JP2016018350A - Batch server maintenance method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform maintenance to a batch server in which a business batch program operates for a long time, while suppressing influence on batch business or online business related to batch business as much as possible.SOLUTION: A batch server maintenance method for performing maintenance to a batch server in which a business batch program 11 is being executed, includes: a step (a) in which a maintenance program 12 makes the batch server set a stop request flag 14; a step (b) in which the business batch program 11 makes the batch server determine whether the stop request flag 14 has been set for every timing of a check point CP, and when the flag has been set, makes the batch server set a stop completion flag 16 to stop execution; and a step (c) in which the maintenance program 12 makes the batch server determine whether the stop completion flag 16 related to the business batch program 11 has been set, and when the flag has been set, makes the batch server performance maintenance to the batch server.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an operation management technology for server equipment, and more particularly to a technology effective when applied to a maintenance method for server equipment that performs batch processing for a long time.

  To maintain the normal operation of an information processing system that includes server devices, maintenance processing such as applying patches, bug fixes, and other software updates, periodic reboots, etc. is necessary. Since the server under maintenance cannot basically perform business processing, various measures are taken in order not to affect the continuation of the overall business.

  For example, Japanese Patent Laid-Open No. 2012-174220 (Patent Document 1) discloses a server management apparatus that manages reboot of each server in a system including a plurality of servers, and a second after executing reboot for the first server group. When rebooting is not completed for the servers that make up the server group instructed to reboot and the reboot planning unit that plans the reboot order for executing the reboot for the server group of the server group, the server group degeneration operation is performed by separating the server that does not complete the reboot. Decommissioning operation determination unit that determines whether or not it is possible and even if there is a reboot error in the servers that constitute the first server group, the determination that the degeneration operation of the first server group can be performed by the degeneration operation determining unit is completed Rebooting the second server group with a reboot instruction By a control unit, while reducing the burden of the maintenance personnel, technology that can execute the reboot of the server in the maintenance time has been described.

  In JP 2003-15894 (Patent Document 2), in a computer system including at least one cluster composed of a plurality of computers, a comprehensive monitoring device manages configuration information and the like of the plurality of computers. By determining the patch application schedule based on the configuration information and the like, the patch application process is performed on the operating systems operating on the computers constituting the computer system without stopping the entire computer system. The technology is described.

JP 2012-174220 A JP 2003-15894 A

  In conventional technology, server maintenance is based on a server group or cluster configuration in which multiple servers that can perform the same processing exist, and a maintenance stop time is provided for each of the servers in turn. In the meantime, the maintenance process is efficiently performed.

  The above method can be applied to a server that performs online business (hereinafter may be referred to as “business server”), but a server that performs batch business (hereinafter referred to as “batch server”). Is difficult to apply. Online business usually has a characteristic that a large number of transactions that can be completed in a short time, so it is configured to start multiple business servers and improve processing capacity and availability by load balancing. It is relatively easy to disconnect and stop the maintenance target server during business.

  On the other hand, batch operations usually do not require a high degree of immediacy, so there may be cases where multiple systems of batch servers are prepared and processing capacity and availability are not improved, and a single server is configured. In addition, since a large amount of data is collectively processed, the processing time may be long, and it is difficult to stop the business batch program being processed at a desired timing for maintenance. .

  SUMMARY OF THE INVENTION An object of the present invention is to provide a batch server maintenance method for performing maintenance on a batch server in which a business batch program operates for a long time while minimizing the influence on batch business and related online business.

  The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  Of the inventions disclosed in this application, the outline of typical ones will be briefly described as follows.

  A batch server maintenance method according to a representative embodiment of the present invention is a batch server maintenance method for performing maintenance on a batch server in which one or more batch programs are being executed, and (a) for the batch server A step in which a maintenance program started to perform maintenance causes the batch server to set a stop request flag; and (b) checkpoints at which the batch programs arrive at the batch server at predetermined intervals during execution. At each timing, it is determined whether or not the stop request flag is set. If the stop request flag is not set, the execution is continued. If the stop request flag is set, a stop completion flag is set. And (c) the maintenance program Causing the batch server to determine whether or not the stop completion flag for each batch program is set, and when all the stop completion flags are set, causing the batch server to perform maintenance. It is what you have.

  Among the inventions disclosed in the present application, effects obtained by typical ones will be briefly described as follows.

