CN108519920A

CN108519920A - A kind of scheduling retries method and device

Info

Publication number: CN108519920A
Application number: CN201810211538.5A
Authority: CN
Inventors: 何青松; 周仕成; 王新盟; 李联邦
Original assignee: Koubei Shanghai Information Technology Co Ltd
Current assignee: Koubei Shanghai Information Technology Co Ltd
Priority date: 2018-03-14
Filing date: 2018-03-14
Publication date: 2018-09-11
Anticipated expiration: 2038-03-14
Also published as: CN108519920B

Abstract

The application provides a kind of dispatch and retries method, the method includes：Specific abnormal conditions when handling abnormal according to job scheduling determine that job scheduling handles Exception Type：Exception Type is handled according to identified job scheduling, it is determined whether is needed to carry out scheduling next time and is retried；And if the determination result is YES, then further comprise obtaining scheduling strategy corresponding with job scheduling processing Exception Type；If desired it is dispatched and is retried next time, then dispatched next time the time retried using scheduling strategy determination, and executed scheduling next time when reaching the time and retry.Accordingly, the technical solution of the application is abnormal for different work dispatch deal, takes different scheduling strategies, improves the abnormal efficiency of processing, saves system resource when scheduling retries.

Description

A kind of scheduling retries method and device

Technical field

This application involves computer scheduling method fields, and in particular to a kind of scheduling retries method and device.The application is also It is related to a kind of for dispatching the electronic equipment retried and a kind of computer readable storage medium.

Background technology

In the large-scale network architecture, a business action request generally requires to complete by multiple systems calling, Including calling, data calculating, data storage etc. between system than relatively time-consuming operation.If calling abnormal, business between encountering system Situations such as data exception, interface call time-out, initiation system needs, which are scheduled, to be retried, to continue to promote the action of execution business Request.Scheduling retry strategy directly determine business action how long can be advanced to successfully, dispatch consumption system how many resource retried Etc. practical problems.Therefore, scheduling retries the selection of strategy for promoting execution business action request successful execution in close relations.

In the prior art, scheduling retries the mode that strategy uses equal difference time interval, copes with all abnormal scenes.It should Method is there are apparent defect, because the interval time at equal difference interval cannot reasonably determine that time interval is too small, wastes and is System resource, time interval then increase be advanced to successful time cost greatly, can not be dispatched for some and retry successful exception It is scheduled to retry and also be not necessarily to.For example, the abnormal Scene dispatch in part cannot be advanced to success forever, manpower intervention solution is needed Certainly, abnormal be scheduled in this way is retried with regard to being not necessarily to.For another example, the abnormal scene in part cannot be advanced to success the short time, Therefore the scheduling in the short time, which retries, can waste time and system resource.In addition, there is some in a short time may be used extremely Success is retried with scheduling, under equal difference time interval strategy, scheduling weight next time may be just carried out by longer time interval Examination, it is less efficient.

Invention content

The application provides a kind of dispatch and retries method this method, it is intended to solve existing scheduling and retry the abnormal efficiency of method processing When low, scheduling retries the problem of waste system resource.

A kind of scheduling provided by the present application retries method this method, including：It is specific when handling abnormal according to job scheduling Abnormal conditions determine that job scheduling handles Exception Type；Exception Type is handled according to identified job scheduling, it is determined whether is needed It is dispatched and is retried next time；And if the determination result is YES, then further comprise obtaining different with job scheduling processing The corresponding scheduling strategy of normal type；If desired it is dispatched and is retried next time, then determined next time using the scheduling strategy It dispatches the time retried, and executes scheduling next time when reaching the time and retry.

Optionally, described to determine that job scheduling handles exception class according to specific abnormal conditions of job scheduling processing when abnormal Before type, further include：After the action request for receiving action initiator, the information of request server checkout action initiator； After the information for acting initiator for receiving server transmission is by the information of verification, to action, initiator confirms current action Ask successful execution；According to the action request, the job scheduling of operation needed for the action request is carried out.

Optionally, the job scheduling processing Exception Type includes at least following one kind：Solution can not be retried by scheduling Exception, rely on that operation processing is abnormal, accidental processing is abnormal, system exception or calls time-out exception.

Optionally, if job scheduling processing Exception Type is the exception that can not retry solution by scheduling, Job scheduling determined by the basis handles Exception Type, it is determined whether it needs to carry out in the step of scheduling retries next time, Judging result is no；And it further performs the step of：Execute alarm operation.

Optionally, if job scheduling processing Exception Type is system exception or calls time-out abnormal, the basis Identified job scheduling handles Exception Type, it is determined whether it is yes that needs are dispatched the judging result retried next time；Institute It is that Fibonacci passs ordered series of numbers strategy to state scheduling strategy, i.e., passs ordered series of numbers determination according to Fibonacci and dispatch next time between the time retried Every.

Optionally, if job scheduling processing Exception Type is that accidental processing is abnormal, make determined by the basis Industry dispatch deal Exception Type, it is determined whether it is yes that needs are dispatched the judging result retried next time；The scheduling strategy Short time every strategy, as dispatch next time the time interval that retries be the short time every.

Optionally, described according to determining if job scheduling processing Exception Type is that rely on operation processing abnormal Job scheduling handle Exception Type, it is determined whether to be dispatched the judging result retried next time be yes to needs；The scheduling Strategy for for a long time every strategy, i.e., dispatch next time the time interval that retries be for a long time every.

Optionally, after execution alarm operation, further include：The different of solution can not be retried described in determination by scheduling Often by Account Closure when causing, the credit information of the corresponding user of the account is obtained, the credit information includes credit Grade, credit score；According to preset credit management rule, credit score, the credit grade of the user are adjusted.

Optionally, described to be dispatched next time the time retried using scheduling strategy determination, and when reaching the time Scheduling next time is executed to retry specially：Strategy determination is passed using the Fibonacci and dispatches the time retried next time；It determines Pass whether the time dispatched and retried next time determined by ordered series of numbers is more than preset time interval threshold values according to Fibonacci；If not It is more than then to execute scheduling next time when reaching the time to retry.

