JPH02132529A - Automatic monitor switch controller - Google Patents

Abstract

PURPOSE:To shorten the service stop time by using a substitute processor to perform the business work which is so far carried out by a faulty processor and then using this processor again when the relevant fault is recovered for execution of a diagnostic program. CONSTITUTION:An automatic monitor switch controller 100 consists of a processor interface part 110, a switch control part 120, and a process part 130. The part 130 includes a memory part 130-1 which stores the control tables/blocks and a group of process programs and an execution control part 130-2 which interprets and carries out successively the instruction trains of the process programs. Then the controller 100 receives the information on occurrence of faults from a control system 200 and a business work system 220 and selects a processor based on the stored control information to substitute the faulty processor to perform its business work. The controller 100 instructs the substitution of the due process with addition of the substitute business work. In such a way, the service stop time can be shortened.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an automatic monitoring and switching control device for information processing systems such as computer systems, and in particular to automation of switching of management systems in multiple information processing system configurations. Regarding. [Prior art] Information processing system, i.e., computer system 24
Japanese Patent Laid-Open No. 7548/1983 discloses an operating method for a redundant system using a standby system in addition to the current system in order to minimize service stoppage time due to hourly service. Tokukai 1986
- The technology described in No. 7548 aims to save power and labor by keeping the standby system powered off and automatically switching over when a failure occurs in the active system.

[Problem to be solved by the invention]

Regarding the conventional technology, the technology described in Japanese Patent Application Laid-Open No. 60-7548 is such that in a redundant system consisting of an active system and a standby system, the power of the standby system is always turned off, and when a failure is detected in the active system, This relates to a control device that simultaneously resets the active system, connects the input/output devices to the standby system, turns on the power to the standby system, and automatically issues an initial program loading command. Therefore, the aim is to achieve power-saving operation in a redundant system consisting of an active system and a standby system. However, the above conventional technology requires (1) a failure detection procedure within the active system, (2) a job re-execution procedure after the standby system is started, (3) a processing procedure within the automatic switching control device, and (4) There is no disclosure of the power-on control method for the standby system, (5) the procedure for changing the connection relationship of input/output devices, and (6) the restart method after the active system is restored. leaving a problem.

Incidentally, as online services have become more popular in recent years, it is necessary to minimize service outage time due to failures. One solution is to form a system with multiple processors, and each processor has its own operating system.
This would be a method using a complex computer system in the form in which the system operates.

In a compound computer system configuration, one of the processors centrally manages the entire system (this is called the management system), and other processors handle business processing (
This is called a business system), but in this system configuration, issues remain such as how to detect and immediately automatically switch over a faulty system, and how to control reconnection after the faulty system is recovered.

Therefore, an object of the present invention is to provide a mechanism for monitoring the presence or absence of an abnormality in the management system, and a mechanism for delegating the functions of the management system to the business system when an abnormality is detected, in a multicomputer system configuration consisting of a management system and a business system. An object of the present invention is to provide an automatic monitoring switching control device equipped with a control means and a control means for running a diagnostic program in a management system. Another object of the present invention is to provide a control means for monitoring the presence or absence of an abnormality in the business system in a multicomputer system configuration consisting of a business system of a management system, and a system for causing the management system to perform the functions of the business system on behalf of the management system when an abnormality is detected. An object of the present invention is to provide an automatic monitoring switching control device equipped with a control means for running a diagnostic program on the working dog stem.

Another object of the present invention is to connect a group of input/output devices connected to a failed processor to another processor, and to switch a process currently being executed or waiting to be executed on the group of input/output devices to an alternative processor. An object of the present invention is to provide an automatic monitoring switching control device equipped with a control means for continuous processing by a processor.

Another object of the present invention is to provide automatic monitoring switching that includes a control means for incorporating a faulty processor into a multicomputer system configuration and re-executing the tasks assigned before the fault occurs when fault countermeasures for the faulty processor are completed. The purpose is to provide a control device.

Another object of the present invention is to provide an automatic monitoring and switching control device that makes it possible to realize stopless service using a multi-computer system configuration.

[Means to solve the problem]

In order to achieve the above object, the automatic monitoring switching control device of the present invention is connected to each computer (processor) of a compound computer system, and controls channel bus selection of each input/output device group. The automatic supervisory switching control device sequentially checks the operating status of each processor. Furthermore, the automatic monitoring switching control device of the present invention is also equipped with a mechanism for urgently receiving failure states independently detected within each computer system.

