JPS6010666B2 - Computer system monitoring method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for monitoring a crab computer system comprising a plurality of central processing units.

Conventionally, in order to collect the processing contents and monitor the operating status of a computer system consisting of multiple central processing units, the first
As shown in the figure, the monitoring device SUP and the monitored processing device C
The PU-1 to CPU-n are connected one-to-one, and each processing unit CPU-1 to CPU-n transfers processing contents, etc. between the interface units 1 and ln of the SUP using the program PA. I was doing it.

Therefore, in a monitoring device, as the number of monitored target devices increases, the interface part for collecting processing contents and monitoring becomes enormous, and the monitoring device itself becomes large. Large-scale remodeling was necessary. An object of the present invention is to eliminate the above-mentioned drawbacks, and to provide a small-sized monitoring device that can easily follow an increase in the number of monitored devices without increasing the number of interface units even if the number of monitored devices increases. It is in.

Therefore, the feature of the present invention is that a plurality of processing devices to be monitored are connected to each other through an interface that was conventionally connected to a monitoring device so that processing contents can be collected from a specific processing device and operation can be monitored. , a specific processing device and a monitoring device are connected using the same interface as in the past.

Then, a specific processing device has a program that collects and monitors the processing contents of other processing devices, exchanges information about the collected processing contents and monitoring status with the monitoring device, and processes monitoring operations from the monitoring device. The program to be carried out is hawked, and
A processing device connected to a specific processing device has a built-in program for processing the processing contents and monitoring operations of the processing device. This allows a specific processing device to perform operations that were conventionally performed by a monitoring device, and there is no need to increase the number of interface units of the monitoring device even if the number of devices to be monitored increases. Next, one embodiment of the present invention will be described in detail using the drawings.

FIG. 2 is a block diagram of an embodiment of the present invention.

In FIG. 2, the monitoring device SUP is the central processing unit CPU.
-1 via an input/output interface WTF used for normal input/output device control, and is controlled as one input/output device by the program PA in the central processing unit CPU-1. On the other hand, the central processing unit CPU-
2 to CPU-n are connected in series by an input/output interface INTF via a central processing unit CPU-1 and an inter-channel coupling device CTCA for controlling each central processing unit from the central processing unit CPU-1 as an input/output device. connected to. The central processing unit CPU-1 has the above-mentioned program PA.
In addition, a program PB is placed to control the central processing units CPU-2 to CPU-n as input/output devices, and to collect processing contents and monitor operations of the CPU-2 to CPU-n. Correspondingly, a program PC for operating as an input/output device for the CPU-1 according to instructions from the CPU-1 is installed in the CPU-2 to CPU-n. At that time, CPU-
Instructions between CPU-1 and CPU-2 to CPU-n include CPU-1 and CPU-2 to CPU-n.
A read command for collecting the processing contents of CPU-2 to CPU-n and a test 1/0 command for monitoring the operation of CPU-2 to CPU-n as normal input/output devices are defined.
In addition, between the CPU-1 and the monitoring device SUP, the data collected from the CPU-2 to CPU-n by the program PA of the CPU-1 is provided.
A write command for transferring processing information and monitoring information to the SU and a response command to an interrupt from the SUP are defined. Figure 3 is a schematic configuration diagram of the monitoring device SUP.
an interface unit 1 for exchanging information with the CPU-1;
A monitor interrupt unit 2 monitors the CPU-1 in response to a write command from the CPU-1, a timer unit 3 provides a time signal to the monitor interrupt unit 2, and a C signal sent in response to a write command from the CPU-1.
A data identification unit 4 that identifies the processing content or monitoring information data of the PU-12 to CPU-n, an abnormality alarm unit 5 that displays an abnormality or issues an alarm based on the data from the data identification unit 4, and a data display unit that displays the processing content. It is equipped with 6th grade.

Now, monitor the entire system as follows.

