In the cloud CI process, after the developer submits the code to the gerrit code base 30, the cloud CI checkin pipeline 32 can be automatically triggered. The Jenkins pipeline concurrently executes unit testing, code static checking, compilation and build jobs. After the job is successful, the compiled version will be submitted to the artifact library at the service granularity. With the development of devops+ (a combination of Development and Operations, i.e., integrated development and operation), related static code scanning, compliance governance, security monitoring and other requirements have also been developed and added to the cloud CI detection link. For example, in Figure 3 There are five types of scanning checks: UT, SM, KW, Coverity, and Package. Therefore, in the actual process, five matching dockers (application container engines) are needed to support it.

[0050] from Figure 3 It can be seen that behind a very simple code submission for the cloud integration platform user (code submitter), there are multiple or even dozens of calls and feedbacks from Cloud CI. Once a fault occurs, it must be quickly located and fed back, otherwise it will cause a large number of user complaints and affect the user experience:

[0051] This embodiment provides a cloud continuous integration alarm processing system corresponding to the cloud continuous integration alarm processing method in embodiment 1. Figure 4 A schematic diagram of the cloud continuous integration alarm processing system is shown below:

[0052] The cloud continuous integration alarm processing system 4 includes a database 400, a log processing module 402, a status display module 404, a log application module 406, and an alarm knowledge base 408. The log processing module 402 obtains and processes the detection execution log of the inspected code on the cloud integration platform, queries the detection execution log status in real time, and feeds back the query results of the status information to the status display module 404 for display.

[0053] At the same time, if the status information indicates that the inspected code is successfully executed under a certain detection project, the log processing module 402 can determine the execution index of the inspected code under the detection project according to the detection execution log. In addition, the log processing module 402 can also generate a code submission record for the current submission of the inspected code after determining the execution index of the inspected code under each detection project according to the detection execution log, and store the code submission record in the database 400. For example, the log processing module 402 generates a code submission record for the current submission of the inspected code according to the order in which the inspected code undergoes each detection project. The code submission record may include the commit ID of the inspected code, the submitter identification information, the initial entry time of the cloud continuous integration, the KW scan details, the SM scan details, the verifyCI duration, the mergeCI duration, etc. The code submission records stored in the database can be used for statistical analysis by the log application module 406 to perform execution index fitting, thereby identifying potential alarms.

[0054] Subsequently, the log application module 406 can compare the execution index with the corresponding execution index threshold, thereby identifying whether the inspected code has an immediate alarm in the detection project and which immediate alarms exist. If the status information queried by the log processing module 402 indicates that the inspected code has failed to execute under a certain detection project, the log application module 406 can extract keywords from the detection execution log, and then use a preset keyword-alarm mapping table to determine the immediate alarm corresponding to the extracted keyword.

[0055] After the log application module 406 identifies the immediate alarm, it determines the target response event and triggers the corresponding target response event according to the alarm-response mapping table provided by the alarm knowledge base 408. The alarm-response mapping table is actually a set of message mapping MAP<R,A> , where R represents alarm and A represents response event. The alarm knowledge base 408 forms alarm rules through artificial knowledge aggregation, alarm classification and typical log analysis in the early stage, and the operation and maintenance personnel provide conventional processing solutions corresponding to each alarm as a matching response event. A response event can be composed of one or more sub-response events. In some examples of this embodiment, the content in the alarm knowledge base 408 can also be learned and optimized based on historical data.

[0056] The following describes the details of specifying the alarm-response mapping table:

[0057] In an example of this embodiment, the alarm-response mapping table R includes an alarm code and a corresponding alarm content. In some examples, the first digit in the alarm code is used to characterize the abnormality category corresponding to the alarm, and the second and third digits are used to identify the abnormality subcategory of the specific link of the cloud CI. For example, the following Table 1 shows the corresponding relationship between the values ​​of the first digit in the alarm code and the abnormality category:

[0058] Table 1

[0059] The first digit value of the alarm code The abnormal category to which the alarm belongs 0 Unknown Type 1 Configuration Error 2 Human error 3 Gerrit server issues 4 Product library problem 5 mvn proxy repository problem 6 Virtual machine environment issues 7 Code Problem

[0060] Table 2 shows the corresponding relationship between the second and third digits of the alarm code and the abnormal sub-categories:

[0061] Table 2

[0062]

[0063]

[0064] For example, combining Table 1 and Table 2, we can see that "210" means that the branch in the configuration stage is locked, "012" means an unknown error in the initial stage, "416" means the artifact library return code in the artifact library component, "403" means access denied, which is generally a network problem, and "418" means that the artifact library link has timed out or crashed.

[0065] The following describes the settings of response events in the alarm-response mapping table: Optionally, a response event can be formed by "atomicizing" the sub-response events. Specifically, a sub-response event is provided first, and each sub-response event has a different impact factor. The impact factor here can be any value between [0,1], which represents the impact of the sub-response event on the CI process, environment, and CI service capabilities. For example, see Table 3:

[0066] Table 3

[0067]

[0068] Each response event may be composed of one or more sub-response events mentioned above, and the impact factor of the synthesized response event is the sum of the impact factors of each sub-response event.

[0069] Since one alarm in the alarm-response mapping table can correspond to multiple response events at the same time, the process of selecting the target response event by the log application module 406 is introduced below:

[0070] The log application module 406 can first determine the key cost indicators X corresponding to each response event, such as code pull time, verifyCI time, mergeCI time, product library upload time, labor cost, resource cost, etc. Then, the log application module 406 determines the estimated cost corresponding to each response event according to the following formula:

[0071]

[0072] Among them, Y is the estimated cost corresponding to a response event, n is the total number of key cost indicators X corresponding to the response event, i is the i-th key cost indicator corresponding to the response event, and X i represents the value of the i-th key cost indicator, W i Represents the weight of the i-th key cost indicator.

[0073] In this embodiment, after the log application module 406 determines the estimated cost of each response event corresponding to the immediate alarm, it selects a response event with the lowest estimated cost as the target response event of the immediate alarm. After the log application module 406 determines the target response event corresponding to the immediate alarm, it can trigger the target response event.

[0074] The cloud continuous integration alarm processing system provided in this embodiment can effectively identify abnormal alarms in the cloud CI link and trigger corresponding response events according to the alarm rules, thereby ensuring the robustness of the cloud CI environment, reducing the workload of operation and maintenance personnel, and improving user satisfaction.

[0075] Embodiment three:

[0076] The present embodiment provides a storage medium, which may store one or more computer programs that can be read, compiled and executed by one or more processors. In the present embodiment, the storage medium may store a cloud continuous integration alarm processing program, and the cloud continuous integration alarm processing program can be executed by one or more processors to implement the process of any one of the cloud continuous integration alarm processing methods introduced in the aforementioned embodiments.

[0077] This embodiment also provides a server, such as Figure 5 As shown: the server 5 includes a processor 51, a memory 52 and a communication bus 53 for connecting the processor 51 and the memory 52, wherein the memory 52 can be the storage medium storing the cloud continuous integration alarm processing program. The processor 51 can read the cloud continuous integration alarm processing program, compile and execute the process of implementing the cloud continuous integration alarm processing method described in the above embodiment:

[0078] Processor 51 identifies the immediate alarm corresponding to the inspected code based on the detection execution log of the inspected code on the cloud integration platform, and then queries the alarm-response mapping table to determine the target response event corresponding to the immediate alarm. The alarm-response mapping table contains the correspondence between each alarm and each response event; finally, processor 51 triggers the target response event corresponding to the immediate alarm.

[0079] In some examples of this embodiment, when the processor 51 queries the alarm-response mapping table to determine the target response event corresponding to the immediate alarm, if there are at least two response events corresponding to the immediate alarm in the alarm-response mapping table, the processor 51 determines the estimated cost corresponding to each response event corresponding to the immediate alarm, and the estimated cost is the estimated cost of executing the response event in at least one of the human cost, event cost and resource cost. Then, the processor 51 selects the target response event corresponding to the immediate alarm according to the estimated cost of each response event.

[0080] In one example, when the processor 51 determines the estimated costs corresponding to each response event corresponding to the immediate alarm, for any response event corresponding to the immediate alarm, the processor 51 determines each key cost indicator X corresponding to the response event; and then determines the estimated cost corresponding to the response event according to the following formula:

[0081]

[0082] Among them, Y is the estimated cost corresponding to the response event, n is the total number of key cost indicators X corresponding to the response event, i is the i-th key cost indicator corresponding to the response event, and X i represents the value of the i-th key cost indicator, W i Represents the weight of the i-th key cost indicator.

[0083] In some examples of this embodiment, after the processor 51 triggers the target response event corresponding to the immediate alarm, it also determines the actual cost of the response event corresponding to the triggered alarm, and then updates the weight W corresponding to each key cost indicator based on the actual cost combined with the logistic regression algorithm.

[0084] Optionally, when the processor 51 identifies the immediate alarm corresponding to the inspected code based on the detection execution log of the inspected code on the cloud integration platform, it can first query the detection execution log of the inspected code to determine the status information of the execution of the inspected code under each detection project. For any detection project, if the status information indicates that the inspected code is successfully executed under the detection project, the processor 51 determines the execution index of the inspected code under the detection project based on the detection execution log, and compares the execution index of the inspected code under each detection project with the corresponding execution index threshold to identify the immediate alarm in each detection project.

[0085] For any detection item, if the status information indicates that the detected code fails to execute under the detection item, the processor 51 extracts the keywords in the detection execution log, and then queries the keyword-alarm mapping table to determine the immediate alarm corresponding to the keyword. The keyword-alarm mapping table contains mapping relationships between multiple keywords and multiple alarms.