Optionally, described to be dispatched next time the time retried using scheduling strategy determination, and when reaching the time Scheduling next time is executed to retry specially：The time retried is dispatched using the short time next time every strategy determination；To described The execution counting how many times for dispatching the time step retried next time are determined every strategy using the short time；In the execution number When being less than preset number threshold values, executes scheduling next time and retry.

Optionally, it is determined using the scheduling strategy described before dispatching the time retried next time, further includes：True The fixed dependence operation processing Exception Type is to obtain account freezing information, the account freezing by account freezing and when causing Information includes that account freezes the time limit；Freeze whether the time limit is less than and preset freezes time limit threshold values described in determination；If so, executing next Step.

Optionally, freeze described in the determination time limit whether be less than it is preset freeze time limit threshold values after, held described Before the step of row next step, further include：When by the account freezing time limit, whether detection account freezing has released；If It is to execute next step.

Optionally, before dispatching the time retried next time using scheduling strategy determination, further include：Determining It states and relies on operation processing Exception Type when being caused by account insufficient funds, obtain the associated account number letter of the account of insufficient funds Breath；Using the associated account number as dispatching the object retried next time.

Optionally, the job scheduling processing is abnormal generates with the next stage：Local data processing stage or/and calling are remote Journey system processing stage.

Optionally, the accidental processing Exception Type, including handled caused by network jitter abnormal.

The application also proposes that a kind of scheduling retries device, which is characterized in that described device includes：Exception Type determines single Member, specific abnormal conditions when being configured to handle abnormal according to job scheduling determine that job scheduling handles Exception Type；Scheduling Tactful acquiring unit is configured to handle Exception Type according to identified job scheduling, it is determined whether needs to carry out next time Scheduling retries；And if the determination result is YES, then further comprise obtaining corresponding with job scheduling processing Exception Type Scheduling strategy；Scheduling retries unit, is configured to if desired be dispatched next time and retries, then uses the scheduling strategy true It fixs the time once dispatched and retried, and executes scheduling next time when reaching the time and retry.

Optionally, described device further includes：Request unit is verified, is configured to ask in the action for receiving action initiator After asking, the information of request server checkout action initiator；Confirmation unit is configured in the action for receiving server transmission After the information that the information of initiator passes through verification, current action request successful execution is confirmed to action initiator；Scheduling unit, It is configured to, according to the action request, carry out the job scheduling of operation needed for the action request.

Optionally, described device further includes：Execution unit, if it is institute to be configured to the job scheduling processing Exception Type The exception of solution can not be retried by scheduling by stating, then job scheduling determined by the basis handles Exception Type, it is determined whether It needs to carry out in the step of scheduling retries next time, after judging result is no, executes following steps：Execute alarm operation.

Optionally, it is if the scheduling strategy acquiring unit is further configured to the job scheduling processing Exception Type System exception calls time-out abnormal, then job scheduling determined by the basis handles Exception Type, it is determined whether need into It is yes that row dispatches the judging result retried next time；The scheduling strategy is that Fibonacci passs ordered series of numbers strategy, i.e., according to striking Poona Contract passs ordered series of numbers determination and dispatches the time interval retried next time.

Optionally, it is if the scheduling strategy acquiring unit is further configured to the job scheduling processing Exception Type Accidental processing is abnormal, then job scheduling determined by the basis handles Exception Type, it is determined whether needs are adjusted next time It is yes to spend the judging result retried；The scheduling strategy short time, as dispatching the time interval retried next time was every strategy Short time every.

Optionally, it is if the scheduling strategy acquiring unit is further configured to the job scheduling processing Exception Type It is abnormal to rely on operation processing, then job scheduling processing Exception Type determined by the basis, it is determined whether need to carry out next The secondary judging result retried of dispatching is yes；The scheduling strategy is between dispatching the time retried next time every strategy for a long time Be divided into for a long time every.

Optionally, described device further includes：Credit information acquiring unit is configured to that scheduling can not be passed through described in determination The exception of solution is retried by Account Closure and when causing, obtains the credit information of the corresponding user of the account, the credit Information includes credit grade, credit score；Credit information adjustment unit is configured to, according to preset credit management rule, adjust The credit score of the whole user, credit grade.

Optionally, the scheduling retries unit and is configured specifically for：Strategy is passed using the Fibonacci to determine next time Dispatch the time retried；Determine whether the time for being passed according to Fibonacci and dispatching and retrying next time determined by ordered series of numbers is more than default Time interval threshold values；It is retried if not exceeded, then executing scheduling next time when reaching the time.

Optionally, the scheduling retries unit and is configured specifically for：It is adjusted next time every strategy determination using the short time Spend the time retried；To the execution number for dispatching the time step retried next time every strategy determination using the short time It counts；When the execution number is less than preset number threshold values, executes scheduling next time and retry.

Optionally, described device further includes account freezing information acquisition unit, is configured to determining the dependence operation Processing Exception Type is to obtain account freezing information by account freezing and when causing, and the account freezing information includes that account is frozen Tie the time limit；Freeze time limit judging unit, be configured to determine described in freeze whether the time limit is less than and preset freezes time limit threshold values；If It is that determination executes next step.

Optionally, described device further includes：Detection unit when being configured to by the account freezing time limit, detects account Number freeze whether to have released；If so, determination executes next step.

Optionally, described device further includes interlock account information acquisition unit, is configured to determining the dependence operation When processing Exception Type is caused by account insufficient funds, the associated account number information of the account of insufficient funds is obtained；Scheduling retries Object determination unit is configured to using the associated account number as dispatching the object retried next time.

The application also proposes a kind of for dispatching the electronic equipment retried, which is characterized in that the electronic equipment includes：It is aobvious Show device；Processor；Memory, the program retried for storing scheduling, described program are held when reading execution by the processor The following operation of row：Specific abnormal conditions when handling abnormal according to job scheduling determine that job scheduling handles Exception Type；According to Identified job scheduling handles Exception Type, it is determined whether needs to carry out scheduling next time and retries；And if judging result is It is then to further comprise obtaining scheduling strategy corresponding with job scheduling processing Exception Type；If desired it carries out next Secondary scheduling retries, then is dispatched next time the time retried using scheduling strategy determination, and when reaching the time under execution Primary scheduling retries.

The application also proposes a kind of computer readable storage medium, is stored thereon with scheduling and retries program, which is characterized in that When the program is read out by the processor execution, following operation is executed：Specific abnormal conditions when handling abnormal according to job scheduling, really It is set for industry dispatch deal Exception Type；Exception Type is handled according to identified job scheduling, it is determined whether needs to carry out next Secondary scheduling retries；And if the determination result is YES, then further comprise obtaining opposite with job scheduling processing Exception Type The scheduling strategy answered；If desired it is dispatched and is retried next time, then determine what scheduling next time retried using the scheduling strategy Time, and execute scheduling next time when reaching the time and retry.

It is provided by the present application to dispatch the technical solution retried, for different types of exception, it is determined whether it is necessary to carry out Scheduling retry and it is determine that it is necessary to be scheduled retry when obtain scheduling retry strategy.When the technical solution must not in determination It is scheduled when retrying, does not just execute scheduling and retry, avoid to this It is not necessary to which the scheduling carried out is retried to system resource Waste.When the technical solution it is determine that it is necessary to be scheduled retry after, be just further taken at identified Exception Type Adaptable scheduling strategy, and the time retried is dispatched using scheduling strategy determination next time and is reached in the time laggard Row scheduling retries, so that being retried by the scheduling of corresponding strategy to corresponding Exception Type, that is, determines that corresponding scheduling retries It time rather than it is applicable in unified strategy without exception determines identical time interval, this is conducive to the time spent when solving abnormal not Reasonable problem, and can solve the problems, such as scheduling retry time synchronization system resource occupancy it is excessive, that is, solve existing Technology uses equal difference time interval determination scheduling next time to retry, and solution caused when the time is extremely less efficient, is provided to system Source occupies more problem.

In addition, in preferred technical solution, for the exception of solution can not be retried by scheduling, just without scheduling weight Examination, directly carries out manpower intervention, avoids scheduling and retries the time to be spent and the occupancy to system resource, improves place Manage efficiency.

In preferred technical solution, for system exception or calls time-out abnormal, it is gradual that ordered series of numbers is passed using Fibonacci The time interval retried is dispatched in raising next time, can first be retried by the short time every being scheduled, after by long-time every being adjusted Degree retries so that first attempt quickly to solve it is abnormal, when failing quickly to solve abnormal again in long-time every rear such exception of solution, Required system resource realization is retried to the abnormal efficiency of processing, scheduling to distribute rationally.

It is abnormal for relying on operation processing in preferred technical solution, it determines and is dispatched next time every rear for a long time It retries, avoids the waste dispatched and retried to system resource in multiple short time interval.

It is abnormal for accidental processing in preferred technical solution, weight is dispatched next time after determining short time interval Examination improves efficiency of such exception of processing.

In short, this present embodiment can be directed to different work dispatch deal exception, different scheduling strategies is taken, is improved It handles abnormal efficiency, save system resource when scheduling retries.

Description of the drawings

Fig. 1 is the flow chart of one embodiment provided by the present application for dispatching the method that retries；

Fig. 2, which is that the application is shown in FIG. 1, dispatches the application scenarios schematic diagram for retrying embodiment of the method；

Fig. 3 is the structural schematic diagram of one embodiment provided by the present application dispatched and retry device.

Specific implementation mode

Many details are elaborated in the following description in order to fully understand the application.But the application can be with Much implement different from other manner described here, those skilled in the art can be without prejudice to the application intension the case where Under do similar popularization, therefore the application is not limited by following public specific implementation.

Fig. 1 is the flow chart of one embodiment provided by the present application for dispatching the method that retries, and Fig. 2 is shown in the application Fig. 1 Scheduling retry the application scenarios schematic diagram of embodiment of the method.It is adjusted with reference to application scenarios shown in Fig. 2 to illustrate in Fig. 1 Degree retries the technical solution of method.

Fig. 2 can be understood as executing the business action scene withholdd on network.In this scenario, business of withholing action is asked The execution asked is generally required by repeatedly calling corresponding processing program, respective handling function, corresponding computer interface, corresponding calculating Machine system etc. is than relatively time-consuming operation.If informing that user withholds success, needs after the completion of waiting for so multinomial calling again User waits for longer time.

In order to inform in time user withhold success or not as a result, in practice usually before executing multinomial scheduler task, Scheduling initiation system is usually after the action request for receiving action initiator, the letter of request server checkout action initiator Breath.As long as user by information checking, directly notifies the request of withholing of user to execute completion, every tune is subsequently carried out again With completing actual business of withholing, with previously to client notify it is consistent.

The information of the user, for example, can be the action request information of user, the credit grade information (such as excellent) of user, Credit score information (such as 701 points).The verification can be business verification card, such as verify whether the action request is payment Action request.

The verification can also be data check, such as whether judge the credit grade information of user, credit score information Meet pre-provisioning request, for example it is 650 points or more that credit grade, which is good above, credit score,.

The verification can also be regular verification, for example, whether the action request of withholing of user meets scheduled time rule Then, such as whether within 8 points of morning period to this period of 6 pm.

After the information for acting initiator for receiving server transmission is by the information of verification, then it is true to action initiator Recognize current action request successful execution.For example, when the action request information of user is to ask withhold information, and credit grade Divide for excellent, credit score 701 when meeting pre-provisioning request, then can receive the notice by information checking of server transmission.

Wherein, action request initiator can be true user, can also be some system, and scheduling initiation system refers to It is some computer system.

The above process is known as synchronizing the stage of accepting, at this stage, as long as the information of user is by associated check, then directly logical The action request successful execution for knowing user improves the efficiency that user handles action request to save the time.

It is above-mentioned synchronize the stage of accepting after, scheduling initiation system just starts to execute multinomial calling, for example calls corresponding position Program, respective handling function, corresponding computer interface, corresponding computer system etc. are managed than relatively time-consuming operation, that is, basis The action request carries out the job scheduling of operation needed for the action request.

For example, when receive server transmission by the notice of information checking after, in order to deduct one in the account of user When the amount of money of part, need by " promoting order " " obtaining user account fund information ", " whether amount of money is no less than in verification account Amount to be charged " " allots oneself ", for progress such as the execution required program of this series of steps, functions It calls.The process is known as the asynchronous process stage.