Therefore, when the automatic monitoring switching control device of the present invention detects a failure in each processor group or receives a notification of the occurrence of a failure from each processor, it issues a command to stop the failed processor and stops the processor. After
Based on the management information stored in advance in the automatic monitoring switching control device of the present invention, six substitute processors are determined which are equipped with a control means for selecting a processor to take over the task that was being executed by the faulty processor. Then, in order to switch the input/output device group connected to the faulty processor to the substitute processor, the channel path selection line is controlled via the switching control unit, and the connection relationship is switched.

Next, the automatic monitoring switching control device of the present invention issues a control means to the proxy processor to add the details of the proxy work and instruct the proxy processor to perform the process. As a result, the proxy processor performs the proxy business in addition to the main business.

On the other hand, for a processor in which a failure has occurred, the automatic monitoring and switching control device of the present invention includes a control means for instructing the processor to execute a diagnostic program. As a result, the faulty processor can be automatically diagnosed before maintenance personnel arrive. Once the maintenance personnel of the computer system has repaired the faulty part, the maintenance personnel only needs to input information to the effect of "recovery" from the console device of the service processor.

When the automatic supervisory switching control device receives this "recovery" information, it starts the operating system of the recovered processor. Generally, this process is performed by initial program loading (Initial Program Loading).
This is called em Loading (IPL) processing. Upon receiving a report that the IPL processing of the processor has been completed, the automatic monitoring and switching control device issues a command to the substitute processor to stop the task that it had previously performed, and the input/output device group also returns to the original processor. Change the connection to. Thereafter, control means is provided for causing the recovered processor to perform the original task.

This makes it possible to disconnect and reconnect a faulty processor without stopping service to end users in the operating mode of a complex computer system configuration, and also to realize stopless service.

[Effect]

The automatic monitoring and switching control device of the present invention operates by being connected to each processor via a dedicated line in a compound computer system configuration. Furthermore, channel path selection is controlled for each input/output device group.

Therefore, since monitoring of the operating state of each processor is carried out in parallel with the original operation of each processor, it does not induce malfunction in the conventional operation of each processor.

In addition, the automatic monitoring switching control device of the present invention automatically performs operations that were conventionally performed by maintenance personnel, system operators, and administrators of computer systems, and prevents frequent erroneous operations due to human error. This provides effects such as shortening the time required to change the system operation mode. Furthermore, although the automatic monitoring switching control device of the present invention is applied to a multiple processor configuration, it is also applicable to a single processor configuration.

In that case, by being able to automatically execute a diagnostic program after detecting a failure, the effect of reducing service outage time can be obtained.

〔Example〕

An embodiment of the present invention will be described below with reference to FIGS. 1 to 15-b. FIG. 1 is a diagram clearly showing the configuration of the automatic monitoring switching control device of the present invention, and also shows the connection relationship with each processor group and input/output device group. Number 10 in the diagram
0 is an automatic monitoring switching control device of the present invention, which is composed of a processor interface section 110, a switching control section 120, and a processing section 130. The processing unit 130 includes a memory unit 130-1 that stores a control table/control block and a group of processing programs, and an execution control unit 130-2 that sequentially interprets and executes a sequence of instructions of the processing program. 130-2 may be constructed of anything as long as it has an arithmetic function such as a microprocessor.

Referring to FIG. 1, in the configuration example of the compound computer system shown in FIG.
0 and the business system 220), but there is no limit to the number of processors.

The automatic monitoring switching control device 100 of the present invention is a management system 2
00 and business system 220 and signal line group Ql-0-Q
2-7, and instructs each input/output device group to select a channel path via a control signal line group 125 to channel path switching control devices 230-234. Note that the switching control section 120 is made up of a circuit that transmits the switching instruction signal Q18 from the processing section 130 to the corresponding signal line in the signal line group Q5. In this embodiment, FIG. 2 is a diagram showing the control circuit of the processor interface section 110, FIG. 3 is a diagram showing a list of commands issued to each processor from the automatic monitoring switching control device, and FIG. FIG. 5 is a diagram showing a list of request information from each processor, and is a diagram showing the configuration of a control table, control blocks, and processing program group of the processing section 130 in the automatic monitoring switching control device 100. 6 to 14 are flowcharts of each control table, details of control blocks, and processing programs. 15-a and 15-b are diagrams showing the operation flow of the monitoring program 210 within the management system 200 and the business system 220.

After describing the outline of the operation using FIG. 1, a detailed explanation will be given. In FIG. 1, among the symbols other than those described above, symbols 235 to 238 are input/output device groups, symbols Q6-0 to n9-1 are channel data lines, and symbol 201 is a service terminal of the management system 200. A processor 221 is a service processor of the business system.

Reference numerals 202 and 222 are operating systems (OS).
The OS 202 and ○S 222 may be the same OS or may be different OSs.

203 is a management program that operates under the management system 200, 223 is a business program that operates under the business system 220, and 224 is a business system 2.
20 management programs. Therefore, code 203
The management program 224 will be distinguished from the general management program, and the management program 224 from the sub-management program.

Note that the management system 200 and the business system 220 are connected by a signal line Q3. Instead of this signal line Q3,
Communication between the management system 200 and the business system 220 may be performed via the automatic monitoring switching control device 100 of the present invention. In this case, the commands and request information shown in FIGS. 3 and 4 shown later may be added.

System fault detection means include (1) a method in which the automatic monitoring switching control device 100 periodically inspects the operating computer system; (2) a method in which the operating computer system detects a fault and automatically monitors the computer system; There is a method of reporting to the switching control device 100. In this embodiment, the operation according to method (1) will be explained first, and then the operation according to method (2) will be explained.

The automatic monitoring switching control device 100 issues the command No. 5 in FIG. 3 via the signal line Ql-0.

The SVP 201 interprets commands from the automatic monitoring switching control device 100 and performs corresponding operations. The state of the hardware is inspected by the SVP 201, and control is passed to the monitoring program 210 to inspect the software including the OS. The monitoring program 210 inspects the operating state and sends the result to the SVP 201.
Return to. The SvP 201 returns the presence or absence of a failure, including the hardware test results, to the automatic monitoring switching control device 100 via the signal RIAQ2-0. The automatic monitoring switching control device 100 examines the data on the signal line Q2-0 to check for the occurrence of a failure. The above test will be repeated for the number of processors that make up the complex computer system. That is, the test signal is Ω1-i, and the response signal is Q2-i. In the example of FIG. 1, signal lines Q1-7 and Q2-7 are used for the seventh business system 220. On the other hand, if the computer system in operation (2) detects a fault, the request information shown in FIG.
It is reported to the automatic monitoring switching control device 100 via i. At this time, the automatic monitoring switching control device 100 operates independently (asynchronously) of the inspection procedure for each of the computer systems 2oO to 220.

The automatic monitoring switching device IR 1 0 0 is as described above (1
), (2), the processor interface unit 110 receives the test information reported via the line Q2-i. The processor interface unit 110 is connected to the signal line group Q11.
Processor identification information and fault information are sent to the processing unit 13 via
Pass to 0. The processor identification information can be identified by a processor number. If the processing unit 130 determines the state of the failure and determines that the processor that has caused the failure should be stopped,
A command to stop the processor is issued via the signal line group 010. 5 to 15 show the memory section 1 in the processing section 130.
30-1 shows the operation flow of the control table, control block, and processing program stored in 30-1, and will be described in detail later.

The stop command is sent to the faulty processor via @Q1-i, and after the stop processing is performed, an instruction to run the diagnostic program instead of the OS is sent to the line Ql-i.
issued through.

If the processor in which the failure has occurred is the management system 200, the overall management program 203 that collectively manages the multifunction computer system will be executed under the business system 220. This may be done by issuing a command to the sub-management program 224 to provide an integrated management function, or by issuing a command to the sub-management program 224 under the business system 220.
24 is not running, it is sufficient to newly run the overall management program. In any case, these commands are transmitted from the automatic monitoring switching control device 100 to the signal line 121.
-emitted via j. At this time, the signal &! Q2
-j is sent the type information of the agency work. Note that, before instructing the automatic monitoring switching control device 100 to perform the work on behalf of the user, the automatic monitoring switching control device 100
The input/output devices used by the failed system are
Reconnect to the system that performs the work on your behalf. In the example shown in FIG. 1, the business system 220 acts as a proxy for the management system 200, so the input/output devices 235 to 238 are the business system 220.
The connection is changed so that it can be used from 20 onwards.