  That is, according to the representative embodiment of the present invention, it is possible to perform maintenance on a batch server in which a business batch program operates for a long time while minimizing the influence on batch business and related online business. Become.

It is the figure which showed the outline | summary about the example of the batch server maintenance method which is one embodiment of this invention. It is the figure which showed the outline | summary about the structural example of the information processing system to which the batch server maintenance method of one embodiment of this invention is applied. It is the flowchart which showed the outline | summary about the example of the flow of a process of the business batch program in one embodiment of this invention. It is the flowchart which showed the outline | summary about the example of the flow of the process of the maintenance program in one embodiment of this invention. It is the flowchart which showed the outline | summary about the example of the flow of a process of the business server program in one embodiment of this invention. It is the figure which showed the outline | summary about the example of the data structure of the batch management table in one embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

  As described above, it is difficult for a batch server in which a business batch program operates for a long time to stop a business batch program being processed, and therefore it is difficult to provide a stop time for maintenance at a desired timing.

  In addition, among the batch servers on which the business batch program operates, data is transmitted by file transfer from other business servers, for example, once every 15 minutes or once every 1000 cases. In some cases, access is made at predetermined intervals. In the case of such a configuration, for example, when the business server is a system that is always operating for 24 hours 365 days or the like, the batch server needs to be always operable, and maintenance is required. It is difficult to provide a stop time for. In addition, even when maintenance is performed on the batch server, it is necessary to take measures such as controlling other business servers to stop file transfer and prevent access to the batch server.

  Therefore, the batch server maintenance method according to an embodiment of the present invention provides checkpoints that can be temporarily stopped in a business batch program at predetermined intervals, and maintenance jobs, etc. in a storage area that can be referred to from the business batch program. When the stop request flag is set by, the business batch program terminates voluntarily at the next checkpoint, and then performs maintenance processing (execution of maintenance jobs, etc.) on the batch server. It is possible. In addition, after the completion of the maintenance, the business batch program that has been terminated once is restarted from the checkpoint after the flag is cleared.

  This makes it possible to temporarily stop the business batch program and secure a stop time for maintenance by setting the stop request flag at the timing when the maintenance process is required while operating the business batch program normally. .

  FIG. 1 is a diagram showing an outline of an example of a batch server maintenance method according to an embodiment of the present invention. The business batch program 11 operating on the batch server is provided with checkpoints CP at predetermined intervals. The business batch program 11 determines whether or not the stop request flag 14 is set every time the checkpoint CP arrives. To check. On the other hand, the maintenance program 12 is activated, for example, at a timing when the maintenance process needs to be performed, and sets a stop request flag 14 (sets to ON).

  When the stop request flag 14 is set to ON, the business batch program 11 detects this at the timing of the next checkpoint CP, and then sets the stop completion flag 16 (sets to ON) and ends abnormally. The maintenance program 12 executes a desired maintenance process after confirming that the stop completion flag 16 is set to ON, that is, that the business batch program 11 has ended. When the maintenance process is completed, the maintenance program 12 cancels the stop request flag 14 and the stop completion flag 16. Thereafter, the business batch program 11 can resume processing from the checkpoint CP that ended abnormally last time.

<System configuration>
FIG. 2 is a diagram showing an outline of a configuration example of an information processing system to which the batch server maintenance method according to the embodiment of the present invention is applied. The target information processing system has, for example, a configuration in which a batch server 10 and one or more business servers 20 are connected so that they can communicate with each other via a network 30 such as a LAN (Local Area Network). Have.

  The batch server 10 is a server device that performs batch operations. For example, the batch server 10 includes one or more business batch programs 11 implemented by software operating on middleware such as an OS (Operating System) and a DBMS (DataBase Management System) (not shown). , And the maintenance program 12, etc., which can be executed manually or automatically. Each data includes a batch management table 13, a stop request flag 14, and a server list 15 implemented by a database, a file, a memory table, and the like.

  The business batch program 11 is a batch program that executes business processing related to the online business executed in the business server 20. For example, a plurality of processing request data and log files generated by business processing for a certain period of time in each business server 20 are acquired via the network 30, and are registered in a batch, asynchronously with the online business in the business server 20. A program that performs this processing is assumed.

  As described above, the business batch program 11 is implemented with checkpoints that can be paused spontaneously at predetermined intervals (for example, once every 15 minutes or once every 1000 cases). If the stop request flag 14 described later is ON at each checkpoint, the necessary information is recorded so that the process can be resumed from the checkpoint, and the process is temporarily terminated abnormally. It shall be.

  The maintenance program 12 is a program for performing a maintenance process on the batch server 10 and is automatically or manually started at the timing of executing the maintenance. As the maintenance process, for example, the business batch program 11 is executing a process such as patch application or version upgrade of middleware such as the OS of the batch server 10 or other software including the business batch program, reboot of the batch server 10, etc. In some cases, processing that cannot be executed is included.

  The batch management table 13 is a table for managing the execution status of the business batch program 11. The execution status also includes a status that is once terminated at a check point for maintenance of the batch server 10. The contents of the batch management table 13 will be described later.

  The stop request flag 14 is a flag that is set to ON by the maintenance program 12 in order to request the business batch program 11 to stop when the maintenance program 12 is started to perform maintenance of the batch server 10. . The flag can be set on a storage device such as a memory or HDD (not shown) that can be referred to from each business batch program 11. You may make it set and hold | maintain on the batch management table 13 or another database.

  The server list 15 is data that holds a list of destination information and the like of the business server 20 that needs to be notified (to make a stop request) when performing maintenance processing of the batch server 10. As described above, the business server 20 to be notified is, for example, the business server 20 that accesses the batch server 10 such as file transfer at a predetermined interval, and does not allow the batch server 10 to perform the access during maintenance. It is something that needs to be done. If the business server 20 to be notified is fixed, it may be unnecessary. The server list 15 holds, for example, a list of the host name and IP address of each business server 20, and the maintenance program 12 notifies each target business server 20 of a stop request based on this information. To do.

  The business server 20 is a server device that provides an online business function to a client terminal or the like (not shown) via the network 30 or the like. For example, the business server 20 is based on software that runs on middleware such as an OS, DBMS, or Web server program (not shown). It has a business server program 21 to be implemented. Further, it has data such as a stop request flag 22 implemented by a database, a file, a memory table, or the like. The business server program 21 may have a plurality of business servers 20 of different types and contents, or a configuration in which a plurality of one type of business server 20 are provided in parallel to perform load distribution processing. .

  The business server program 21 is a program that performs online business processing based on a request from a client terminal or the like. In the present embodiment, the network 30 stores a plurality of processing request data, log files, etc. accumulated so far at a predetermined interval (for example, once every 15 minutes or once every 1000 processes). The file is transferred to the batch server 10 via, and has a function of requesting batch processing by the business batch program 11. At this time, if a stop request flag 22 to be described later is ON, file transfer to the batch server 10 is not performed (for example, wait until the flag is released or hold and continue processing). It shall be implemented.

  The stop request flag 22 is a flag that is set to ON upon receiving a stop request notification from the maintenance program 12 when the maintenance program 12 is activated in the batch server 10. The flag can be set on a storage device such as a memory or HDD (not shown) that can be referred to from the business server program 21, for example.

<Process flow>
FIG. 3 is a flowchart showing an outline of an example of the processing flow of the business batch program 11 on the batch server 10. When the business batch program 11 starts processing, it registers information in a batch management table 13 to be described later, and then starts loop processing that repeats processing sequentially for each processing step (S01 to S05).

  In each processing step, first, it is determined whether or not a checkpoint has arrived (S02). As described above, checkpoints may be implemented to arrive at a predetermined interval such as once every 15 minutes or once every 1000 processes, or a series of processing steps of a business batch program It may be embedded directly in an appropriate place. If the checkpoint has not yet arrived, the target processing step is executed as usual (S04), and the process proceeds to the next processing step to continue the loop processing (S05, S01). When all the processing steps have been executed and the loop processing is completed, the business batch program 11 is terminated as normal termination (S06).

  If a checkpoint has arrived in step S02, it is determined whether the stop request flag 14 is set (set to ON) on the batch server 10 as a process at the time of the checkpoint (S03). ). If the stop request flag 14 is not set to ON, the process proceeds to step S04 and the execution of the target processing step is continued.

  If the stop request flag 14 is set to ON in step S03, the processing of the business batch program 11 is temporarily stopped. That is, information necessary for resuming processing from the checkpoint after the temporary stop is recorded (S07), the stop completion flag 16 is set to ON (S08), and the business batch program 11 is ended as an abnormal end ( S09). In the present embodiment, the stop completion flag 16 is set for each business batch program 11 on the batch management table 13 described later.

  Through the above processing, the business batch program 11 is voluntarily suspended (in this embodiment, not suspended but abnormally terminated) when the stop request flag 14 is set to ON at each checkpoint. Is possible.

  FIG. 4 is a flowchart showing an outline of an example of the processing flow of the maintenance program 12 on the batch server 10. The maintenance program 12 is activated when a maintenance job such as patch application needs to be executed in the batch server 10. When activated, first, the stop request flag 14 for temporarily stopping the business batch program 11 is set to ON on the batch server 10 (S11).

  Thereafter, a stop request is notified to the related business server 20 to stop access to the batch server 10 (S12). Information about the related business server 20 can be acquired with reference to the server list 15, for example. In the present embodiment, the PUSH type notification that notifies the stop request information from the batch server 10 to each business server 20 is used. However, each business server 20 appropriately notifies the batch server 10 of the status of the stop request flag 14. It is good also as a PULL type notification which inquires for.

  Thereafter, it is determined whether or not all the business batch programs 11 are stopped (S13). That is, it is determined whether all the target business batch programs 11 have the stop completion flag 16 set to ON. In the present embodiment, as described above, the status of the stop completion flag 16 for each business batch program 11 is managed on the batch management table 13. Not only the business batch program 11 but also all business servers 20 that have notified the stop request in step S12 may confirm that access to the batch server 10 has been stopped.

  If all the business batch programs 11 are not stopped in step S13, after waiting for a predetermined time (S14), the process returns to step S13 to confirm the stop of the business batch program 11 again. When all the business batch programs 11 are stopped, a desired maintenance process is executed (S15). Here, the maintenance program 12 itself may perform processing such as patch application on the OS or the like of the batch server 10 or notify the system administrator or the like that the maintenance processing is ready. Thus, manual maintenance may be encouraged.

  When the maintenance process ends, the maintenance program 12 sets all the stop flags to OFF to cancel them (S16), and ends the process. The stop flag includes a stop request flag 14 on the batch server 10, a stop request flag 22 on the business server 20, and a stop completion flag 16 on the batch management table 13.

  Through the above processing, the maintenance program 12 (or system administrator or the like) can execute the maintenance processing after forcibly forming a work time frame by temporarily suspending all business batch programs 11 that have been operating for a long time. .

  FIG. 5 is a flowchart showing an outline of an example of the processing flow of the business server program 21 on the business server 20. When starting the process, the business server program 21 starts a loop process that sequentially repeats the process for each process step (S21 to S27).

  In each processing step, first, it is determined whether or not the processing content is access to the batch server 10 (file transfer in the present embodiment) (S22). If it is not file transfer, the target processing step is executed as usual (S23), and the process proceeds to the next processing step to continue the loop processing (S27, S21). When all the processing steps are executed and the loop processing is finished, the business server program 21 is finished.

  If the processing step is file transfer in step S22, it is determined whether or not the stop request flag 22 is set (set to ON) on the business server 20 (S24). As described above, instead of confirming the state of the local stop request flag 22, the batch server 10 may be inquired about the state of the stop request flag 14. If the stop request flag 22 is not set to ON, the file transfer process to the batch server 10 is executed as usual (S25), and the process proceeds to the next process step to continue the loop process (S27, S21).

  On the other hand, if the stop request flag 22 is set to ON in step S24, after waiting for a predetermined time (S26), the process returns to step S24 to confirm the state of the stop request flag 22 again. That is, the file transfer to the batch server 10 is temporarily reserved until the stop request flag 22 is set to OFF (cancelled).

  In the example of FIG. 5, while the file transfer to the batch server 10 is reserved, the processing is also stopped for the normal processing steps. Here, when the business server program 21 is configured to provide a user interface such as a screen to a client terminal or the like, for example, a stop request flag 14 is set on the batch server 10 as a notification to the user via the screen. If it is only, a message such as “May take longer to process” is displayed, and if the stop completion flag 16 is set, a message such as “Please wait about XX minutes until processing is completed” is displayed. It may be. If possible, the file transfer may be asynchronous with the normal processing step, and only the file transfer may be reserved and the usual processing step may be continued.

  With the above processing, the business server program 21 can stop access to the batch server 10 by file transfer or the like while the batch server 10 is performing maintenance processing.

<Data structure>
FIG. 6 is a diagram showing an outline of an example of the data configuration of the batch management table 13 on the batch server 10. The batch management table 13 is a table for managing the execution status of the business batch program 11, and includes items such as a program name, start date / time, stop completion flag, checkpoint, stop date / time, restart date / time, and end date / time. .

  The program name item holds information such as a name and ID for uniquely identifying the target business batch program 11. The item of the start date / time holds time stamp information at the time when the target business batch program 11 is started. The item of the stop completion flag holds flag information indicating that the target business batch program 11 has temporarily stopped processing based on the stop request flag 14 being set to ON. The checkpoint item holds information for specifying a checkpoint when the target business batch program 11 is temporarily stopped. The check point is a starting point when the target business batch program 11 is re-executed.

  The stop date / time item holds time stamp information at the time when the target business batch program 11 is temporarily stopped (in this embodiment, not suspended but abnormally terminated). The item of the restart date / time holds time stamp information at the time when the target business batch program 11 is re-executed after being suspended. The end date / time item holds time stamp information when the target business batch program 11 completes processing.

  Note that the data configuration (items) of each table shown in FIG. 6 is merely an example, and any other table configuration or data configuration may be used as long as similar data can be held and managed. .

  As described above, according to the batch server maintenance method according to an embodiment of the present invention, checkpoints that can be temporarily stopped are provided in the business batch program 11 at predetermined intervals, and are referred to from the business batch program 11. In a possible storage area, when the stop request flag 14 is set by the maintenance program 12, the business batch program 11 is voluntarily terminated at the next checkpoint. Further, when the stop request flag 14 is set, the business server program 21 on the business server 20 stops access to the batch server 10 by making a stop request to the related business server 20.

  With these processes, it is possible to secure a stop time for maintenance while operating the business batch program 11 as usual, and to perform a maintenance process on the batch server 10. In addition, after the completion of the maintenance, the business batch program 11 which has been once ended after the flag is released can be restarted from the checkpoint.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiments. However, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention. Needless to say. For example, the above-described embodiment has been described in detail for easy understanding of the present invention, and is not necessarily limited to the one having all the configurations described. In addition, it is possible to add, delete, and replace other configurations for a part of the configuration of the above-described embodiment.

  Moreover, in each said figure, the control line and the information line have shown what is considered necessary for description, and do not necessarily show all the control lines and information lines on mounting. Actually, it may be considered that almost all the components are connected to each other.

  The present invention can be used in a maintenance method for server equipment that performs batch processing for a long time.

DESCRIPTION OF SYMBOLS 10 ... Batch server, 11 ... Business batch program, 12 ... Maintenance program, 13 ... Batch management table, 14 ... Stop request flag, 15 ... Server list, 16 ... Stop completion flag,
20 ... business server, 21 ... business server program, 22 ... stop request flag,
30 ... Network,
CP ... Checkpoint

Claims (4)

A batch server maintenance method for performing maintenance on a batch server in which one or more batch programs are being executed,
(A) a maintenance program started to perform maintenance on the batch server causes the batch server to set a stop request flag;
(B) Each batch program causes the batch server to determine whether or not the stop request flag is set at every checkpoint timing that arrives at a predetermined interval during execution, and the stop request flag is set If not, continue execution, and if the stop request flag is set, respectively, a step of setting a stop completion flag to stop execution,
(C) The maintenance program causes the batch server to determine whether or not the stop completion flag relating to each batch program is set, and when all the stop completion flags are set, A process of performing maintenance;
A batch server maintenance method. In the batch server maintenance method according to claim 1,
In the step (a), the maintenance program further causes the batch server to notify a stop request to another server that can access the batch server,
Furthermore, (d) The other server which received the notification of the said stop request from the said batch server has the process of stopping the access with respect to the said batch server, The batch server maintenance method. In the batch server maintenance method according to claim 1,
(E) the maintenance program causing the batch server to release the stop request flag and the stop completion flag after the maintenance for the batch server is completed;
(F) The restarted batch program causes the batch server to resume execution from the checkpoint when execution is stopped in the step (b);
A batch server maintenance method. In the batch server maintenance method according to claim 2,
Further, (g) when the other server has the stop request flag or the stop completion flag set on the batch server, a screen for notifying the client terminal that processing takes time is displayed. A batch server maintenance method.