O＆M visualizes automated job platform and implementation methodTechnical field
The present invention relates to internet operation maintenance technical field, and in particular to a kind of O＆M visualizes automated job platformAnd implementation method.
Background technology
With internet years development, O＆M technology is gradually of interest for masses, and the O＆M for more having an impact global range is bigCan etc. many field technologies of tissue share, make the continuous iteration upgrading of technology in industry, the related tools production of O＆M is gradually improved and veryTo being commercialized, market it is more well-known have a blue whale intelligence cloud, but due to its wider, Er Qieshi that is that commercial product covers O＆M functionImplemented based on public cloud, be more suitable for architecture layer system keeper and use.
O＆M is responsible for the operation maintenance work of the business production system of internet scientific ＆ technical corporation, and for O＆M in itselfCore value is exactly the steady in a long-term of system operation, and limitation change frequency is more few better, and industry data shows that failure is sent outRaw probability, which is from the change of operation system in itself more than 70%, to be caused, so management becomes it in the even more weight of maintenance workWeight.
Therefore the whole life cycle of each change of plan is ensured all in the range of O＆M control, and it includes：Change windowMouth, concrete operations content, each role for changing link, process and output, feedback of status and risk control etc. in real time.
Internet industry competition development is swift and violent, and the system per company is almost all to be updated in iteration everyday, all Zhou FabuTo everyday changing, the working strength of operation maintenance personnel is very big, if being always in manual work, in addition to inefficiency, in teamEveryone experience is different with ability, and processing means also can be variant, it is impossible to forms standard and is easier to trigger human accident, givesCompany and individual bring unnecessary loss.
The content of the invention
This hair is related to the management and execution of O＆M change operation, is more suitable for business and application attendant uses, solutionThe following two technical problem that prior art of having determined is present：
1st, O＆M can control the detailed content of each link during implementation is changed, and can be according to risk judgmentManual intervention is made, the operation transparent of whole cycle, data visualization is allowed.
2nd, work that is simple and repeating is become into automatic flow operation.
The technical solution adopted by the present invention is：
A kind of O＆M visualizes automated job platform, including task processing module and interface control module；
Task processing module includes component, task and task groups；One task groups includes at least one task, task groups controlTask suction parameter transmission in the implementation procedure and implementation procedure of each task in processed group；One task is included at leastThe component that one order is performed, the execution of each component of task control, and the component output parameter of previous component is passed toNext component；
Interface control module includes request receiving interface, performs interface, message push interface, log interface and state processingInterface；Ask receiving interface to obtain title and parameter that task groups are performed, create task groups data and generate task groups ID, will appointBusiness group data Cun Chudao stores DB, and task groups ID is pushed in message queue；Interface is performed to take from message queue in real timeGo out on missions a group ID；State processing interface pushes interface and is connected with performing interface and message, holding in logger task group implementation procedureRow state change；Message pushes interface and performs the execution journal that interface is sended over, while PUSH message is updated information toIn message queue；Log interface takes out the daily record in message queue, and real time propelling movement is into web consoles.
A kind of O＆M visualizes automated job platform implementation method, including job platform change operation, and job platform becomesMore operation includes task processing module operation and interface control module operation, and task processing module includes component, task and taskGroup, interface control module includes request receiving interface, performs interface, message push interface, log interface and state processing interface；
Task processing module operation is comprised the following steps that：
Write the atomic operation collection required for job platform change operation；
The N number of change atomic operation collection to be performed of component package, N is at least 1；
M component is carried out layout by task, and M is at least 1；
Task groups perform task, and title and parameter that task groups are performed are sent to request receiving interface；
Interface control module operation is comprised the following steps that：
Ask receiving interface to obtain title and parameter that task groups are performed, and be sent to execution interface；
Perform interface to perform, retry or fix tentatively atomic operation, and be sent to state processing interface, while by execution journalIt is sent to message and pushes interface；
Execution state change in state processing interface logger task group implementation procedure, and status information is pushed to messagePush interface；
Message pushes interface state processing the interface status information pushed and the execution journal for performing interface transmission, togetherWhen PUSH message update information in message queue；
Log interface takes out the daily record in message queue, and task status is carried out into real-time exhibition.
Further, the task status carries out real-time exhibition by web consoles, and a task groups include at least oneTask a, task includes at least one component, and each component has a drop down button in web consoles, controlled by webPlatform processed checks the progress of task and needs the component of execution, and the daily record situation that component is performed is checked by drop down button.
Further, the atomic operation collection is instruction or script.
Further, task by M component in order or concurrent fashion carry out layout.
Further, task groups perform task temporally, scope or order perform.
A kind of intra-company IT service managements automate platform butting method, comprise the following steps：
Task groups are compiled in the change produced during company service is managed；
User initiates service procedure request；
Interface control module processing service procedure request；
Trigger task groups；
Complete service operation task PUSH message；
Complete service procedure.
A kind of Surveillance center's failure monitoring automates platform butting method, comprises the following steps：
Most common failure is compiled into task groups；
Surveillance center sends warning information；
Interface control module handles warning information；
Surveillance center sets up event, and automatic triggering task groups perform task；
Complete alarming processing pushed information；
Surveillance center's alarm clearing.
A kind of O＆M cloud resource management automation platform butting method, comprises the following steps：
Task groups are compiled into the management and control of O＆M cloud resource and change；
Ask the service of O＆M cloud resource；
Interface control module handles O＆M cloud resource service request；
Task groups are triggered by O＆M cloud resource interface；
Complete O＆M cloud resource processing PUSH message；
Complete the service of O＆M cloud resource.
Beneficial effects of the present invention are as follows：
1st, the pain spot of O＆M change management has been solved just, it is ensured that O＆M can control each during implementation is changedThe detailed content of link, and can make manual intervention according to risk judgment, allows the operation transparent of whole cycle, data visualizationChange.
2nd, ensure that everyone work is quickly substituted by standard practice change, changed significantly there is provided O＆MThe efficiency of management.
Brief description of the drawings
Fig. 1 is task processing module structure chart.
Fig. 2 is interface control module structure chart.
Fig. 3 is O＆M visualization job platform changing process figure.
Fig. 4 is service management automation platform butting method flow chart.
Fig. 5 is Surveillance center's failure monitoring automation platform butting method flow chart.
Fig. 6 is O＆M cloud resource management automation platform butting method flow chart.
Fig. 7 is job platform visualization schematic diagram.
Fig. 8 is the task detailed process schematic diagram using issue.
Fig. 9 is daily record implementation status schematic diagram.
Embodiment
Hereinafter, the present invention is further elaborated in conjunction with the accompanying drawings and embodiments.
Fig. 1 is task processing module structure chart of the present invention, including component, task, three functional modules of task groups.Each work(Energy module declaration is as follows：
Component：For encapsulating specific O＆M operation, such as one script or an interface interchange, each component is onlyIt is responsible for once single operating process；The part of component is as follows：
Component ID, it is globally unique；
The affiliated task ID of component, it is globally unique, for associated task；
Component states：Special key words
Component content：The entity of the component is performed, if it is to call script that component, which is type, content is exactly script sheetBody；
Component input parameter：One component performs required parameter, for example：Ssh (a Secure Shell, safetyShell protocol) remotely connect, it is necessary to host (main frame), user (logon account), password (login password)；It is certain types ofComponent can be made up of default parameters and custom parameter；
Component output parameter：Mark the component to need derived variable parameter after performing, performed for subsequent components；
Component performing module：Perform and execution journal is stored by pipe method, update execution state；
Component classification：, can be by following a few class (as shown in table 1, nonessential, the Yong Huyi of component clustering in programmingCan voluntarily it be extended according to design).
The component classification of table 1
Task：Stream is performed for creating component, a task includes the component that multiple orders are performed, each group of task controlThe execution of part, and the implementing result (component output parameter) of previous component is passed into next component, in addition, appointBusiness can also control execution, pause of subsequent components etc. to operate, and when one of component performs mistake, component states can be returned directlyIt is back to task.
The part of task is as follows：
Task ID, it is globally unique, for associated task group and component；
The affiliated task groups of task, for associated task group；
Component index：One group of tactic component ID, the execution for associating multiple components；
Task suction parameter：For receiving and sending the parameter in task required for first invocation component；
Task execution module：Performed for control assembly, update task status.
Task groups：Multiple tasks are organized together into one task groups of composition, task groups are right according to default implementation strategyTask carries out the suction parameter in the implementation procedure and implementation procedure of each task in packet execution, task groups control groupTransmission.
The part of task groups is as follows：
Task groups ID：It is globally unique, for associated task；
Task is indexed：One group of task ID list included in the task groups；
Task groups implementation strategy：The concurrent implementation strategy of control task；
Task groups suction parameter：For receiving and creating task, task groups can be generated corresponding according to the parameter gotTask.
Task groups performing module：Performed for control task, update task groups state.
Fig. 2 is interface control module structure chart of the present invention, including asks receiving interface 2, state processing interface 7, execution to connectMouth 4, message push the modules such as interface 5, log interface 6.
Interface control module includes request receiving interface 2, performs interface 4, message push interface 5, log interface 6 and stateProcessing Interface 7；Ask receiving interface 2 to obtain title and parameter that task groups are performed, create task groups data and generate task groupsID, stores DB8, and task groups ID is pushed in message queue 3 by task groups data Cun Chudao；Interface 4 is performed in real time from disappearingCease and task groups ID is taken out in queue 3；State processing interface 7 pushes interface 5 and is connected with performing interface 4 and message, logger task groupExecution state change in implementation procedure；Message pushes interface 5 and receives the execution journal that execution interface 4 is sended over, and pushes away simultaneouslyInformation updating information is sent into message queue 3；Log interface 6 takes out the daily record in message queue 3, and real time propelling movement is controlled to webIn platform 1 processed.
Fig. 3 is the changing process figure of O＆M job platform, it can be seen that the change operation of O＆M job platform includes taskProcessing module is operated and interface control module operation.
Task processing module operation is comprised the following steps that：
Write job platform and change the atomic operation collection such as instruction or script required for operation；
The N number of change atomic operation collection to be performed of component package, N is at least 1；
By M component in order or concurrent fashion carries out layout, M is at least 1 to task；
Task groups temporally, scope or order perform task, and the title and parameter that task groups are performed be sent to pleaseSeek receiving interface 2.
Interface control module operation is comprised the following steps that：
Ask receiving interface 2 to obtain title and parameter that task groups are performed, and be sent to execution interface 4；
Perform interface 4 and perform, retry or fix tentatively atomic operation, and be sent to state processing interface 7, while by the execution dayWill is sent to message and pushes interface 5；
Execution state change in the logger task group implementation procedure of state processing interface 7, and status information is pushed to disappearedBreath pushes interface 5；
Message pushes the reception state Processing Interface 7 of interface 5 status information pushed and the execution day for performing the transmission of interface 4Will, while PUSH message is updated information in message queue 3；
Log interface 6 takes out the daily record in message queue 3, and task status is carried out into real-time exhibition.
Interface control module and task processing module are most important two functional modules of the present invention.
By interface control module, instruction that each technical attendant of company is write and script can be made by connecingThe existing remote scheduling of cause for gossip and execution, script file can be distributed in batches, and have the parameter of unified technical requirements in Interface designAnd specification, the difference in functionality design of multiplex roles, except attendant without logging into specific host server performs operation, and operateProcess and result also can return platform by interface message and daily record real time propelling movement, so as to realize standardization and the whole course visible of operationChange.
By task processing module, technical attendant can be the ability of instruction and script edit on web consolesComplete and store, and realize that versioned is managed, so as to be order and the script transparence of each attendant, that is, operateContent can be audited and be multiplexed, also make routine work become instruction or set of scripts solidify, these set of scripts pass through taskDesign is packaged into component until task groups, by control task group, realizes the association between different operating with being connected, and by pre-The rule execution first set, makes it possible to be performed in series or concurrently performs.These task groups by multiple authentication and improve after,A thematic affairs in maintenance work will be become, can be arranged to touch periodically or in the case where a certain condition is satisfiedHair is performed, and semi-automatic or full automatic working system action is realized to a certain extent, is reduced human intervention, is finally achieved and carryElevator ties up efficiency.
Embodiment one, intra-company's IT service managements automation platform butting method.
Standardization is docked with intra-company platform, if any change can be produced to corporate business system by being related to, includingFor business datum additions and deletions change look into, the continuous integrating of program etc. can become the job task group operation of standard, task groupsPowerful policing feature can make flow adapt to a variety of demand scenes.Dock in-company IT service management platforms such as Fig. 4 institutesShow, step is as follows：
Task groups are compiled in the change produced during company service is managed；
User initiates service procedure request；
Interface control module processing service procedure request；
Trigger task groups；
Complete service operation task PUSH message；
Complete service procedure.