For this purpose, the processing unit 1 in the automatic monitoring switching control device 100
A command signal from 30 is sent to the switching control section 120,
Channel path switching control device 23 via signal line group Q5
Indicated from 0 to 234. As a result, the channel path for the input/output device group is switched from the management system 200 to the business system 220, and the channel data line 07-
Data can be transmitted and received via Q1 to Q9-1. Here, the diagnostic program 20 for the management system 200
4 in the main memory (not shown) of the management system 200,
If the channel path of the input/output device 235 already exists, the channel path of the input/output device 235 is also switched. When the original o-gram 205 is loaded from the input/output device 235 into the main memory in the management system 200 and executed, the connection of the input/output device 235 is changed after the diagnostic program is executed.

Furthermore, this switching operation is not performed for input/output device groups that are already used in the business system 220.

Maintenance personnel for the computer system arrive while the diagnostic program 204 is running under the faulty processor, i.e., the management system 200 in the example of FIG. 1, and find that the cause of the fault is due to hardware. An analysis is performed to determine whether the problem is caused by software or not. If the cause is hardware, replace the circuit pointed out by the diagnostic program. If the problem is caused by software, fix the program that caused the problem. Thereafter, the maintenance person inputs information indicating "recovery" from the console device (not shown) of the SVP 201. This information is transmitted to the automatic supervisory switching control device 100 via the signal line Q2-0, and the processing unit 130 in the automatic supervisory switching control device 100 starts up the operating system of the processor.

This is the process p L (Initi
A command to start alProgram Loading) will be issued.

When the IPL processing of the management system 200 is completed, the signal line Q
2-0, the automatic monitoring switching control device 100 issues the command No. 11 in FIG. Signal line Ql-j (signal line Ql-7 in the example in Figure 1)
).

Next, in order to connect the input/output devices 235 to 238 to the original management system 200 again, the automatic monitoring switching control device
The command signal from the processing unit 130 in 00 is switched to the switching control unit 1.
20, and is instructed to the channel/bus switching control devices 230-234 via the signal line group Q5. As a result, the channel path for the input/output device group returns from the business system 220 to the management system.

The above is an outline of the operation of the automatic monitoring switching control device 11100 of the present invention. Next, details of the operation will be explained using the figures from FIG. 2 onwards. No. 211A is automatic monitoring switching control device 10
FIG. 2 is a circuit diagram of the processor interface unit 110 in FIG. In the figure, numeral 1 is a decoder DEC, and numeral 2 is an encoder ENC. Code 3 is command register CREG
, 4 is the data register DREG, and 5 is the command work register C: WREG. Code 6 is data
Work register DWREG, numerals 7 and 8 are gate circuits. Here, CWREG and DWREG are connected to each processor and signal lines Ql-i and Q2-i in FIG. 1, and form a pair for each processor. Therefore, in FIG. 2, subixes such as CWREGs and DWREG1 are added.

The code Q10-1 is a mode signal line sent from the processing unit 130 in FIG. 1, and the value is "1" when transmitting data to the processor 200 etc., and the value is "0" when receiving data. . Code 1210-2 has address buste, CIllREGi, DWREG
The value of i of + is specified and sent from the processing unit 130 in FIG. Code Ql3 is the command data bus, code 1
214, Ql5 is the data path. The code 016 has a value of tt 1 pp when request information is sent from the processor group via the signal line Q2-i, and the encoder E
It serves as an input to NC2, and causes the processor number to be output as signal ifllll-1. Therefore, the input signals of the encoder ENC2 are output signals from the DWREG* corresponding to each processor, as shown in the figure from 0 to ■.

Further, the symbols Q10-1 to Qlo-3 in FIG. 2 correspond to the signal line group Q10 in FIG. 1, and the symbols Ω11-1 to All-2
corresponds to the signal line group Q11.