When being scheduled in the asynchronous process stage, if called between encountering system, abnormal, business datum is abnormal, interface calls Situations such as overtime, scheduling initiation system needs, which are scheduled, to be retried, and the scheduling retries to refer to executing again to be made to required The calling of program, function etc., so as to continue promote execution business action request, until by business action request execute at Work(.After business action request runs succeeded, phase one could have been executed with the action request message success for prenoticing user It causes.

In the asynchronous process stage, scheduling initiation system is scheduled when retrying, certain scheduling is needed to retry strategy, for example, How long after be scheduled retry again, this will directly determine business action how long can be advanced to successfully, dispatch to retry and disappear The practical problems such as consumption system how many resource.Therefore, scheduling retries the selection of strategy for promoting the success of execution business action request It executes in close relations.

It is exactly to determining scheduling plan during the asynchronous process shown in Fig. 2 that scheduling shown in Fig. 1, which retries method, The slightly solution of this key problem, illustrates in detail below.Method is retried below in conjunction with Fig. 1 to the scheduling to carry out in detail It describes in detail bright.

Step S101, specific abnormal conditions when handling abnormal according to job scheduling determine that job scheduling handles exception class Type.

The operation refers to the term " Job " in computer realm, is to execute series of steps to complete one or more The general name of item task.Such as in user in the network platform does shopping scene, when needs deduct a part of money in the account of user When volume, it may be necessary to by " obtaining user account information ", " whether amount of money is no less than amount to be charged in verification account ", " deduct Certain amount of money " and etc., this series of steps is properly termed as an operation.

The job scheduling refers to complete one or more tasks, it usually needs executes certain in a certain order A or certain steps.Such as the preceding paragraph execution withhold task when, need to execute successively in order " obtaining user account information ", " whether amount of money is no less than amount to be charged in verification account ", " the certain amount of money of deduction " and etc..When executing some step, just It is possible that being called to executing program, function etc. used in the step.The scheduling retries, and refers to executing again pair The calling of required program, function etc..

The job scheduling processing is abnormal, refers to that job scheduling handles implementation procedure since some step can not successful execution Caused by entire operation process the case where can not normally promoting.All it is the abnormal specific exception of common job scheduling processing below Situation and possible producing cause：Job scheduling processing is abnormal caused by Account Closure, since Account Closure can lead to nothing Method is withholdd from the account, and then causes processing abnormal；Exception is handled caused by account freezing or account insufficient funds, if account Freeze to be likely to result in no normal direction account reimbursement etc. in family；Job scheduling processing caused by network jitter is abnormal, refers to due to network System is unstable and the job scheduling processing in the short time caused by can not being attached with account etc. is abnormal；System exception or tune With overtime abnormal, for example job scheduling processing caused by can not successfully calling some program, function etc. for a long time is abnormal.

The job scheduling processing Exception Type refers to the classification that various specific job schedulings are handled with abnormal conditions.It is aobvious So, the classification of different modes can be carried out from different perspectives according to different purposes；The same specific job scheduling processing is different Often, several types can be belonged to according to different mode classifications.In the present embodiment, since the main purpose of classification is to determine scheduling Therefore the scheduling strategy retried is classified from the influence angle to scheduling strategy.With above-mentioned caused by Account Closure For dispatch deal exception.In some scenes, Account Closure will lead to not withhold from the account, withhold even if executing again Request, still can not successful execution, this one kind processing is abnormal to retry solution by scheduling, be properly termed as that scheduling can not be passed through Retry the Exception Type of solution.For another example, processing caused by above-mentioned account freezing is abnormal.In some scenes, due to the account It may be freezed by bank account processing system, if withholing to the account, need to wait for, rely on bank account processing system It thaws the account, this processing can be included into extremely relies on operation processing Exception Type.For another example, above-mentioned network jitter causes Processing it is abnormal, since network jitter is usually accidental, infrequently, and can be handled again quickly, thus this Class processing is abnormal to be properly termed as accidental processing Exception Type.For another example, there are system exception or the abnormal conditions for calling time-out abnormal, It needs the time for waiting for random length that could call again, then can be included into the same Exception Type, i.e. system exception or calling It is overtime abnormal.Due to that in many job scheduling systems, can be generated when job scheduling occurs and handles abnormal and correspond to specific exception The exception code of situation, then the specific abnormal conditions according to job scheduling processing when abnormal, determine that job scheduling processing is abnormal Type can be the exception code corresponding to specific abnormal conditions when handling abnormal according to detected job scheduling, determine Job scheduling handles Exception Type.For example, when the specific abnormal conditions when a kind of processing of job scheduling is abnormal occur (such as by Can not withhold from the account in Account Closure), correspondingly detect that the preset exception code of generation is " 00000001 " When, determine that job scheduling processing Exception Type is the exception that solution can not be retried by scheduling.

In the present embodiment, above-mentioned job scheduling processing is abnormal to be generated with the next stage：Local data processing stage, Call remote system processing stage.This is because when executing certain action requests, processing is abnormal to be both likely to occur in local system System, and be likely to occur in the other systems in network.But either dispatch deal anomaly occurs in the number on local system According to processing stage, the data processing stage being still happened in other remote systems, these processing are abnormal can be by the application Scheme handled.

Step S102 handles Exception Type, it is determined whether needs are dispatched next time according to identified job scheduling It retries；And if the determination result is YES, then further comprise obtaining tune corresponding with job scheduling processing Exception Type Degree strategy.

The corresponding scheduling strategy of the processing Exception Type with the job scheduling, to the effect that pass through how long Performance objective task is just again attempted to, and whether needs to carry out other relevant treatments.For example, just being held again by 2 seconds Capable task of withholing.