By the service routinized in service management, all job task is compiled in packing, is formed from asking to auditing to work orderWhole process hardware and software platform is completed to operation to processing, and user only needs to concern and initiates application, judges rationality and compliance on streamExamination ＆ approval, so as to gradually reduce artificial work, really improve flow path efficiency.
Embodiment two, Surveillance center's failure monitoring automation platform butting method.
Operation system is stable except change causes accident, is secondly exactly all kinds of basic failures, such as server overload, applicationDaily record excessively causes full disk write, server hardware damage, network jitter etc., finds that failure is Surveillance center in the very first timeResponsibility, but very first time response and processing, can just repair failure within the most short time, recover stable, and the response of people and placeReason is all to need time always upper ten minutes level slower than system automation processing, therefore is got through by platform and Surveillance center,The task groups for handling most common failure are realized and editted, after alarm triggered gives a warning information, Surveillance center can set up thingPart and then system docking trigger the troubleshooting tasks carrying of matching to job platform automatically, just can in unattended situation,Repair the failure.Idiographic flow is as shown in figure 5, comprise the following steps：
Most common failure is compiled into task groups；
Surveillance center sends warning information；
Interface control module handles warning information；
Surveillance center sets up event, and automatic triggering task groups perform task；
Complete alarming processing pushed information；
Surveillance center's alarm clearing.
Embodiment three, O＆M cloud resource management automation platform butting method.
Corporate data center can more and more turn to cloud computing technology, the popularization of public cloud and privately owned cloud intoIt is ripe so that O＆M resource management becomes enhanced convenience fast, therefore equally can be with butt-joint operation for resource management and control and changePlatform is realized, so that make resource management change controllable all the more and efficient, but it needs cloud resource platform also to provide related behaviourMaking interface can dock, specifically as shown in fig. 6, comprising the following steps：
Task groups are compiled into the management and control of O＆M cloud resource and change；
Ask the service of O＆M cloud resource；
Interface control module handles O＆M cloud resource service request；
Task groups are triggered by O＆M cloud resource interface；
Complete O＆M cloud resource processing PUSH message；
Complete the service of O＆M cloud resource.
Instrument, which can connect all, based on same O＆M job platform needs the scene of operation service system, is allowed toMore standard, more close rule, it is more efficient.
Example IV, the visualization of O＆M job platform.
For some more great operation tasks, such as big version is issued, using dilatation and migration, major change planDeng, being just highly desirable to whole monitoring ensures that any link is performed according to plan, and again perform avoid carrying out again in window phase it is artificialOperation, causes the time to elongate even One's cake is dough so as to delay whole process and go wrong.
Task groups are designed as with the good task scheduling of pre-edit to lead to there is provided the ability combination web platforms of task layoutThe visualization that web platforms realize O＆M job platform is crossed, it is specific as shown in Figure 7.
Task groups can include multiple tasks, you can so that the affairs to be done are edited good berth script in advance in advance, and in behaviourAll set each workflow sequence formation work plan before making, while having multiple component operations in each task, can be controlledThe execution of task processed, and the daily record output of the progress and detailed process of whole concern task.Fig. 8 is programmed hairThe task detailed process of the application issue of cloth in the works.
The progress bar of the macro-progress of current application issue can be paid close attention in taskbar, and can be known below taskThe component operation performed, the responsible specific works of each component operation and its current state are needed, and for each componentOperation clicks on drop down button and sees the daily record situation specifically performed, as shown in Figure 9.
Each step of component Job execution and state are known from specific data, if it find that problem can be positioned at onceTo in place of the details of particular problem, and in component task status also can Feedback go out the state of failure, it is ensured that problem firstTime is captured and shows.
The present invention is preferably write by python language developments, and runs on Linux system, is not had for computer hardwareParticular/special requirement.
Although the present invention is disclosed as above with preferred embodiment, it is not for limiting the present invention, any this areaTechnical staff without departing from the spirit and scope of the present invention, may be by the methods and techniques content of the disclosure above to this hairBright technical scheme makes possible variation and modification, therefore, every content without departing from technical solution of the present invention, according to the present inventionAny simple modifications, equivalents, and modifications made to above example of technical spirit, belong to technical solution of the present inventionProtection domain.