[0086] In some examples of this embodiment, after the processor 51 determines the execution index of the inspected code under each detection item according to the detection execution log, it will also generate a code submission record for the current submission of the inspected code according to the order in which the inspected code undergoes each detection item, and store the code submission record in the database. Optionally, the code submission record includes at least part of the submission ID of the inspected code, the submitter identification information, the initial entry time of the cloud continuous integration, the KW scan details, the SM scan details, the verifyCI duration, and the mergeCI duration.

[0087] In some examples of this embodiment, the processor 51 may also perform execution indicator fitting according to the code submission records stored in the database, and then determine potential alarms for the cloud integration platform based on the fitting result analysis.

[0088] Obviously, those skilled in the art should understand that all or some steps, systems, and functional modules / units in the above disclosed methods can be implemented as software (which can be implemented with program code executable by a computing device), firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component can have multiple functions, or a function or step can be performed by several physical components in cooperation. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, executed by a computing device, and in some cases, the steps shown or described can be performed in a different order than herein, and the computer-readable medium can include a computer storage medium (or a non-transitory medium) and a communication medium (or a transient medium). As is well known to those of ordinary skill in the art, the term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storing information (such as computer readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and may include any information delivery medium. Therefore, the present invention is not limited to any particular combination of hardware and software.

[0089] The above contents are further detailed descriptions of the embodiments of the present invention in combination with specific implementation methods, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A cloud continuous integration alarm processing method, comprising: Identify the instant alarm corresponding to the inspected code according to the inspection execution log of the inspected code on the cloud integration platform; Querying an alarm-response mapping table to determine a target response event corresponding to the immediate alarm, wherein the alarm-response mapping table includes a correspondence between each alarm and each response event; If there are at least two response events corresponding to the immediate alarm in the alarm-response mapping table, then for any response event corresponding to the immediate alarm, determine each key cost indicator X corresponding to the response event; The estimated cost corresponding to the response event is determined according to the following formula: The Y is the estimated cost corresponding to the response event, the n represents the total number of key cost indicators X corresponding to the response event, the i represents the i-th key cost indicator corresponding to the response event, and the X i represents the value of the i-th key cost indicator, W i represents the weight of the i-th key cost indicator, wherein the estimated cost is the estimated cost of executing the response event in at least one of the aspects of human cost, event cost and resource cost; Selecting a target response event corresponding to the immediate alarm according to the estimated cost of each of the response events; A target response event corresponding to the immediate alarm is triggered.

2. The cloud continuous integration alarm processing method according to claim 1, characterized in that: After the target response event corresponding to the immediate alarm is triggered, the method further includes: Determining the actual cost of the response event corresponding to triggering the alarm; The weight W corresponding to each key cost indicator is updated according to the actual cost in combination with the logistic regression algorithm.

3. The cloud continuous integration alarm processing method according to claim 1 or 2, characterized in that: The step of identifying the instant alarm corresponding to the inspected code according to the detection execution log of the inspected code on the cloud integration platform includes: Querying the detection execution log of the detected code to determine the execution status information of the detected code under each detection item; For any detection item, if the status information indicates that the detected code is successfully executed under the detection item, then determining the execution index of the detected code under the detection item according to the detection execution log; The execution index of the inspected code under each of the inspection items is compared with the corresponding execution index threshold, and the immediate alarm in each of the inspection items is identified.

4. The cloud continuous integration alarm processing method according to claim 3, characterized in that: The step of identifying the instant alarm corresponding to the inspected code according to the detection execution log of the inspected code on the cloud integration platform also includes: For any detection item, if the status information indicates that the detected code fails to execute under the detection item, extracting keywords from the detection execution log; The keyword-alarm mapping table is queried to determine the immediate alarm corresponding to the keyword, wherein the keyword-alarm mapping table contains mapping relationships between multiple keywords and multiple alarms.

5. The cloud continuous integration alarm processing method according to claim 3, characterized in that: After determining the execution index of the inspected code under each of the inspection items according to the inspection execution log, the method further includes: Generate a code submission record for the current submission of the inspected code in the order in which the inspected code undergoes each inspection item, wherein the code submission record includes at least part of the submission identifier commitID of the inspected code, the submitter identification information, the time of first joining the cloud continuous integration team, the KW scanning details, the SM scanning details, the verifyCI duration, and the mergeCI duration; The code submission record is stored in a database.

6. The cloud continuous integration alarm processing method according to claim 5, characterized in that: The cloud continuous integration alarm processing method also includes: Performing execution indicator fitting according to the code submission records stored in the database; Potential alarms for the cloud integration platform are determined based on the fitting result analysis.

7. A server, comprising a processor, a memory and a communication bus; The communication bus is used to realize the connection and communication between the processor and the memory; The processor is used to execute one or more programs stored in the memory to implement the steps of the cloud continuous integration alarm processing method according to any one of claims 1 to 6.

8. A storage medium, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the cloud continuous integration alarm processing method according to any one of claims 1 to 6.

Images