It is described " Exception Type to be handled according to identified job scheduling, it is determined whether need to carry out scheduling next time and retry； And if the determination result is YES, then further comprise obtaining scheduling plan corresponding with job scheduling processing Exception Type Slightly " it is specifically as follows following at least one situation：

If the job scheduling processing Exception Type is system exception or calls time-out abnormal, determined by the basis Job scheduling handles Exception Type, it is determined whether it is yes that needs are dispatched the judging result retried next time；The scheduling plan Slightly Fibonacci passs ordered series of numbers strategy, i.e., passs ordered series of numbers determination according to Fibonacci and dispatch the time interval retried next time.It is described Fibonacci passs ordered series of numbers and refers to such a ordered series of numbers 2, and 3,5,8,13, its main feature is that each single item is equal to the sum of first two.Example Such as, if currently calling time-out, for example certain function is called, has invoked 2 seconds and runs succeeded not yet, this one kind processing exception is System exception calls time-out abnormal.To this kind of system exception or call time-out abnormal, it is determined whether needs are adjusted next time In the step of degree retries, judging result is yes；Further determine that the time interval dispatched and retried next time is 3 seconds.If again It is not carried out success, then further determines that the time interval dispatched and retried next time is 5 seconds.

If the job scheduling processing Exception Type is that accidental processing is abnormal, at job scheduling determined by the basis Manage Exception Type, it is determined whether it is yes that needs are dispatched the judging result retried next time；The scheduling strategy short time every Strategy.For example, due to the stable and accidental connection error in system operation portion, this abnormal one kind processing is that accidental processing is abnormal.It is right This kind of accidental processing is abnormal, it is determined whether needs to carry out in the step of scheduling retries next time, judging result is yes；Further really Fix that once to dispatch the time interval that retries be the short time every for example being 0.5 second.

If the job scheduling processing Exception Type is to rely on operation processing exception, operation tune determined by the basis Degree processing Exception Type, it is determined whether it is yes that needs are dispatched the judging result retried next time；The scheduling strategy is length When interval strategy.For example, exception is handled caused by account freezing, the reason is that the account is freezed by bank account processing system , in some scenes, if withholing to the account, needs to wait for, relies on the bank account processing system defrosting account, this Kind processing is abnormal for dependence operation processing Exception Type.Exception Type is handled to this kind of dependence operation, it is determined whether need to carry out In the step of scheduling retries next time, judging result is yes；Further determine that the time interval dispatched and retried next time is longer Time interval, such as through 2 days.

If the job scheduling processing Exception Type is the Exception Type that can not retry solution by scheduling, the basis Identified job scheduling handles Exception Type, it is determined whether it is no that needs are dispatched the judging result retried next time；With And if judging result is no, and further perform the step of：Execute alarm operation.For example, since Account Closure can not be from this Account is withholdd, even if executing request of withholing again, can not still be succeeded, this abnormal one kind processing is that can not be retried by scheduling The Exception Type of solution.To this kind of Exception Type that can not retry solution by scheduling next time, it is determined whether needs are adjusted In the step of degree retries, judging result is no；Alarm operating procedure is directly executed to remind manpower intervention, is carried out artificial corresponding Processing.Described in determination can not by scheduling retry cause by the exception of solution is by Account Closure when, can also obtain described The credit information of the corresponding user of account, the credit information include credit grade, credit score.For example, can obtain Message level information, such as " excellent ", " good " etc..It can also be credit score information, for example " 651 " are divided, " 701 " grade.In addition, Credit score, the credit grade of the user can also be adjusted according further to preset credit management rule.For example, The preset credit management rule can be：It closes account before not withholing, then reduces a grade or by credit score Number turns down 50 points.So, in this step, credit grade and credit score are just directly turned down.Following technology effect can be obtained accordingly Fruit：In abnormal cases, for the exception that can not retry solution by scheduling, manpower intervention is directly reminded, improves scheduling weight The efficiency of examination.Meanwhile by the credit information adjustment to user, improving the credit safety of transaction.

After to be dispatched according to step S102 determinations and retry and get corresponding scheduling strategy next time, continue by Described in following step S103, dispatched next time the time retried using scheduling strategy determination, and when reaching the time It executes scheduling next time to retry, be explained in detail below.

If desired step S103 is dispatched and is retried next time, then determine scheduling weight next time using the scheduling strategy The time of examination, and execute scheduling next time when reaching the time and retry.

When based on step S102 determine job scheduling processing Exception Type, whether need carry out next time scheduling retry and After obtaining scheduling strategy corresponding with the job scheduling processing Exception Type when if the determination result is YES, just use described Scheduling strategy determination is dispatched the time retried next time, and is executed scheduling next time when reaching the time and retried.

For example, when system exception is determined based on step S102 or calls time-out abnormal, and determine and need to carry out next time Scheduling is retried and further got passs ordered series of numbers plan with the system exception or the abnormal corresponding Fibonacci of calling time-out Slightly, i.e., according to Fibonacci pass ordered series of numbers determination dispatch the time interval retried next time after, using the scheduling strategy determine It dispatches next time the time retried, and executes scheduling next time when reaching the time and retry；Plan is being retried using above-mentioned scheduling After slightly, the transition of system resource is caused to consume in unsuccessful always be repeatedly executed at predetermined intervals in order to avoid scheduling is retried, it can To retry the strategy that strategy setting one terminates scheduling and retries in conjunction with different scheduling, for example, being passed using the Fibonacci In the case of strategy, following step is specifically executed：First, strategy is passed using the Fibonacci and determines what scheduling next time retried Time.It runs succeeded not yet for example, having invoked 2 seconds, determines that the time interval dispatched and retried next time is 3 seconds, if It is not carried out success again, determines that the time interval dispatched and retried next time is 5 seconds.Secondly, it determines and passs number according to Fibonacci Dispatch whether the time retried is more than preset time interval threshold values determined by row next time.If not exceeded, then in the time Scheduling next time is executed when arrival to retry.For example, identified dispatch the time interval retried 5 seconds less than preset next time Time interval threshold values 20 seconds then executes scheduling next time and retries after 5 seconds.After if scheduling retries repeatedly, again really The time interval that degree of setting the tone retries is more than 20 seconds, then can terminating progress, scheduling retries next time.