First, CPU-1 starts monitoring device SUP by program PA.
A write command is issued to the CPU-1, and the SUP starts monitoring the CPU-1. When the SUP starts monitoring based on the above write command, the monitoring interrupt section 2, which receives the rise signal from the timer section, interrupts the CPU-1 at regular intervals, and issues a response command to the interrupt. The CPU-1 is monitored by monitoring whether it is boiled within a certain period of time. On the other hand, when CPU-1 starts monitoring SUP, program PB causes CPU-2 to
Test 1/0 instructions are issued to CPU-n sequentially.
The CPU-2 to CPU-n are connected to the CPU-1 via a channel coupling device CTCA, and the CTCA is managed by a device number at the same time as a normal input/output device.
Therefore, the CPU-1 can detect the operating states of the CPU-2 to CPU-n by sequentially specifying the device numbers in the program PB and issuing test 1/0 commands. The program PCs in CPU-2 to CPU-n are
It controls the response when the PU receives a test 1/0 command and transmits the information to the CPU-1 when stopping the operation. CPU-1 uses the above program PB
monitors CPU-2 to CPU-n, issues a test 1/0 command with the CTCA device number corresponding to CPU-2 to CPU-n at regular intervals, and checks whether there is a response or not and the contents of the response. As a result, when an abnormality is detected in CPU-2 to CPU-n, the information is passed to the PA.

As a result, program PA issues a write command and P
SUP the monitoring information of CPU-2 to CPU-n obtained from B
send to. In the SUP, the data identification unit 4 determines the content based on the information obtained from the CPU-1, and indirectly sends the information to the CPU-1.
2 - Detects an abnormality in CPU-n. Also, as mentioned above,
SUP is monitoring CPU-1, and
It is also possible to detect an abnormality of 1. Therefore, SUP
This allows CPU-1 to CPU-n to be monitored. Note that the operation of collecting the processing contents of the CPU-1 to CPU-n is also the same as the monitoring operation. FIG. 4 shows another embodiment of the present invention, which is an improvement on the embodiment shown in FIG.

That is, in the embodiment shown in FIG. 2, if CPU-1 fails, it becomes impossible to collect and monitor the processing contents of CPU-2 to CPU-n. In order to improve this point, in Figure 4, the number of CPUs connected to the SUP is increased by one, and normally CPU-1
Collects and monitors the processing contents of -2 to CPU-n and exchanges SUP information, and in the event of a failure of CPU-1, CPU-n
2 also collects and monitors the processing contents of CPU-3 to CPU-n and exchanges information with SUP.
In FIG. 4, the same reference numerals as in FIG. 2 have the same functions. Further, the program PX placed in the CPU-2 operates in the same way as the CPU-1, with PC being placed when the interface switching unit SW is on the a side, and PA and PB being placed when it is on the b side. As is clear from the above description, the following effects can be obtained by adopting the monitoring method of the present invention.

[1] The interface part with the monitored system is 1
Therefore, even if the number of processing devices to be monitored increases, it can be easily followed without modifying the monitoring device, and the monitoring device can be small or inexpensive. ■ For the system to be monitored, the already developed program can be used as is, and the number of target processing devices can be easily increased.

[BRIEF DESCRIPTION OF THE DRAWINGS] FIG. 1 is a diagram showing a conventional monitoring system, FIG. 2 is a diagram showing an embodiment of the monitoring system according to the present invention, and FIG. 3 is a schematic diagram of the monitoring device shown in FIG. 2. FIG. 4 is a diagram showing another embodiment of the monitoring system according to the present invention. CPU-1 to CPU-n...Central processing unit, P
A, PB, PC...0 program, CTCA...
... Channel coupling device, SUP ... Monitoring device, INTF ... Interface line, 1, Interface section. Figure 1 Figure 2 Figure 3 Figure 4

Claims (1)

[Claims] 1. A computer system comprising a plurality of central processing units 1 to n and a monitoring device for monitoring the computer system, wherein one central processing unit 1 and other central processing units 2 to n is connected to the central processing unit 1 through an interface that can monitor the operation and collect the processing contents, so that the central processing unit 1 can monitor the operation of the central processing units 2 to n and collect the processing contents, and the central processing unit 1 and a monitoring device to exchange processing contents and monitoring information collected from the central processing units 2 to n. 2. A patent claim characterized in that the monitoring device exchanges processing contents and monitoring information of the central processing units 2 to n with the central processing unit 1, and also monitors the operation of the central processing unit 1. Scope of the above: A monitoring method for an electronic computer system according to item 1.