JPH031694B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention relates to an operation automation method in a virtual computer system that can automatically respond to messages from a virtual computer and input commands to the virtual computer. be.

[Conventional technology and problems]

Conventionally, AOF (Automatic Operation Facility)
etc., the operation automation of the OS (Operation System) is being carried out, but in systems using this, initial program loading, master scheduler initialization, job entry system Operations such as startup are required, and automatic operations such as independent utilities cannot be performed.

In a virtual computer system, multiple virtual computers are managed by one control program. Some of these virtual computers cannot use the automatic operation function.

[Purpose of the invention]

An object of the present invention is to provide an operation automation method in a virtual computer system that enables automatic operation of a virtual machine immediately after starting a control program in the virtual computer system.

[Structure of the invention]

Therefore, the operation automation method for a virtual computer system of the present invention, in a virtual computer system having a plurality of virtual machines, intercepts input/output requests to the virtual console issued by the server virtual machine and the virtual machine under test. The server virtual machine is equipped with input/output request notification means for notifying a virtual machine, and operation notification means for notifying an operation from the server virtual machine to the tested computer, It is characterized by having a function of analyzing input/output requests to the console and notifying the operation notification means of the corresponding operation, and a function of logging messages from the virtual machine under test.

[Embodiments of the invention]

Hereinafter, the present invention will be explained with reference to the drawings.

FIG. 1 is a conceptual diagram of one embodiment of the present invention. In FIG. 1, 1 is a control program called AVM (hereinafter simply referred to as AVM), 2-1 to 2-n are virtual computers (hereinafter simply referred to as VM),
3 is a server virtual computer (hereinafter simply referred to as server VM), 4-1 to 4-n are virtual consoles,
Reference numeral 5 indicates an exit routine, 6 indicates a virtual punch, 7 indicates a file for message logging data, and 8 indicates a file for operation data. In addition, virtual console 4-1 to 4-
n is a dummy value. Furthermore, AVM is
It stands for Advanced Virtual Machine.

The system in Figure 1 connects VM2 via AVM1.
-i (i=1, 2, ... or n) and server VM3
Data is exchanged between server VM3 and server VM3.
This is a system that controls VM2-i. Specifically, the i/0 command issued from VM2-i to virtual console 4-i is sent to the server via VM1.
VM3 intercepts the message sent from VM2-i, and sends input data and commands to the server VM3.
The VM2-i is controlled from the VM2-i.

The exit routine 25 of the AVM 1 has a function of notifying the server VM of a virtual console input/output request for the tested VM 2-i, and a function of notifying the tested VM 2-i of an operation from the server VM 3. . The server VM has a function of analyzing input/output requests to the virtual console 4-i of the tested VM 2-i and notifying the exit routine 5 of the corresponding operation, and
It has functions such as logging messages of 2-i.

FIG. 2 explains the flow of processing performed in response to a WRiTE request from the VM under test.
Note that the CP body is the AVM excluding exit routine 5.
It means part 1. When the VM2-i under test issues a WRjTE request, a program check interrupt is issued to the hardware. The CP body checks whether the WRiTE request is for the DUMMY console or not, and if yes, passes control to exit routine 5. Exit routine 5
creates a pseudo attention to server VM3 and notifies server VM3 of this. server VM
Upon receiving this attention, No. 3 issues a DiAGNOSE (diagnosis) command to accept the request. When the CP body receives this DiAGNOSE command, the server
It is checked whether or not it is a DiAGNOSE command from VM3, and if Yes, control is passed to exit routine 5. Exit routine 5 passes the console address and WRiTE request to server VM3, and
Notify the tested VM 2-i of completion of startup in response to the WRiTE SiO command. This is done by setting condition code CC to zero. The server VM3 issues a WRiTE DiAGNOSE command. The CP body takes in this DiAGNOSE command and checks whether it is a DiAGNOSE command from the server VM3. If Yes, control is transferred to exit routine 5, and exit routine 5 transfers the WRiTE data to server VM3 and at the same time
Channel end CE and device end DE are sent to 2-i. The server VM3 writes the sent WRiTE data to the message logging data file 7.

FIG. 3 is a diagram illustrating the flow of processing performed when commands are input to the VM under test. When attempting to input a command to the VM2-i under test, first the server VM3 receives attention.
Send the DiAGNOSE command. When the DiAGNOSE command is sent, the CP main body receives the command from server VM3.
It is checked whether it is a DiAGNOSE command or not, and if Yes, control is transferred to exit routine 5. Exit routine 5
creates a pseudo attention to the VM2-i under test, sends it to the server VM3, and reflects this attention on the VM2-i under test. When the tested VM 2-i receives this attention notification, it issues a READ request. The CP body checks whether this READ request is for the DUMMY console or not, and if Yes, transfers control to exit routine 5. Exit routine 5 creates a pseudo attention to the server VM3 and reflects this pseudo attention to the server VM.
When the server VM receives this attention notification, it issues a DiAGNOSE command to accept the request.
In the CP main body, this DiAGNOSE command is
If yes, control is transferred to exit routine 5. Exit routine 5
notifies server VM3 of the console address and READ request, and at the same time sends the READ request.
Tested VM2-i to check startup completion for SiO command
to notify. When the server VM3 receives these console addresses and READ requests, it reads the operation data from the operation data file 8, and then reads the READ request.
Send the DiAGNOSE command. The CP body is
It is checked whether the DiAGNOSE command is from the server VM3, and if Yes, control is transferred to the exit routine 5. Exit routine 5 is the server VM
3 transfers the operation data read from file 8 to the tested VM2-i, and after the data is completed, notifies the channel end CE and device end DE, and at the same time returns a response to the DiAGNOSE command to the server VM. . Exit routine 5 transfers the operation data to the tested VM2-i and at the same time executes READ.
Notify server VM3 of completion of the DiAGNOSE command. The server VM3 writes in the message logging data file 7 that the operation data has been sent in response to the READ request.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, it is possible to automatically operate all virtual machines immediately after starting the control program.

[Brief explanation of the drawing]

Figure 1 is a conceptual diagram of one embodiment of the present invention, Figure 2 is a diagram explaining the flow of processing performed in response to a WRiTE request from a VM under test, and Figure 3 is a diagram illustrating the flow of processing performed in response to a WRiTE request from a VM under test.
FIG. 2 is a diagram illustrating the flow of processing performed when inputting a command to a VM. 1...Control program called AVM, 2-1
to 2-n...virtual computer, 3...server virtual computer, 4-1 to 4-n...virtual console, 5...
Exit routine, 6...virtual punch, 7...file for message logging data, 8...file for operation data.

Claims (1)

[Claims] 1. In a virtual computer system having a plurality of virtual machines, a server virtual machine, an input/output request notification means that intercepts an input/output request issued by the virtual machine under test to the virtual console and notifies the server virtual machine, and and an operation notification means for notifying the computer under test of an operation from the computer,
The server virtual machine has a function of analyzing input/output requests from the virtual machine under test to the virtual console and notifying the corresponding operation to the operation notification means, and a function of logging messages from the virtual machine under test. An operation automation method in a virtual computer system characterized by comprising: