JPH0235528A - Control system for virtual computer system - Google Patents

Abstract

PURPOSE:To correctly set taken-over information at the time of the restoring of a main operating system by starting a virtual computer monitor right before the main operating system finishes and inputting virtual input data. CONSTITUTION:If the main operating system 5 is disabled to operate more owing to trouble, its contents are reported to a system trouble detection part 6 registered previously in the system 5. The detection part 6 gathers and processing trouble information of the system 5 and reports the trouble of the system 5 to the virtual computer monitor right before the system 5 finishes. The monitor 7 on receiving the information makes a comparison to judge whether or not the contents of a specific register Z are a specific pattern Y. When so, the control is passed to the end processing control part 8 of the monitor 7 and the virtual computer 9 is informed of the end. The virtual computer 9 indicates the execution of end processing to an operating system 10.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a control method for a virtual computer system. In particular, the present invention relates to a control method for a job-based board-based computer system in which a virtual computer monitor operates as a process under the control of an operating system and simulates a plurality of virtual computers.

〔overview〕

The present invention is a control method for a virtual computer system that uses a special control program to create a plurality of virtual computers in one computer system and allows different operating systems to operate on each virtual computer. Control is transferred to the virtual machine monitor just before the main operating system becomes inoperable and terminates, and a termination notification is sent to the virtual machine running on the virtual machine monitor, which is registered in advance when the operating system on that virtual machine starts. By inputting the input data for the virtual machine, the operating system running on the virtual machine is terminated, and the recovery information is set correctly so that it will start up correctly when restarted.

[Conventional technology]

BACKGROUND ART Conventionally, a job-based board computer system has been used in which a virtual computer monitor that operates as a process under the control of an operating system simulates a plurality of virtual computers and runs an operating system on them.

[Problem that the invention seeks to solve]

In such conventional virtual computer systems, when the operating system becomes unable to operate any further due to a hardware or program failure, it executes termination processing, collects failure information, and shuts down the central processing unit. However, because the termination process of the operating system running on the virtual machine simulated by the virtual machine monitor has not been executed, the file status may be in an incomplete state or the takeover during recovery may be incomplete. There is a problem where the operating system on the virtual machine does not start up correctly when trying to run the operating system on the virtual machine again because the information is not set correctly.

In other words, because the operating system on the virtual machine stops without the termination process being executed, it has the disadvantage that when you try to run the operating system on the virtual machine again, it may not work properly. There is.

The present invention eliminates these drawbacks by ensuring that when the operating system becomes inoperable due to some kind of failure, it reliably executes termination processing, correctly sets the inheritance information upon recovery, and correctly bootstraps the operating system upon re-execution. The purpose is to provide a method that can be used.

[Means for solving problems]

The present invention provides a central processing unit, an input/output control device connected to a plurality of peripheral devices, a main storage device connected to the central processing unit and this input/output control device, and a main storage device on the main storage device. A system failure detection means for detecting a failure when a failure occurs in the operating system, and a plurality of virtual machines simulated by a virtual machine monitor, and the system detects the failure when the main operating system becomes inoperable due to a failure. In a control method for a virtual computer system including means for stopping a central processing unit, means for notifying the virtual machine monitor of termination of the operating system in the virtual machine, and when the virtual machine runs its own operating system. means for registering input data for giving a termination notification to the operating system;
The present invention is characterized by comprising means for inputting the input data from the input device of the virtual machine and instructing the operating system on the virtual machine to notify the completion when the main operating system is terminated.

[Effect]

While the virtual machine monitor is operating as a process under the control of the operating system, the main operating system may become unable to operate any further due to a hardware or program failure. At this time, just before the main operating system becomes inoperable and ends, the virtual machine monitor is notified of the end.

In response to this notification, the virtual machine monitor inputs input data for termination notification that has been registered in advance at the startup of the operating system/system included in the virtual machine to the virtual machine running by the virtual machine monitor from the virtual input device. Executes termination processing of the operating system included in the virtual machine.

Through this termination process, the file status is adjusted, the inheritance information at the time of recovery is set correctly, and the operating system can be started correctly when the operating system is executed again.

〔Example〕

Next, embodiments of the present invention will be described based on the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

The embodiment of the present invention includes a central processing unit 1, a main storage device 2,
It includes an input/output control device 3 and a peripheral device 4.

The main storage device 2 includes a main operating system 5, a virtual machine monitor 7, and a system failure detection unit 6,
The virtual computer monitor 7 has a simulated virtual computer 9,
The virtual machine 9 includes an operating system 10, termination notification input data (instruction data) 12 for terminating the operating system 10, and input data 12 for termination notification.
and a virtual input device 11 for inputting.

When the main operating system 5 becomes unable to operate any further due to a hardware failure or a program failure, the contents are reported to a system failure detection unit 6 registered in advance in the main operating system 5. The system failure detection unit 6 collects termination processing failure information of the main operating system 5, and notifies the virtual machine monitor 71 that the main operating system 5 has failed immediately before the system is to be stopped. As a means of notification, when the main operating system 5 transfers control to the virtual machine monitor 7, a specific register
This is implemented by setting a specific pattern Y to .

The virtual machine monitor 7 that has been notified of the failure compares whether the contents of the specific register Z match the specific pattern Y, and if they are the same, sends control to the virtual machine termination processing control unit 8 in the virtual machine monitor 7. Transition. The virtual machine termination processing control unit 8 instructs the virtual machine 9 to terminate. In the virtual machine 9, when the operating system 10 that is running is started on the virtual machine 9, input data 12 for termination notification registered in advance is input from the virtual input device 11 controlled by the virtual machine 9. Instructs the operating system 10 to execute termination processing. The virtual machine 9 then periodically monitors the termination of the operating system 10, and notifies the virtual machine termination processing control unit 8Δ2 when the operating system 10 has terminated.

The virtual machine termination process control unit 8 stops the central processing unit 1 when it receives notifications of the termination process of the operating system 10 from all of the operating virtual machines 9.

FIG. 2 is a flowchart showing the flow of operation of the embodiment of the present invention. The main operating system 5 enters a fault state, the system fault detection unit 6 of the main operating system 5 detects the fault, and the virtual machine monitor 7 and the virtual machine termination processing control unit 8 after notifying the virtual machine monitor 7 of the fault.
, and the operation of the virtual machine 9.

Steps a and j show the operations of the virtual machine monitor 7, steps b and f show the operations of the virtual machine termination processing control unit 8, and steps c to e show the operations of the virtual machine 9.

A pattern Y is set in the register X from the system failure detection unit 6 in the main operating system 5, and control of the virtual machine monitor 7 is transferred.

At the starting point, the virtual machine monitor 7 checks whether pattern Y is stored in register X (step a).

If the content is other than Y, the normal processing of the virtual machine monitor 7 is executed, and if the content is Y, the virtual machine termination processing control unit 8 is executed.

The virtual machine termination process control unit 8 notifies all virtual machines 9 that are operating at that time of the termination process (
Step b). The notification is carried out by displaying a completion notification in a part of the memory of the virtual machine 9.

The virtual machine 9 that has been notified of termination checks whether the operating system 10 is running (step C).
If it is running, input data 12 for termination notification registered in advance in the virtual machine 9 is input from the virtual input device 11 (step d).

The termination notification input data 12 on the virtual machine 9 is automatically created when the operating system 10 is started and the virtual machine 9 determines when the type and hardware configuration of the operating system 10 is defined. There are two methods: one in which the information is input by an operator, and the other in which the information is entered through operator intervention.

After that, the virtual machine 9 monitors the termination of the operating system 10 (step f). When the virtual machine 9 detects that the operating system 10 is missing, it notifies the virtual machine termination processing control unit 8 of this fact. The virtual machine termination processing control unit 8 waits for completion notifications from all the virtual machines 9 to which termination instructions were given in step b (step J). When all completion notifications are notified to the virtual machine termination processing control unit 8, this is notified to the virtual machine monitor 7 and the central processing unit 1 is stopped (step k
).

〔Effect of the invention〕

As explained above, according to the present invention, control is transferred to the virtual machine monitor immediately before the main operating system terminates, and the termination is registered in advance in the virtual machine running on the virtual machine monitor when the operating system is started. By inputting notification input data from a virtual input device, when the main operating system becomes unable to operate due to a failure, etc., the operating system can be reliably terminated, the file status adjusted, and recovered. This has the effect of correctly setting the time takeover information and allowing the operating system on the virtual machine to start up correctly when it is executed again.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the overall configuration of an embodiment of the present invention. FIG. 2 is a flowchart showing the operation flow of the embodiment of the present invention. 1...Central processing unit, 2...Main storage device, 3...
Input/output control device, 4... Peripheral device, 5... Main operating system, 6... System failure detection unit, 7
. . . Virtual machine monitor, 8 . . Virtual computer termination processing control unit, 9 . . . Virtual machine, 10 . instruction data).

Claims (1)

[Claims] 1. A central processing unit (1) and an input/output control device (3) connected to a plurality of peripheral devices (4).
), a main storage device (2) connected to this central processing unit and this input/output control device, and a device on this main storage device that detects a failure when it occurs in the main operating system (5). comprising a system fault detection means (6) and a plurality of virtual machines (9) simulated by a virtual machine monitor (7), and stops the central processing unit when the main operating system becomes inoperable due to a fault. In the control method for a virtual computer system, the control method includes means for notifying the virtual machine monitor of termination of the operating system of the virtual machine; and means for notifying the virtual machine monitor of termination of the operating system of the virtual machine; and means for registering input data for giving a termination notification, inputting the input data into the virtual machine monitor from the input device of the virtual machine when the main operating system is terminated, and controlling the operating system on the virtual machine. (10) A control method for a virtual computer system, comprising means for instructing a termination notification to a computer.