For another example when being determined that accidental processing is abnormal based on step S102, and determine that needing to carry out scheduling next time retries And further get with the accidental processing abnormal corresponding short time every strategy, i.e., the time retried is dispatched next time Between be divided into the short time every later, dispatched next time the time retried using scheduling strategy determination, and when reaching the time Scheduling next time is executed to retry；In addition, it is the plan for terminating the termination scheduling of foundation and retrying also to be arranged one for it with number of retries Slightly.Specifically implementation procedure is：First, the time retried is dispatched using the short time next time every strategy determination.For example, due to The stable and accidental connection error in system operation portion, this abnormal one kind processing is that accidental processing is abnormal.It is different to this kind of accidental processing Often, determine that it is the short time every being 0.5 second to dispatch the time interval that retries next time.Secondly, the short time is used to described The execution counting how many times for dispatching the time step retried next time are determined every strategy；It is less than preset time in the execution number When number threshold values, executes scheduling next time and retry.For example, when determine every time dispatch next time short time for retrying every after, counter (counter) all plus 1, when executing number and being less than preset executions number threshold values 20 times for such as 5 times, it is heavy to execute scheduling next time Examination.Wherein, counter is the counter for allowing user any position in sets of threads to create reference.Counter configuration allows user Maximum value (such as 20 times) is configured, counter can be set to after being recycled to 20 times to terminate.

For another example relying on operation processing exception when determining based on step S102, can also continue to determine that the dependence is made Industry handles Exception Type whether by account freezing and when causing, if being caused by account freezing, can obtain account freezing letter Breath, the account freezing information includes that account freezes the time limit.For example, it may be to freeze 24 hours that account, which freezes deadline information,.Into And freeze whether the time limit is less than and preset freezes time limit threshold values described in can continuing to determine.For example, account freezes deadline information is Freeze 24 hours, the preset time limit threshold values that freezes is 48 hours, and definitive result is "Yes" at this time.If so, freezing by the account When tying the time limit, it can also detect whether account freezing has released.For example, when by account freeze the time limit 24 it is small when after, detection Whether account is freezed to have released.If having released, just carries out scheduling next time and retry, i.e., under being determined using the scheduling strategy It once dispatches the time retried, and executes scheduling next time when reaching the time and retry.Accordingly, operation processing exception is being relied on Type is to freeze time limit valve less than preset by freezing the time limit described in determination under the abnormal conditions by account freezing and when causing When value just execute next time scheduling retry, avoid freeze the time limit it is long when and carry out it is invalid dispatch retry, improve scheduling The efficiency retried.

For another example when based on step S102 determine rely on operation processing Exception Type, can also continue to determine described according to Operation is relied whether to handle Exception Type by being caused by account insufficient funds.If so, the account of insufficient funds can also be obtained Associated account number information.For example, current account insufficient funds, just obtain the associated account information of current account.Further, by institute Associated account number is stated as dispatching the object retried next time.For example, due to current account insufficient funds, directly detained next time Money operate, still possibly can not successful execution be therefore scheduled and retry directly against other associated accounts, i.e., to it He executes operation of withholing by interlock account.Thus when needs dispatch and retried next time, under being determined using the scheduling strategy It once dispatches the time retried, and executes scheduling next time when reaching the time and retry.Accordingly, it is handled in the dependence operation Under abnormal conditions when Exception Type is caused by account insufficient funds, by regarding associated account number as what scheduling next time retried Object improves the specific aim that scheduling retries.

According to the technical solution of first embodiment above, one is described again below when there is reimbursement business processing exception, The processing procedure carried out according to the technical solution of first embodiment above.

Step S201, specific abnormal conditions when handling abnormal according to job scheduling determine that job scheduling handles exception class Type.

To the account of user execute reimbursement business without running succeeded, this abnormal conditions may due to Account Closure and It is caused, it is also possible to caused by account is frozen, it is also possible to due to network system it is unstable and can not be with account Caused by being attached, it is also possible to be caused by can not successfully calling some program for a long time.Since Account Closure will Lead to not to account reimbursement, even if executing reimbursement task again, still can not successful execution, one kind processing is abnormal to lead to for this It crosses scheduling and retries solution, be properly termed as that the Exception Type of solution can not be retried by scheduling.Since account is handled by bank account When system is freezed, if carrying out reimbursement to the account, needs to wait for, relies on the bank account processing system defrosting account, this place Abnormal can be included into of reason relies on operation processing Exception Type.It can not be realized with the account of user caused by network jitter It is usually accidental, infrequently to connect this exception, and can be handled again quickly, can so this kind of processing is abnormal To be known as accidental processing Exception Type.And the abnormal abnormal conditions of time-out such as call some program, need to wait for random length when Between could call again, then can be included into system exception or call time-out abnormal.

The above-mentioned job scheduling processing that may occur is abnormal, can be according to corresponding to detected specific abnormal conditions Exception code, come determine job scheduling handle Exception Type.Such as since Account Closure is without normal direction account reimbursement, correspondingly examine When the preset exception code for measuring generation is " 00000003 ", determines and make abnormal to be that can not retry the different of solution by scheduling Often.

Step S202 handles Exception Type, it is determined whether needs are dispatched next time according to identified job scheduling It retries；And if the determination result is YES, then further comprise obtaining tune corresponding with job scheduling processing Exception Type Degree strategy.

If it is when being caused due to can not successfully call some program for a long time that can not execute reimbursement, it is determined that need to carry out Scheduling retries next time, and scheduling strategy is that Fibonacci passs ordered series of numbers strategy, i.e., passs ordered series of numbers determination according to Fibonacci and adjust next time Spend the time interval retried.For example, if current call time-out, for example call certain function, have invoked execute not yet within 2 seconds at Work(, then it is 3 seconds to dispatch the time interval retried next time.If being not carried out success again, dispatch next time retry when Between between be divided into 5 seconds.

If can not execute reimbursement be unstable by network system and caused by can not being attached with account, it is determined that need It is dispatched and is retried next time, the scheduling strategy short time is every strategy.For example, it is 0.5 to dispatch the time interval retried next time Second.

If it is caused by being freezed by bank account processing system by account that can not execute reimbursement, it is determined that needs carry out down Primary scheduling retries, and scheduling strategy is for a long time every strategy.For example, it is 2 days to dispatch the time interval retried next time.

If it is caused by being canceled by account that can not execute reimbursement, it is determined that it need not be dispatched and be retried next time, And it further performs the step of：Execute alarm operation.Manpower intervention is reminded by directly executing alarm operating procedure, into pedestrian Work is handled accordingly.Further, it is also possible to obtain the credit information of the corresponding user of the account, the credit information includes letter With grade, credit score.Furthermore it is also possible to according further to preset credit management rule, the letter of the user is adjusted With score, credit grade.

If desired step S203 is dispatched and is retried next time, then determine scheduling weight next time using the scheduling strategy The time of examination, and execute scheduling next time when reaching the time and retry.

When based on step S202 determine job scheduling processing Exception Type, whether need carry out next time scheduling retry and After obtaining scheduling strategy corresponding with the job scheduling processing Exception Type when if the determination result is YES, just use described Scheduling strategy determination dispatch next time the time retried, and when reaching the time execute next time scheduling retry, specifically divide with Under several situations：

If it is when being caused due to can not successfully call some program for a long time, to be passed according to Fibonacci that can not execute reimbursement The time interval retried is dispatched in ordered series of numbers determination next time, and is executed scheduling next time when reaching the time and retried.In order to avoid Scheduling is retried causes the transition of system resource to consume in unsuccessful always be repeatedly executed at predetermined intervals, and can set a termination The strategy retried is dispatched, following step is specifically executed：Determine the time dispatched and retried next time whether more than between the preset time Every threshold values.It is retried if not exceeded, then executing scheduling next time when reaching the time.For example, the scheduling next time determined retries Time interval 5 seconds be less than preset time interval threshold values 20 seconds, then execute scheduling next time after 5 seconds and retry.If After scheduling retries repeatedly, determine that the time interval that scheduling retries is more than 20 seconds, then can terminate progress next time again Scheduling retries.

If can not execute reimbursement is unstable by network system and can not be attached with account when causing, by the short time every Strategy determination is dispatched the time retried next time, and is executed scheduling next time when reaching the time and retried.In addition, also being set for it It is the strategy for terminating the termination scheduling of foundation and retrying that one, which is set, with number of retries.Specifically implementation procedure is：First, to the use The short time determines the execution counting how many times for dispatching the time step retried next time every strategy；Do not surpass in the execution number When crossing preset number threshold values, executes scheduling next time and retry.For example, ought determine every time dispatch next time short time for retrying every Afterwards, counter (counter) all plus 1, when executing number and being less than preset execution number threshold values 20 times for such as 5 times, under execution Primary scheduling retries.Wherein, counter is the counter for allowing user any position in sets of threads to create reference.Counter is matched Setting allows user configuration maximum value (such as 20 times), can set counter to after being recycled to 20 times to terminate.

If can not execute reimbursement be due to account is freezed by bank account processing system and when causing, can obtain account jelly Information is tied, the account freezing information includes that account freezes the time limit.For example, it may be that freeze 24 small that account, which freezes deadline information, When.And then freeze whether the time limit is less than and preset freezes time limit threshold values described in can continuing to determine.For example, preset freeze the time limit Threshold values is 48 hours, and definitive result is "Yes" at this time.If so, when by the account freezing time limit, account jelly can also be detected Whether knot has released.For example, when by account freeze the time limit 24 it is small when after, detection account freeze whether to have released.If It through releasing, just carries out scheduling next time and retries, i.e., dispatched next time the time retried every scheduling strategy determination using long-time, and Scheduling next time is executed when reaching the time to retry.Accordingly, freeze the time limit less than preset by freezing the time limit described in determination When threshold values just execute next time scheduling retry, avoid freeze the time limit it is long when and carry out it is invalid dispatch retry, improve tune Spend the efficiency retried.

If can not execute reimbursement be due to account is freezed by bank account processing system and when causing, fund can also be obtained The associated account number information of insufficient account.Further, using the associated account number as dispatching the object retried next time.For example, It is scheduled and retries directly against other associated accounts, i.e., operation of withholing is executed to other interlock accounts, to need Scheduling next time is carried out when retrying, is dispatched next time the time retried every scheduling strategy determination using long-time, and in the time Scheduling next time is executed when arrival to retry.Accordingly, by the way that associated account number to be used as to the object dispatched and retried next time, tune is improved Spend the specific aim retried.

Fig. 3 is the structural schematic diagram of one embodiment provided by the present application dispatched and retry device.Due to device embodiment It is substantially similar to embodiment of the method, so describing fairly simple, relevant part refers to the embodiment of the method for above-mentioned offer Corresponding explanation.Device embodiment described below is only schematical.

Degree provided by the present application retries device, including：Exception Type determination unit 301, scheduling strategy acquiring unit 302, Scheduling retries unit 303.

In the present embodiment, Exception Type determination unit 301 is specific when being configured to handle abnormal according to job scheduling Abnormal conditions determine that job scheduling handles Exception Type；Scheduling strategy acquiring unit 302 is configured to according to identified work Industry dispatch deal Exception Type, it is determined whether need to carry out scheduling next time and retry；And if the determination result is YES, then into one Step includes obtaining scheduling strategy corresponding with job scheduling processing Exception Type；Scheduling retries unit 303, is configured to If desired it is dispatched and is retried next time, then the time retried is dispatched using scheduling strategy determination next time, and at this Between reach when execute next time scheduling retry.

Optionally, which is characterized in that the scheduling retries unit and is configured specifically for：Strategy is passed using the Fibonacci The time retried is dispatched in determination next time；Determine that the time dispatched and retried next time determined by ordered series of numbers is passed according to Fibonacci is No is more than preset time interval threshold values；It is retried if not exceeded, then executing scheduling next time when reaching the time.

The application propose it is a kind of for dispatching the electronic equipment retried, since electronic equipment embodiment is substantially similar to method Embodiment, so describing fairly simple, relevant part refers to the corresponding explanation of the embodiment of the method for above-mentioned offer. Electronic device described below embodiment is only schematical.

The application also proposes a kind of for dispatching the electronic equipment retried, including：Display；Processor；Memory is used In the program that storage scheduling retries, described program executes following operation when reading execution by the processor：According to operation tune Specific abnormal conditions when degree processing is abnormal determine that job scheduling handles Exception Type；It is handled according to identified job scheduling Exception Type, it is determined whether need to carry out scheduling next time and retry；And if the determination result is YES, then further comprise obtaining Scheduling strategy corresponding with job scheduling processing Exception Type；If desired it is dispatched and is retried next time, then use institute It states scheduling strategy determination to dispatch next time the time retried, and executes scheduling next time when reaching the time and retry.

The application also proposes a kind of computer readable storage medium, due to the basic phase of computer readable storage medium embodiment It is similar to embodiment of the method, so describing fairly simple, relevant part refers to the correspondence of the embodiment of the method for above-mentioned offer Explanation.Calculation as described below machine readable storage medium storing program for executing embodiment is only schematical.

A kind of computer readable storage medium that the application also proposes, is stored thereon with scheduling and retries program, the program quilt When processor reads execution, following operation is executed：Specific abnormal conditions when handling abnormal according to job scheduling, determine operation tune Degree processing Exception Type；Exception Type is handled according to identified job scheduling, it is determined whether needs to carry out scheduling weight next time Examination；And if the determination result is YES, then further comprise obtaining scheduling corresponding with job scheduling processing Exception Type Strategy；If desired it is dispatched and is retried next time, then dispatched next time the time retried using scheduling strategy determination, and Scheduling next time is executed when the time reaches to retry.

Although the application is disclosed as above with preferred embodiment, it is not for limiting the application, any this field skill Art personnel are not departing from spirit and scope, can make possible variation and modification, therefore the guarantor of the application Shield range should be subject to the range that the application claim defined.

In a typical configuration, computing device includes one or more processors (CPU), input/output interface, net Network interface and memory.

Memory may include computer-readable medium in volatile memory, random access memory (RAM) and/or The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium Example.

1, computer-readable medium can be by any side including permanent and non-permanent, removable and non-removable media Method or technology realize information storage.Information can be computer-readable instruction, data structure, the module of program or other numbers According to.The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), dynamic random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), fast flash memory bank or other memory techniques, CD-ROM are read-only Memory (CD-ROM), digital versatile disc (DVD) or other optical storages, magnetic tape cassette, tape magnetic disk storage or Other magnetic storage apparatus or any other non-transmission medium can be used for storage and can be accessed by a computing device information.According to Herein defines, and computer-readable medium does not include non-temporary computer readable media (transitory media), is such as modulated Data-signal and carrier wave.

2, it will be understood by those skilled in the art that embodiments herein can be provided as method, system or computer program production Product.Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the application Form.It can be used in the computer that one or more wherein includes computer usable program code moreover, the application can be used The computer program product implemented on storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) Form.

Claims

1. a kind of scheduling retries method, which is characterized in that the method includes：

Specific abnormal conditions when handling abnormal according to job scheduling determine that job scheduling handles Exception Type；

Exception Type is handled according to identified job scheduling, it is determined whether is needed to carry out scheduling next time and is retried；And if sentence Disconnected result is yes, then further comprises obtaining scheduling strategy corresponding with job scheduling processing Exception Type；

If desired it is dispatched and is retried next time, then dispatched next time the time retried using scheduling strategy determination, and Scheduling next time is executed when the time reaches to retry.

2. scheduling according to claim 1 retries method, which is characterized in that when described abnormal according to job scheduling processing Specific abnormal conditions determine job scheduling processing Exception Type before, further include：

After the action request for receiving action initiator, the information of request server checkout action initiator；

After the information for acting initiator for receiving server transmission is by the information of verification, confirm to action initiator current Action request successful execution；

According to the action request, the job scheduling of operation needed for the action request is carried out.

3. scheduling according to claim 1 retries method, which is characterized in that the job scheduling processing Exception Type is at least Including following one kind：

The exception of solution can not be retried by scheduling, relied on operation processing exception, accidental processing exception, system exception or called super Shi Yichang.

4. scheduling according to claim 3 retries method, which is characterized in that if job scheduling processing Exception Type is The exception that solution can not be retried by scheduling, then job scheduling processing Exception Type, determination determined by the basis are No needs carried out in the step of scheduling retries next time, and judging result is no；And it further performs the step of：Execute alarm behaviour Make.

5. scheduling according to claim 3 retries method, which is characterized in that if job scheduling processing Exception Type is System exception calls time-out abnormal, then job scheduling determined by the basis handles Exception Type, it is determined whether need into It is yes that row dispatches the judging result retried next time；The scheduling strategy is that Fibonacci passs ordered series of numbers strategy, i.e., according to striking Poona Contract passs ordered series of numbers determination and dispatches the time interval retried next time.

6. scheduling according to claim 3 retries method, which is characterized in that if job scheduling processing Exception Type is Accidental processing is abnormal, then job scheduling determined by the basis handles Exception Type, it is determined whether needs are adjusted next time It is yes to spend the judging result retried；The scheduling strategy short time, as dispatching the time interval retried next time was every strategy Short time every.

7. scheduling according to claim 3 retries method, which is characterized in that if job scheduling processing Exception Type is It is abnormal to rely on operation processing, then job scheduling processing Exception Type determined by the basis, it is determined whether need to carry out next The secondary judging result retried of dispatching is yes；The scheduling strategy is between dispatching the time retried next time every strategy for a long time Be divided into for a long time every.

8. scheduling according to claim 5 retries method, which is characterized in that the use scheduling strategy determines next It is secondary to dispatch the time retried, and execute scheduling next time when reaching the time and retry specially：

Strategy determination is passed using the Fibonacci and dispatches the time retried next time；

Determine whether the time for being passed according to Fibonacci and dispatching and retrying next time determined by ordered series of numbers is more than preset time interval Threshold values；

It is retried if not exceeded, then executing scheduling next time when reaching the time.

9. scheduling according to claim 6 retries method, which is characterized in that the use scheduling strategy determines next It is secondary to dispatch the time retried, and execute scheduling next time when reaching the time and retry specially：

The time retried is dispatched using the short time next time every strategy determination；

To the execution counting how many times for dispatching the time step retried next time every strategy determination using the short time；

When the execution number is less than preset number threshold values, executes scheduling next time and retry.

10. scheduling according to claim 7 retries method, which is characterized in that next being determined using the scheduling strategy Before the time that secondary scheduling retries, further include：

When determining that the dependence operation processing Exception Type is caused by account insufficient funds, the account of insufficient funds is obtained Associated account number information；

Using the associated account number as dispatching the object retried next time.