JPH02171833A - Fault recovering system for virtual space - Google Patents

Abstract

PURPOSE:To recover the fault of a virtual space without stopping the virtual space by providing a slave task, which executes work, and a master task to monitor this slave task in the virtual space and regenerating the slave task by the master task when the slave task is reduced by the fault. CONSTITUTION:A slave task 4 and a master task 5 are provided on a virtual space 2 and the master task 5 generates the slave task 4 by slave task generation function 6 and monitors this slave task. When the fault is generated while the slave task 4 executes the work, extinguishment information 10 are generated and the slave task is extinguished. When the master task 5 detects the extinguishment, the extinguishment cause of the slave task 4 is analyzed by end decision function 11 and the slave task 4 is regenerated by the function 6. The slave task 4 executes a work program which is newly given from a control space. Thus, without stopping the virtual space, the fault can be recovered. Then, a recovery time is shortened and the operation efficiency of a system is improved.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an information processing system such as an online system or a database system using a multi-virtual storage type electronic computer, and particularly relates to an information processing system such as an online system or a database system that uses a computer with multiple virtual memory, and particularly to This invention relates to a failure recovery method for virtual space on a multiple virtual memory system that enables improved operability on the surface.

[Conventional technology]

As a conventional technology related to this kind of multiple virtual storage type information processing system, for example, "Latest database system and its application (Benshi Publishing BTT special issue January 1984)"
” and the like are known. In this type of conventional technology, when a fault or the like occurs in a program being executed in a virtual space, the virtual space is stopped and the virtual space is restarted to recover from the fault.

[Problem to be solved by the invention]

The above-mentioned conventional technology recovers from a failure by stopping and restarting the virtual space, so it is necessary to stop and restart the virtual space in order to recover from the failure.
There is a problem in that the operation is cumbersome and it takes a lot of time to recover from a failure.

It is an object of the present invention to solve the problems of the prior art described above, to shorten the failure recovery time of virtual spaces, and to eliminate troublesome operations such as starting and stopping virtual spaces. The objective is to provide a disaster recovery method.

[Means to solve the problem]

According to the present invention, the purpose is to provide a child task in a virtual space for performing a task and a parent task for monitoring the child task,
When a child task disappears due to a failure, this is achieved by the parent task regenerating a new child task.

[For production]

By providing a child task and a parent task in the virtual space, even if the child task that performs the task disappears due to a failure, the parent task can continue, so it is possible to control the virtual space so that it does not stop. The parent task detects the death of a child task and regenerates a new child task in the same way it previously created it.

With these, failure recovery of the virtual space can be performed without stopping the virtual space.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a virtual space failure recovery method on a multiple virtual storage system according to the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a diagram explaining the functions and processing of an embodiment of the present invention;
FIG. 2 shows a block [I] showing the system configuration of an embodiment of the present invention.
The block diagram and FIG. 3 are flowcharts explaining the operation. In Figures 1 and 2, 1 is a multiple virtual storage space, 2 is
is a virtual space, 3 is a business progera J4.4 is a child task, 5
is the parent task, 6 is the child task generation 8! 7 is a child task monitoring function, 8 is a failure, 9 is a failure 1f post-processing, 10 is a child task disappearance notification, 11 is an end determination function, 12 is a child task failure recovery function, 13 is a terminal device, 14 is a communication control 15 is a central processing unit, 16 is a control space, 17 is a business processing space group, 18 is a database, and 19 is a virtual space failure recovery function.

An embodiment of a processing system such as an online system 1 and a database system according to the present invention includes a central processing unit 15, a communication control device, etc., as shown in FIG. 14, it is configured with a plurality of terminal devices 13 and a database 18.

A multiple virtual storage space 1 configured within the central processing unit 15 is constructed by a business processing space group 17 made up of a plurality of virtual spaces 2 and a control space 16 that controls this business processing space group 17. When a message is input from the device 13, the message is passed via the communication control device 14 to an online system or database system built in multiple virtual storage spaces l within the central processing unit 15. II? The II space 16 determines the business process to be performed based on the content of this message, determines the most suitable virtual space 2 from the business process space group 17,
The virtual space 2 is caused to execute the business. The virtual space 2 executes the business program 3 according to instructions from the control space 16, accesses the database 18, etc., and executes business processing. The virtual space failure recovery function 19 in the virtual space 2 monitors the execution of business processes performed within the virtual space 2, and performs failure recovery processing for the virtual space 2 when a failure occurs.

Next, the configuration and operation of each virtual space 2 will be explained in detail with reference to FIG.

Each virtual space 2 is provided with a child task 4 that executes the business program 3 and a parent task 5 that monitors the child task 4.

The parent task 5 generates a child task 4 using the child task generation function 6, and then monitors the child task 4 using the child task monitoring function 7.

When a fault 8 such as a program failure occurs during execution of the business program 3, the child task 4 executes post-failure processing 9, issues a child task disappearance notification 10, and disappears. When the parent task 5 receives the child task disappearance notification 10 from the child task 4 using the task monitoring function 7 and detects the disappearance of the child task,
The termination determination function 11 analyzes the cause of disappearance of the child task.

The failure determination function 11 determines that a failure has occurred based on the post-failure processing 9 of the child task 4 and the child task disappearance notification 10, and executes the child task failure recovery function 12 as necessary. Thereafter, the parent task 5 regenerates the child task 4 using the child task generation function 6, and monitors the business processing of the child task 4 using the child task monitoring function 7. The regenerated child task 4 executes the business program newly given from the control space 16.

One embodiment of the present invention can perform failure recovery in the virtual space 2 by operating as described above.

Typical processes performed by the post-failure process 9 in the embodiment described above include providing information about the failure 8 to the user;
For example, displaying failure summary information on a console device, editing and outputting the contents of a virtual area to a printer device, etc.
This includes releasing the resources related to the child task 4, for example, releasing the virtual area, releasing the reserved sequentially used resources, etc.

Further, a typical process performed by the child task failure recovery function 12 is to release the resources related to the parent task 5 secured between the child task generation function 6, the child task viewing function 7, and the termination determination function 11, for example, Releasing work area for task generation,
This includes closing opened file devices and printer devices.

Next, the operation of the embodiment of the present invention described above will be further explained with reference to the flowchart shown in FIG.

(1) After starting the process, the parent task 5 in the virtual space 2 first performs pre-processing of the oven of the printer device, etc., for example. This process allows the printer device to receive the processing results of the business program 3 executed by the child task 4, failure information, etc. (step 31).

(2) Next, the parent task 5 uses the child task generation function 6 to generate a child task 4, and then the task monitoring function 7 enters a state in which it waits for the child task 4 to contact the task disappearance event communication block. (steps 32 and 33).

(3) When the child task 4 is generated in step 32, the child task 4 starts processing, first registers an exit for failure handling, links to the business program, and executes the business program 3. Start (steps 37.38).

(4) When a failure 8 occurs during the execution of the business program 3, the process moves to the failure processing exit registered in step 37, turns on the failure flag, displays failure information on the console device, and displays the failure information on the printer. Post-failure processing such as editing and outputting the contents of the virtual area to the device is executed (steps 39 to 42).

(5) After executing the post-failure processing in steps 40 to 42 described above, the child task 4 notifies the event communication block for task disappearance of the disappearance of its own task, and disappears (step 43).

(6) When the parent task 5 detects the disappearance of the child task 4 from the child tasks 4 in step 33 described above, it confirms that the failure flag is ON (step 34).

(7) After confirming that the failure flag is ON, the parent task 5 performs post-failure processing, such as closing the printer, and returns to step 31.

This process is a process for reliably outputting the content output in step 42 to a previously reserved printer device (step 36).

(8) When all of the processing of the business program 3 performed by the child task 4 is completed without any failure, the a-task 5 stops the virtual space 2 and ends the processing (step 35).

As mentioned above, in the embodiment of the present invention, in the virtual space 2,
Child task 4 that executes business program 3 and child task 4
The parent task 5 monitors the disappearance of the child task 4, and even if the child task 4 disappears due to a program failure, the parent task can immediately regenerate a new child task. It becomes possible to execute the next business program without stopping it.

〔Effect of the invention〕

As explained above, according to the present invention, even if a failure occurs in a user program, etc. in an online system or database system using a multiplexed virtual memory method, the virtual space used by the user program, etc. is stopped. Since it is possible to recover from a fault in the virtual space without having to run the virtual space, the time required for fault recovery can be shortened, and troublesome system operations such as stopping and starting the virtual space can be eliminated. It is possible to improve the operational efficiency of the system.

[Brief explanation of the drawing]

FIG. 1 is a diagram explaining the functions and processing of an embodiment of the present invention;
FIG. 2 is a block diagram showing the system configuration of an embodiment of the present invention, and FIG. 3 is a flow chart explaining the operation. 1...Multiple virtual storage space, 2...
...Virtual space, 3... Business program, 4... Child task, 5...
... Parent task, 6... Child task generation function, 7... Child task monitoring function, 8...
...Failure, 9...Fault post-processing, 1
0... Child task disappearance notification, 11...
...End determination function, 12... Child task failure recovery function, 13... Terminal device,
Engineering 4...Communication control device, 15...
... Central processing unit, 16 ... Control space, 17 ... Business processing space group, 18 ...
・・・・・・Database, 19・・・・・・・・・
Virtual space failure recovery function. 1st Tawosa 4JI record, 7 openings 853 houses, 1 MaJara 7 Ramu Rimata parent task) task method, A I scolding figure 7 Terimata 7u JLl regret T this 8° floor 9: IO = ``I-9 people's way *ti weir Q 12 + 2 s 2 store'' ! L Shin System Ou Sho 1 Nakadokoro Aoi Ibara K ・Another #Kasaya I7゛Complex Processing Space Group 1B 7'-Taguis 19 5!!

Claims (1)

[Claims] 1. A data processing system using a multiplexed virtual memory method is equipped with a child task that performs work in a virtual space and a parent task that monitors the child task. A virtual space failure recovery method characterized in that a task regenerates child tasks without stopping the virtual space.