FIG. 3 is a list of commands issued from the automatic monitoring switching control device 100 to each processor, and the values shown in FIG.
3. To select a target processor, the processor number is sent from the processing unit 130 in FIG. 1 to the signal line Ω10-2. This causes the processor number value to be changed to the decoder DE
It is decoded by CI, and the value of the output signal line corresponding to the processor number becomes R I I+. For example, if the signal is to be sent to the management system 200 in FIG. 1, the value of the signal line 1217 will be "'1" and the gate circuit 7 will be opened. the result,
The command value shown in Figure 3 is the command register CREG.
3, and the value is held in the command work register CWREGI via the command data bus Q13 and the gate circuit 7. On the other hand, if there is additional information accompanying the command, its value is stored in the data register D.
It is held in REQ4, and the value of signal line QIO-1 is “
'1', that is, the transmission mode is specified, the data work register D is sent via the data bus 14.
W R E G x is held. command work
The values of the register C W R E G 1 5 and the data work register D W R E G 1 6 are transmitted to the management system 200 via the signal line Ql-0.122-0, respectively.
6 or more is the method of sending control information when issuing a command to the management system 200, but when sending to other processor systems, the processor number is sent to the processing unit 130.
If it is sent from the signal line QIO-2 to the decoder circuit DE
C W R E G s corresponding to the output signal of CI,
D W R E G spirit is selected.

A response from the processor is received as follows. Again, for ease of explanation, the management system 200
The operation of the control circuit shown in FIG. 2 will be explained by taking as an example the response from .

The response information from the processor of the management system 200 is sent to the data work register DW via signals gQ2-0.
On the other hand, since the processing unit 130 in FIG. 1 recognizes that it is waiting for a response to a command to the processor, it sets the value of the signal line Q10-1 to "0".
and the processor number is sent to the signal line α10-2. As a result, the gate circuit 8 will open,
Data work register D W R E G x 6
The value of is transferred to data register DR via data bus Ql5.
Retained in EG4.

This is because the value of the signal line flo-1 is "'0", so data can be received.Next, if the processing unit 130 reads the value of the data register DREG4 via the signal line Q11-2, good.

Next, a description will be given of the operation of receiving a failure report when the management system 200 independently detects a system failure, that is, when it detects a failure asynchronously with the automatic monitoring switching control device [100]. Regarding notifications from the processors of the management system 200 and the business system 220, values corresponding to the fault information shown in FIG. 4 are sent out via the signal line Q2-i. Here, i
corresponds to the processor number. Here again, in order to make the explanation concise, a report from the processor of the management system 200 will be explained as an example. Failure notifications from the processor of the management system 200 are sent as data via the signal line Q2-0.
It is held in work register DWREG16. At this time, the value of the signal line 1216 becomes /I II I+, which is encoded by the encoder ENC2. As a result, the processor number that sent the request signal is output to the signal line Qll-1, and this processor number is transferred to the processing unit 130 in FIG.
You can receive it at

The processing unit 130 sends the previously received processor number to the signal line 11110-2, and also sends the previously received processor number to the signal line QIO-1.
Let the value of be “0”. As a result, the gate circuit 8 opens and the data work
After the request information code of register DWREGz6 is held in data register DREG4 via data bus Q15, it is transferred to processing unit 1 of FIG. 1 via data bus Q12-2.
The request information can be recognized at 30.

FIG. 3 shows a list of commands, which are sent to the corresponding processor via CREG3. At this time, DR
Additional information is set in EG4 as necessary and sent to the corresponding processor. FIG. 4 shows the contents of a request when request information from an operating computer system is reported to the automatic monitoring switching control device 100 via the signal line Q2-i.

FIG. 5 shows a processing section 130 in the automatic monitoring switching control device 100.
A control table stored in the memory unit 130-1 of
It shows the structure of the control block and processing program group.

6 shows the configuration of the processor management table, FIG. 7 shows the configuration of the system startup/stop order control block 141 shown in FIG. 5, and FIG. 8 shows the configuration of the device attribute control block 142. Referring to FIG. 5, the processor management table 140 holds the status of each processor, and each entry corresponds to each processor.

Therefore, each entry corresponding to a processor points to the system start/stop order control block 141 and the device attribute control block 142. That is, a system start/stop order control block 141 and a device attribute control block 142 exist for each processor.

On the other hand, the processing program group uses the above-mentioned control table and control block to perform processing.

The processing program group of the automatic monitoring switching control device 100 includes a main program 145, a power on/off program 146, a start/stop program 147, a processor switching program 148, a diagnostic program 149, a recovery switching program 150, and commands. It consists of a sending program 151.

Now, after explaining the control tables and control blocks shown in FIGS. 6 to 8, the operation of the processing program group in the automatic monitoring switching control device 100 will be explained using the operation flow diagrams shown in FIGS. 9 to 14. . Note that FIGS. 15-a and 15-b show the check program CHK2 10 shown in FIG.
FIG. 2 is an operation flow diagram. This check program CH
K2 1 0 operates on the processor side, and operates in cooperation with the service processor SVP 201 .

Therefore, it operates independently of the operating systems OS 202 and 222.

FIG. 6 shows the format of the processor management table 140,
One entry corresponds to each processor. Each entry consists of fields for processor status flag 11, replacement CPU number 12, business code 13, system startup/stop order control block address, and device attribute control block address, and indicates the operating status of each processor. After the inspection, the status is reflected in the status flag 11,
If a failure occurs, the processing program group shown in FIG. 5 will perform the process of switching the task to the processor indicated by the replacement CPU number 12.

FIG. 7 shows the format of the system start/stop order control block 141. This control block stores the procedures for starting and stopping the processors, and the processing program group interprets these and sends them to the processors 200 and 220 via the signal line Q2-i as necessary.
Send to.

FIG. 8 shows the format of the device attribute control block 142. This control block 142 controls the channel path address 1 of the input/output device group connected to the processor.
5 and channel bus address 1 from the alternate processor
6 is stored.

FIG. 9 is a processing flow diagram of the main program 145 shown in FIG. Referring to FIG. 9, in process 21, an interrupt wait state is entered for a certain period of time. Here, waiting for an external interrupt means waiting for a request signal from a processor group connected to the automatic monitoring switching control device 100. In the determination process 22, it is checked whether a request has occurred from the processor or not. If it is a request from the processor, the value of the signal line Ω11-1 shown in FIG. 2 is sent to the decoder DECI as the processor number. , set the value of signal line Q1〇-1 to “O”
Prepare to read DREG4 (processing 23).

On the other hand, if there is no request from the processor, processes 24 to 29 are executed.

In process 24, the processor management table 140 is accessed and preparations are made for checking the entries corresponding to each processor. Here, i is the repetition count. Therefore, in step 25, when i becomes equal to or greater than the number of valid entries n, the process returns to step 21 again. In process 26, the status flag 11 of the processor in question is checked, and the status is checked only for the processors that are in operation. In process 28, the value of processor number i is sent to DECI in FIG. 2, and the value of "operation check" in item number 5 of the command contents in FIG. 3 is sent to CREG3. Next, a command is sent to the corresponding processor by setting the value of the signal line J210-1 to "1". Thereafter, the value of the signal line Q10-1 is set to tL O 17, and preparations are made to receive the result.

In process 29, which is a common process after processes 23 and 28, the value of DREG4 in FIG. 2 is read.

If the value of DREG4 is "02", it is a normal report, and the process returns to step 25 to inspect the next processor.

On the other hand, if the value of DREG4 is not #l O 71,
This means that the factors shown in FIG.
”). If the result of the determination is that it is not a failure recovery report, in process 31 the status flag 11 shown in FIG.
"Failure", control is transferred to the stop routine of the start/stop program 147 to perform a stop process, and then control is transferred to the processor switching program 148.The processing flow of the processor switching program 148 is shown in FIG.
The main processing involves disconnecting a group of devices, connecting them to a replacement processor, and instructing a task changeover. When the control is returned from the processor switching program 148, the control is transferred to the diagnostic program 149, which automatically executes the diagnostic program for the processor that caused the fault (:j.%), and then proceeds to process 25. return.

If it is a failure recovery notification, in step 32 the status flag 11 in FIG. 6 is set to "active" and control is transferred to the recovery switching program 150. The processing flow of the recovery switching program 150 is shown in FIG. 13. In this recovery switching program 150, the processor that was performing the task is stopped, the device is disconnected, and then the corresponding processor is transferred to the processor that has recovered from the failure. Connect devices. Then, it instructs the recovered processor to "switch the business", that is, to resume the business. This allows the processor that has recovered from the failure to resume operations. The above is the operation of the processing program group shown in FIG. 5, and FIGS. 10 to 14 show the processing flow of the sub-program called from the main program 145. Figure 10 shows startup/
Processing flow diagram of stop processing of the stop program 147, 1st
1 is a processing flow diagram of the processor switching program 148, FIG. 12 is a processing flow diagram of the diagnostic program 149, and FIG. 13 is a processing flow diagram of the recovery switching program 150.
FIG. 14 is a processing flow diagram of the command sending program, and all the characteristic operations shown here are shown.

Next, check the check program CHK shown in Figure 1.
The operation of 210 will be explained. 15-a and 15-b are processing flow diagrams of the check program C H K 210. First, in the determination process 35, the monitoring device [i
It is determined whether a command from 1flOO has been issued. This can be done by checking whether a command code has been sent to the signal line Ql-i. When the command comes from the monitoring device, the process shown in FIG. 15-b is performed. The processing at this time will be explained later. If the command is not from the monitoring device gioo, processes 35 to 44 are executed.

If the occurrence of a hardware failure is detected in the determination process 36, the value of the request code REQ is set to 21, and the process 44
Execute. Here, the request code REQ is the value shown in FIG. Similarly, if an event corresponding to the determination condition occurs in determination processing 37 to determination processing 39, the value of the corresponding request code REQ is set. In process 44, the value of the request code REQ is sent back to the signal line Q2-i, and after being reported to the automatic monitoring switching control device 100, the process returns to process 35.

If the command is from the automatic monitoring switching control device 100, the processing shown in FIG. 15-b is performed. That is, process 46
The command code is sent from the signal line Ql-i by the signal line Q2.
- Obtain additional information from i. In branch process 47, the process branches to the process corresponding to the command code obtained in process 46. The numbers in Figure 15-b represent the command code numbers,
This number is a value corresponding to the command content shown in FIG.

Above, we have explained how to switch operations via the automatic monitoring switching control device and how to resume operations after recovery.
The signal line Q3 in the figure may be used to switch the business between the management system 200 and the business system 220. In this case, the signal line Q3 is connected from the management system 200 to the individual business systems 220, and the automatic monitoring switching control device 1i! The processing performed by the management system 200 and the business system 220 will be divided and performed by the management system 200 and the business system 220, respectively.

〔Effect of the invention〕

According to the present invention, in a compound computer system consisting of a plurality of processors, the automatic monitoring switching control device is equipped with a control means that constantly monitors the operating status of each processor, and is capable of detecting the occurrence of a failure in any one of the processors. When the fault is detected, the faulty processor has a control means that causes another substitute processor to execute the work that was being executed in the system, a control means that causes the faulty processor to execute a diagnostic program, and a control means that causes the faulty processor to execute a diagnostic program, and if the fault has been recovered, the faulty processor Since it is equipped with a control means that causes the original processor to execute the task that was being executed, it has the effect of reducing service stoppage time and shortening the failure recovery time of a failed processor.

Furthermore, the automatic monitoring switching control device of the present invention is designed to prevent frequent erroneous operations caused by human error, and to configure This has the effect of shortening the operation time for changes.

In addition, the first @ management system 200 and the business system 22
It is also possible to mutually monitor the operating status of the other system between the two systems. In that case, the alternatives are (1) a method via the signal line 3, and (2) a method via the automatic monitoring switching control device 100. In the case of (1), the processes shown in FIGS. 5 to 14 may be incorporated into the management program 203 and the sub-management program 224. On the other hand, in the case of (2), it is possible to increase the items of the command shown in Figure 3 and the request content shown in Figure 4, and their functions and processing can be easily inferred from the contents disclosed in this embodiment. It is something.

[Brief explanation of the drawing]

FIG. 1 is a diagram succinctly showing the configuration of the automatic monitoring switching control device of the present invention, FIG. 2 is a diagram showing the control circuit of the processor interface section of FIG. 1, and FIG. 3 is a diagram showing the configuration of the automatic monitoring switching control device of the present invention. FIG. 4 is a diagram showing a list of commands issued to each processor. FIG. 4 is a diagram showing a list of request information from each processor. FIG. 5 is a diagram showing a list of request information from each processor. FIG.
FIG. 6 is a diagram showing the configuration of the 30 control tables, control blocks, and processing program groups, FIG. 6 is a diagram showing the configuration of the processor management table, and FIG. 7 is the system startup/stop order control block shown in FIG. 5. 141, FIG. 8 is a diagram showing the configuration of the device attribute control block 142 shown in FIG.
9 is a diagram showing the processing flow of the main program 145 shown in FIG. 5. FIG. 10 is a diagram showing the processing flow of the stop processing of the start/stop program 147. FIG. 11 is a diagram showing the processing flow of the processor switching program 148, and FIG. 12 is a diagram showing the processing flow of the processor switching program 148.
9, FIG. 13 is a diagram showing the processing flow of the recovery switching program 150, FIG. 14 is a diagram showing the processing flow of the command sending program, and FIGS. 15 (a) and (b). is the check program C shown in Figure 1.
It is a diagram showing a processing flow of HK210. DESCRIPTION OF SYMBOLS 100... Automatic monitoring switching control device, 110... Processor interface part, 120... Switching control part, 130... Processing part, 200... Processor of management system, 201, 221... Service - Processor SVP, 203... Management program, 204... Diagnosis program, 210... Check program, 22
0... Processor of business system, 223... Business program, 224... Sub-management program. No. 1 Zllll Graduation No. S No. 1 No. 1 No. O Country No. 1 (α)

Claims (1)

[Scope of Claims] 1. In a composite consisting of a first information processing system having a management function and a second information processing system having a business processing function, means for monitoring the presence or absence of an abnormality in the first information processing system; , command means for switching the management function of the first information processing system to the second information processing system when an abnormality in the first information processing system is detected; and control for stopping the first information processing system. means, a control means for switching an input/output device group connected to the first information processing system to the second information processing system, and a command for starting a diagnostic program for the stopped first information processing system. An automatic monitoring switching control device characterized by comprising means. 2. The means for monitoring the presence or absence of an abnormality in the first information processing system operates independently of the first and second information processing systems, and storage means for storing the states of the first and second information processing systems. a storage means for storing identification information of the second information processing system when a failure of the first information processing system is detected; a state inspection commanding means for the second information processing system; and a task switching system. 2. The automatic monitoring switching control device according to claim 1, further comprising command means. 3. A control means that automatically sends a stop command to the first information processing system when a failure occurrence is detected in the first information processing system, and a command that automatically executes a diagnostic program. 2. The automatic monitoring switching control device according to claim 1, further comprising means. 4. When a status inspection command is issued to the first information processing system, the first information system has a processing means for determining whether its own system is operating normally or abnormally, and immediately issues a command based on the determination result. Claim 1 or 2 is characterized in that it is equipped with a processing means that responds to the control device that has issued the
The automatic monitoring switching control device described in Section 1. 5. In a complex system consisting of multiple information processing systems, a mechanism that sequentially inspects and monitors each information processing system for the presence or absence of an abnormality in its operating state, and a mechanism for checking and monitoring the presence or absence of abnormal operating conditions for each information processing system, as well as a mechanism for performing tasks on information processing systems that are operating abnormally. automatic monitoring switching control characterized by comprising a command means for stopping the information processing system, and a command means for issuing a command for causing the replacement information processing system to perform the work being executed by the information processing system that has caused the failure. Device. 6. In a complex system consisting of a plurality of information processing systems, a control means for determining normal operation and abnormal operation within each information processing system, a control means for reporting the determination result to an external control device, and a control means for receiving reports of abnormal operation. control means for selecting a replacement information processing system based on an abnormality report in an external control device; and a command for issuing a command to cause the selected information processing system to perform the work of the information processing system that has caused the failure. 1. An automatic monitoring switching control device comprising: a means for instructing an information processing system in which the failure has occurred to execute a diagnostic program. 7. In a complex system consisting of multiple information processing systems, when the work of the information processing system that has caused a failure is being performed by another information processing system, the information processing system that has caused the failure issues a request for recovery from the failure. means for receiving the report; a command means for issuing a command to the information processing system that was performing the work based on the fault recovery report to stop the work on its behalf; An automatic monitoring switching control device comprising: a control means for switching connections to a system; and a control means for instructing an information processing system that has recovered from a failure to resume operations. 8. In an automatic monitoring switching control device connected to a complex system consisting of multiple information processing systems, a specific control code is used to determine whether the information processing system is operating normally or abnormally. Any one of claims 1 to 4, characterized by comprising a control means for sending a response code to the information processing system, a control means for receiving a response code from the information processing system, and a control means for making a determination. One automatic monitoring switching control device. 9. In a complex system consisting of multiple information processing systems, the operations of the other information processing systems are performed between a first information processing system that centrally manages the entire complex system and a second information processing system that performs business processing functions. The system is equipped with a control means for monitoring the state, and a processing means for, if an abnormal operating state is detected, executing the work that was being executed by the information processing system that caused the abnormal operating state, on behalf of the information processing system that detected the abnormal operating state. An automatic monitoring switching control device characterized by: