JP2000293401A - Distributed processing computer system provided with software abnormality processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect and specify an abnormal part of software of a distributed processing computer system. SOLUTION: An abnormal part of software is detected and specified by providing relevant output information tables 14, 24, 34 and 44 where relevant output information defining an output message corresponding to the processing condition of an input message is stored in each program where a series of processing of an information processing system is executed with plural programs 12, 22, 32 and 42 and software abnormality processors 13, 23, 33 and 43 which has a selecting means which selects relevant output information about one program from the relevant output information tables and a comparing and collating means which compares and collates an output message of executed results of the one program with the relevant output information selected by the selecting means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a computer system, and more particularly, to a software abnormality in a distributed processing computer system in which a plurality of computers are connected on a network and a series of processes are executed by a plurality of programs. The present invention relates to a detection method and device.

[0002]

2. Description of the Related Art A distributed processing method in which a plurality of processing devices (computers) are connected to a common signal transmission line and a series of processing is distributed and processed by the plurality of processing devices is widely known. In such a distributed processing computer system, JP-A-57-14
According to Japanese Patent No. 6361 (Prior Art 1), a plurality of programs corresponding to distributed processing are stored in a distributed manner in a plurality of processing devices, and each program is started by executing the program. The necessary data is taken from the transmission line into its own processing device and is performed when it is complete. According to this method, a processing device for management for managing the entire distributed processing system is unnecessary, and a series of processes can be distributed by each processing device and executed in an asynchronous free run.

[0003] In such an autonomous distributed computer system, various methods have been proposed for coping with the occurrence of an abnormality in the execution of a program or the like. For example, JP-A-57-1
According to the “abnormality handling method of the autonomous decentralized system” described in Japanese Patent No. 46361 (prior art 2), when a message output request exceeding the allowable density is made, the message output request is delayed for a certain time, It has been proposed to prevent the I / O message buffer from filling up.

[0004] In addition, by monitoring the usage of the reception buffer, storing the transmission message in the retransmission buffer,
The retransmission means of the message in the retransmission buffer makes it possible to avoid data loss due to the reception buffer being full. In addition, it detects programs that are transmitting messages excessively and treats them as abnormal. It is known that this is done (prior art 3).

It is also known that a plurality of processing devices can automatically recognize a change in the configuration of an application program that executes a series of processes in a distributed manner (prior art 4). Furthermore, when a series of processes are executed by a plurality of programs, a method of statically recognizing a relationship between software modules that grasp connections between the programs has been proposed (prior art 5).

On the other hand, Japanese Patent Application Laid-Open No. 2-277126 (prior art 6) discloses a method of monitoring a program.
A method and apparatus for automatically generating a control program for a computer control system are described. Also, Japanese Patent Laid-Open No. 4-8
Japanese Patent No. 4327 (Prior Art 7) proposes a control system in which a plant control system monitors a program in real time in the form of a plant operation plan. That is, a method of monitoring a program in a plant controller for a plant control system that operates in real time mainly in the form of a plant operation plan has been proposed.

[0007]

In a plant information management system, an operation management system, a wide area information processing system, and the like using a computer, a large amount of computer software (programs) is used and operates in real time. are doing. In these systems, software failures can be caused by equipment shutdown or
Immediate remedial action is required because it leads to various service outages.

However, in a distributed processing computer system operating in real time, it is extremely difficult to identify an abnormal part in a short time from a large amount of computer software. In addition, localization of abnormalities, self-healing, and autonomous / cooperative functions have been demanded, such as automation of a plant by labor saving / manpower saving or unmanned computer equipment.

For example, in the prior art 1, the input side program checks the consistency of the input message and discards an abnormal message. However, if the output program does not output a message that should be output, a series of processes may be interrupted. This problem cannot be solved even if a redundant system is constituted by a plurality of programs.

In the prior arts 2 and 3, the receiving buffer can be prevented from being full. However, similarly to the prior art 1, an abnormal case in which a message that should be output by the output side program is not output is considered. Not. In particular, since the related arts 1 to 3 do not grasp the connection relationship between programs, there is a problem in maintenance and operation of software at the time of abnormality.

As a method for solving such a problem, in the prior art 4, the configuration change of the application program configuration can be automatically recognized.
In this document, the relation between a series of programs can be statically recognized. However, even in these prior arts, since no consideration is given to detecting abnormalities in a program operating in real time, software abnormalities in a computer system operating in real time cannot be monitored online. It has not solved the software maintenance problem in the system.

On the other hand, the prior arts 6 and 7 propose a method of monitoring a program in a plant controller for a plant control system, which operates in real time, mainly in the form of a plant operation plan. However, this method has a problem in that when a series of processing is distributed to a plurality of programs by a plurality of processing devices and processed, a connection between programs synchronized with input / output data cannot be grasped.

An object of the present invention is to enable an abnormal part of software to be detected in a distributed processing computer system operating in real time.

It is another object of the present invention to localize an abnormality by protecting the system based on the detected abnormality.

It is another object of the present invention to make possible localization, autonomy, and coordination of abnormalities by complementing detected abnormalities.

Another object of the present invention is to make it possible to easily confirm an abnormal part and an affected area of software as visual information in a computer system operating in real time.

[0017]

SUMMARY OF THE INVENTION The present invention aims to achieve the above object by the following means and a combination of the means.

First, a related output information table storing related output information defining output messages corresponding to processing conditions of an input message for each program that executes a series of processes of the information processing system by a plurality of programs; Selecting means for selecting related output information related to one program from the related output information table; comparing the output message of the execution result of the one program with the related output information selected by the selecting means to output the output message; A software abnormality processing device having a comparison and collation means for detecting the presence / absence of the abnormality.

That is, the output message output from each program is known in advance corresponding to the input message. By defining this as relevant output information, the output message to be output can be output to the relevant output message. If it is not defined in the information, the output message can be determined to be abnormal. Further, since it is determined whether or not the output messages of all the programs are abnormal, there is no abnormality in the input, so that it can be considered that there is some abnormality in the program that outputs the abnormal output message, and the abnormal part can be specified.

In particular, the software abnormality processing device includes an input / output definition information table storing attribute information relating to a program and input / output definition information defining a connection relationship between the program and the input message and the output message; Identifying an abnormal part by including an abnormal part detecting means for specifying an abnormal part and an affected area relating to the output message abnormality detected by the comparing and collating means with reference to the input / output definition information table it can.

In the above case, the software error detection device refers to the related output information, and if an output message output from one program is not defined in the related output information, the software outputs the output message as an invalid message. It is preferable to have an abnormality processing means for preventing output if any. According to this, it is possible to prevent an abnormal operation due to the abnormal output message, so that the system can be protected and the abnormality can be localized.

Further, in the above case, the software abnormality processing device detects the abnormality of the program by the comparison / matching means not outputting the output message defined in the output related information, which should be output originally. At this time, by providing an alternative message output unit that outputs an alternative message instead of the output message related to the abnormality, the abnormality can be complemented, and localization, autonomy, and cooperation of the abnormality can be performed. . Further, interruption of a series of processing can be prevented. In this case, the input section of each program may be provided with an alternative message input means for identifying whether the input message is an alternative message or an original message.

[0023] Further, a display device is provided on the network, and display control means is provided for displaying a connection relationship of another application program which operates upon input of an output message of one application program by picture information,
The display control means can display the abnormal part and the affected area as visual display information. According to this, in a computer system operating in real time, an abnormal part and an affected range of software can be easily confirmed as visual information. In this case, the output of the alternative message can be displayed as visual display information.

[0024]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described based on embodiments. FIG. 1 shows an overall system configuration of an embodiment in which a software monitoring device according to the features of the present invention is applied to a computer system for process control that manages information on rolling plant equipment. The present invention is not limited to this, and can be applied to any business field system as long as a plurality of computers are connected on a network and a series of processes are executed by a plurality of programs.

As shown in FIG. 1, a plurality (4 in the illustrated example)
) Process control computers 10, 20, 30, 40
And one monitoring personal computer 60 are connected to the bus-type network 6 via the network control devices 1 to 5. Each of the control computers 10 to 40 and the monitoring personal computer 60 includes operating systems 11, 21, 31, 41, 61 in association with the network controllers 1 to 5, respectively. The network control devices 1 to 5 perform data exchange control between the control computers 10 to 40 and the monitoring personal computer 60 and the network 6. Each of the operating systems 11 to 41 selectively receives the message 9 with the content code, activates each of the application programs (hereinafter, abbreviated as APP) 12, 22, 32, 42, 45, a message driving unit, and broadcast communication. Has a function. APPs 12, 22, 32, 4
2, 45 selectively receive a required message in accordance with the content code indicating the content of the message, execute a process corresponding to the message, and output the process result. The content code is also added to the message that outputs the processing result, and the message is broadcasted. These points are realized by a known method.

Each of the APPs 12 of the control computers 10 to 40,
The processing results of 22, 32, 42, and 45 are transmitted to the network 6 via the network control devices 1 to 4 as the message 9 with the content code, and are respectively connected to the control computers 10 to 40 or the monitoring personal computers connected to the network 6. It is taken into the computer 60. Each control computer 10
40 to 40 selectively receive the message 9 with the content code by the operating systems 11 to 41 stored in their own control computers, and
Message driven. As a result, each APP executes a predetermined process, and outputs a result of the process as a message with a content code (hereinafter, referred to as a function code) via the operating systems 11 to 41.

Each of the control computers 10, 20, 30, and 40 has an APP1 for managing information on rolling plant equipment.
2, 22, 32, 42 and 45 are stored. APP
Reference numeral 12 has a basic internal configuration including a message input processing unit 121, an application processing unit 122, a message output processing unit 124, and the like. This configuration is based on APP22, APP3,
The same applies to 2, 42, and 45. Control computer 1
0 includes a software abnormality detection processing unit 13, which comprises a message output monitoring unit 131 and a timer monitoring processing unit 132.
Is configured. Similarly, the control computer 20,
Software abnormality detection processing units 23, 33, and 43 having the same configuration are also incorporated in 30, 40. The control computers 10, 20, 30, and 40 respectively include an abnormality detection management information table 14, which stores abnormality detection management information,
24, 34 and 44 are provided. The abnormality detection management information includes an input / output message between APPs stored in each control computer, input / output definition information indicating a connection relationship between APPs, and a message output matrix for monitoring a message output by each APP. (Related output information) and the like. The monitoring personal computer 60 includes a software abnormal part display processing unit 62, and the software abnormal detection processing units 13, 2 incorporated in each control computer.
The abnormality detection information 7 and 8 with content code output from 3, 33 and 43 are input via the operating system 61. Then, based on the input abnormality detection information 7 and 8, an abnormal part of the software and its influence relation are displayed on the monitor CRT 63 in a system flow format. The monitoring personal computer 60 is operated by an operator to designate an online monitor or the like.
Input means 6 including a keyboard (K / B), a mouse, etc.
4 are provided. In the system for detecting a software abnormal part according to the present invention, the functions of the respective parts in FIG. 1 are organically combined to perform the functions.

FIG. 2 shows the entire structure of the function code 9 which is a message with a content code. In the figure, reference numeral 91 denotes a function code number referred to as a content code set in the function code, and reference numeral 92 denotes a transmission source node number, that is, a number unique to the system of the transmission source computer. If necessary, each module-related information held by the computer can be indexed by using the source node number 92. Reference numeral 93 denotes a processing result unique to each APP.

Next, abnormality detection management information (table information) used in the present embodiment will be described with reference to FIGS. FIG. 3 shows the general configuration of the abnormality detection management information tables 14 to 44 of the control computers 10 to 40. The input / output definition information 141, 241, 341, 441,
Message output matrix (related output information table) 1
42, 242, 342, 442, and software abnormality monitoring timer table 145, 245, 345, 445
It consists of. In the figure, input / output definition information 141, 2
Reference numerals 41, 341, and 441 denote general configurations of an input / output definition table, and APP (hereinafter, referred to as an APP) stored in each computer.
This is for storing the relationship between input / output function codes and input / output function codes.
The input / output definition information 141, 241, 341 and 441 are function module names as attribute information of each function module, input function codes input by the function modules, addresses of message output matrices to be used by the function modules, and function modules. Function code information such as output function codes for the number of output function codes for outputting an output message substitution enable / disable flag and a processing result when there is no message output due to the software abnormality of the number of function modules. Here, it is assumed that each functional module has one input and the processing result has a relation of n outputs. In addition, it is assumed that each functional module is basically driven by a message with only one functional code.

Message output matrices 142, 24
Reference numerals 2, 342, and 442 are used as monitoring information of unauthorized message output used by software abnormality detection processing associated with each functional module. These message output matrices are set so as to correspond to each function module as shown in the figure. That is, the function codes (for example, fc11, fc12, fc13,... Fc1m,.
, Fcn1, fcn2, fcn3, ... fcnm) are set and stored for each output pattern (1, ..., n).

Here, the input / output definition tables 141 and 24
1, 341, 441 and message output matrix 1
42, 242, 342, and 442 are used to change statically or configuration information from system information used by the operating systems 11, 21, 31, and 41 having a message driving unit and a broadcast communication function to start an application. It is preferable to have a mechanism in which the content is changed synchronously each time it is performed. Also, there is a wide-area information processing system in which a plurality of computer system networks are organically connected via a gateway. In such a case, however, the above-mentioned multi-computer system on a dedicated software monitoring device is used. By centrally managing related information and intercepting abnormality detection information on each computer network, it is also possible to monitor software in a wide area information processing system. In the above embodiment, the attribute information of the functional module and the attribute information of the input / output message are limited. However, the present invention is not limited to this. For example, the module type, the importance, and the input / output message type, By including other information necessary to monitor the site with visual information, more visible visual information can be provided.

FIGS. 4 to 7 show the abnormality detection management information table 1 of each of the control computers 10 to 40 in this embodiment.
4 shows specific examples of table storage values in 4, 24, 34, and 44. The following description of the operation is based on this table value for ease of description. FIG. 4 shows the storage contents of the input / output definition table 141, the message output matrix 142, and the software abnormality monitoring timer table 145 for the control computer 10.

In FIG. 4, the input / output definition table 141
“A” indicates the number of registered function modules.
B to H indicate definition information on the function module 12. That is, B is the name of the functional module 12, C is the address information of the message matrix 142 to be used, and D is the flag information indicating whether or not an alternative output of the output message as a processing result of the functional module is possible. When this flag information is “0”, it cannot be replaced,
When it is "1" (illustrated example), it is set to be replaceable. E
Are the function code numbers input by the function module 12, F to
H indicates a function code number output by the function module 12. The message output matrix 142 corresponds to the function module, and the input / output definition table 141
Can be indexed from C. Here, in the message output matrix 142, function codes fc2, fc3, and fce output by the function module M1 are registered for each output pattern determined by processing conditions. The software abnormality monitoring timer table 145 corresponds to the function module,
The initial value is 0, which is used only for monitoring the output omission of the function code of each function module. During the monitoring, the name of the function module and the timer value being monitored exist.

FIG. 5 shows the contents of the input / output definition table 241, the message output matrix 242, and the software abnormality monitoring timer table 245 for the control computer 20. In the figure, A in the input / output definition table 241 indicates the number of registered function modules, and B to G indicate definition information on the function module 22. That is, B is the name of the function module 22,
C is address information of the message matrix 242 to be used, and D is flag information indicating whether or not an alternative output of an output message as a processing result of the function module is possible. As a specific example, the alternative is registered. E indicates a function code number input by the function module 22, and FG indicate function numbers output by the function module 22, respectively. E in this example indicates that there is no function code input by the function module 22. The activation of the APP when the input function code is not registered is operated by a different means that is not message driven. For example, depending on the type of the operating system, when the APP is activated, by adding an activation factor, the activated functional module can know the factor to be processed from the added activation factor. it can. The message output matrix 242 corresponds to a function module, and can be indexed from C in the input / output definition table 241. Here, this message output matrix 242 includes the function module M
2 are registered for each output pattern determined by the processing conditions. The software abnormality monitoring timer table 245 corresponds to the function module, and has an initial value of 0, and is used only for monitoring of a function code output leak of each function module. During monitoring, the function module name and the monitoring timer value are monitored. Exists.

FIG. 6 illustrates the contents of the input / output definition table 341, the message output matrix 342, and the software abnormality monitoring timer table 345 for the control computer 30. In the figure, A of the input / output definition table 341 indicates the number of registered function modules, and B to F indicate definition information on the function module 32. That is, B is the name of the function module 32,
C is address information of the message matrix 342 to be used. In this example, C is unused because there is no output message. D is flag information indicating whether or not an alternative output of an output message as a processing result of the function module is possible. In this example, the message is not output because no message is output. E indicates a function code number input by the function module 32, and G indicates that there is no function number output by the function module 32. The message output matrix 342 corresponds to the function module and can be indexed from C in the input / output definition table 341. Here, since there is no output message, this table is unused. Also,
The software abnormality monitoring timer table 345 is also unused.

FIG. 7 illustrates the contents of the input / output definition table 441, the message output matrix 442, and the software abnormality monitoring timer table 445 for the control computer 40. In the figure, A of the input / output definition table 441 indicates the number of registered function modules, BF indicates definition information about the function module 42, and GK indicates definition information about the function module 45. . That is, B is the name of the function module 42, and C is the address information of the message output matrix 442 to be used. Here, since there is no output message, it is unused. D is flag information indicating whether or not an output message as a processing result of the function module can be output as an alternative. In this case, since there is no message output, the replacement is not possible. E indicates a function code number input by the function module 42, and G indicates that there is no function number output by the function module 42. G is the name of the function module 45, and H is the address information of the message output matrix 442 to be used. In this specific example, since there is no output message, it is unused. I is flag information indicating whether or not an alternative output of an output message as a processing result of the function module is possible. In this specific example, the message is not output because no message is output. J indicates the function code number input by the function module 45, and K indicates that there is no function number output by the function module 45. The message output matrix 442 corresponds to the function module and can be indexed from C in the input / output definition table 441. Here, since there is no output message, this table is unused. The software abnormality monitoring timer table 445 is also unused.

FIG. 8 shows a relational diagram that means the storage values of the definition information of FIGS. 3 to 7. To explain this, when the function module 22 operates, a function code FC1 or FCe is output. There is a function module 12 as a function module for inputting the function code FC1. The function module 12 has a function code FC1
Is input, and the processing result is output to the function code FC2 and the function code F.
C3, output as function code FCe. Function code F
Function module 32 inputting C2 and function code FC
That is, "there is no output function code" in the processing result of the function module 42 to which 3 has been input. There is also a function module 45 that uses the function code FCe as an input function code. This is a method used when a common function code is input from each function module and processed, which normally corresponds to alarm processing and the like. Although the function module 45 is not directly related to the present invention, it is a factor to be considered when defining a function code in an output pattern for each processing condition of the message output matrix, such as the function modules 12 and 22.

Next, the operation of the main software function will be described with reference to FIGS. Prior to the description, first, a mechanism of detecting an abnormal part in the system according to the embodiment will be described with reference to FIGS. 9 and 10. FIG. 9 shows a software abnormality detection processing unit 13 using the control computer 10 as an example.
Of the monitoring mechanism of the output message monitoring unit 131 in FIG. In the figure, when the message input processing unit 121 for inputting the function code 9 linked from the operating system 11 operates,
First, the software abnormality monitoring information set processing 131A required for the timer monitoring processing unit 132 is activated, and the message output matrix extraction processing 131B required for continuously monitoring the output function code is activated. . Next, in the application data processing unit 122, the corresponding APP is activated to execute a unique process. After finishing the unique processing, the output function code extraction processing 131C is operated.
This output function code extraction processing 131C activates a function code matching processing 131D for each output pattern between a function code to be output, which is the result of the current application processing, and a message output matrix, which is the processing result of the message output matrix extraction processing 131B. Let it. The output pattern-specific function code collation processing 131D makes it possible to suppress output of an unauthorized function code. Next, when the output of the illegal function code is detected in the comparison / matching result determination 131E, the abnormality detection information 7 on the function module is activated by the software abnormality notification processing 131G.
The software abnormal part display processing 62 in the monitoring personal computer 60 is notified, and the abnormal part and the influence-related information are displayed on the monitor 63. Comparison / matching result judgment 1
When 31E is a normal output, the timer monitoring processing of the function module is canceled by the operation of the software abnormality monitoring information reset processing 131F, and the output function code 9 is linked to the operating system 11 by the message output processing unit 124.

The message input processing unit 121 and the message output processing unit include an operating system and an AP.
It is generally provided as an interface routine between Ps. Therefore, the software interface monitoring information set processing 131A and the message output matrix extraction processing 131B are provided in the input interface routine, and the output function code extraction processing 131 is provided in the output interface routine.
C, function code collation processing 131D for each output pattern, comparison collation result determination 131E, software abnormality monitoring information reset processing 131F, and software abnormality notification processing 13
It is preferred to incorporate 1G.

FIG. 10 illustrates the monitoring mechanism of the timer monitoring processing unit 132 in the software abnormality detection processing unit 13 related to FIG. 9 using the control computer 10 as an example. In the figure, a timer monitoring processing unit 132 is started at the time of system startup and is configured to periodically execute processing with a predetermined delay time. Here, when the monitoring timer value is registered in the software abnormality monitoring timer table 145 of FIG. 4 in the software abnormality monitoring information setting process 131A of FIG. 9, the timer monitoring processing unit 132 sets the monitoring timer value to the time. Increment is repeated at regular intervals until it comes up. Normally, the operation of the software abnormality monitoring information reset processing 131F cancels the timer monitoring processing of the function module, so that the monitoring timer value does not reach time-up.

However, if any abnormality occurs in the processing of the function module and the timer monitoring processing is not canceled, the monitoring timer value times out, and the software error notification processing 132F starts from the alternative message output availability determination 132D. . The basic purpose of the alternative message output availability determination 132D and the alternative message output process 132E is to execute a series of processes as described above because some abnormality occurs in the process of the function module and a predetermined function code is not output. This is to prevent interruption. Here, output message determination 13 of alternative message
2D assumes a case where a function module corresponding to an input function code can and cannot be output alternately. This depends on the target system field or the characteristics of the application. When a function module abnormality is detected due to the time-out of the monitoring timer value, the abnormality detection information 8 on the function module is transmitted to the software abnormality part display processing 62 in the monitoring personal computer 60 by the operation of the software abnormality notification processing 132F. Is displayed on the monitor 63.

FIGS. 11 to 17 supplementarily explain the internal logic of the output message monitoring unit 131. FIG. FIG.
1 is a supplementary explanation of the internal logic of the software abnormality monitoring information set processing 131A in the form of a flowchart. In the figure, the software abnormality monitoring information set processing 131A executes processing S1 for searching the input / output definition table 141 using the input function code linked from the message input processing unit 121, and determines whether there is an output function code of the execution result. Is determined in the processing S2. This is because there is no output of a function code depending on the characteristics of the function module, and there may be a process for only storing data in an internal memory or an external file.
If there is an output function code, the monitoring timer value for monitoring a predetermined time until the processing result of the function module is output as the function code and the name of the function module are replaced with the software corresponding to the function module. Processing S3 to register in abnormality monitoring timer table 145
Execute

FIG. 12 is a supplementary explanation of the internal logic of the message output matrix extraction processing 131B of FIG. 9 and the editing configuration of the message output matrix in the internal memory W1. In the figure, a message output matrix extraction process 131B uses the input / output definition table 141 searched in the software abnormality monitoring information set process 131A to determine whether there is a function code to be output.
This input / output definition table 1 only when
The message output matrix 14 indexed from 41
S2 for editing the file 2 according to the configuration information of the memory W1
Execute

FIG. 13 supplementarily explains the internal logic of the output function code extraction processing 131C of FIG. 9 and the editing configuration of the function code output sequence in the internal memory W2.
In the figure, an output function code extraction process 131C uses the function code sequence 9 created in the application data process 122 to edit according to the information in the memory W2 S2.
Execute 1.

FIG. 14 supplementarily describes the internal logic of the output pattern-specific function code collation processing 131D in the form of a flowchart. In the figure, a function code matching process 131D for each output pattern uses a memory W1 created in a message output matrix extraction process 131B and a memory W2 created in an output function code extraction process 131C, and stores functions stored in the memory W1. Code pattern n
, The processes of S31 to S36 for collating the output function codes for each pattern in the memory W2 corresponding to the number of output function codes. The collation result is used to determine whether or not an illegal function code is output.

FIG. 15 is a supplementary explanation of the internal logic of the software abnormality monitoring information reset processing 131F in the form of a flowchart. In the figure, software abnormality monitoring information reset processing 131F executes processing S41 to S43 for deleting software abnormality monitoring information corresponding to the function module registered in the software abnormality monitoring timer table 145.

FIG. 16 shows the comparison and collation result judgment 131 of FIG.
This is a supplementary explanation of the editing / outputting processing S51 of the notification data related to the software abnormality notification processing 132F related to the function module in FIG. 10 when the unauthorized function code is detected in E. FIG. 17 shows a message configuration 7 of the software abnormality notification data.
That is, the message configuration 7 includes the function code 71, the source node number 72, the illegal function code output abnormality report data 7
3, the abnormal function code output abnormality report data 73
Are input function code 731, function module name 732,
An unauthorized function code string 733 is included.

FIGS. 18 to 21 supplementarily explain the internal logic of the timer monitoring unit 132 shown in FIG. FIG. 18 is a supplementary explanation of the internal logic of the alternative message output availability determination process 132D in the timer monitoring unit 132 in a flowchart format. In the figure, an alternative message output availability determination process 132D targets the function module information for which the monitoring timer value has expired in the software abnormality monitoring timer table 145, and sets an output message alternative availability flag related to the functional module in the input / output definition table. It is determined and executed by extracting and referring to the information 141. If the output message substitution possibility flag is "1", the output message substitution is possible, and if it is "0", processing S61 to S62 in which it is determined that the output message substitution is impossible is executed. FIG. 19 shows the editing / output processing S in the alternative message output processing 132E for the functional module when the alternative message output permission determination 132D determines that the alternative message output is possible.
FIG. 20 shows a supplementary explanation of the structure of the substitute message 9. As shown, the alternative message 9
Is composed of a function code 91, a source node number 92, and APP-specific data 93. Further, the APP unique data 93 includes a flag 931 and alternative APP unique data 932.
Contains. The flag is normal when “0”,
"1" indicates an alternative output message.

FIG. 21 shows the process of editing and outputting data in the software error notification 132F for the function module when the timer monitoring section 132 detects a function module error due to the expiration of the monitoring timer.
1 is shown. FIG. 22 shows a message configuration 8 of the software abnormality notification data. As shown, the message structure 8 includes a function code 81, a source node number 82, and software abnormality notification data 83. Also, the software abnormality report data 83
Is an alternative output execution flag 831, an input code 832,
A function module name 833 is included.

FIG. 23 is a supplementary explanation of the internal logic of the software abnormal part display processing 61 in the form of a flowchart. That is, the software error notification data 7 notified by the operation of the software error notification process 131G in FIG.
Software error notification data 8 notified by the operation of
Is input, and the abnormal part of the software and the influence relation are displayed on the monitor CRT 63 of the monitoring personal computer 60 in a data flow diagram format. As shown in the drawing, the software abnormal part display processing 61 is performed in step S91.
To S94. The process S91 executes a process of inputting the software error notification data 7 or the software error notification data 8 from the operating system 62. The process S92 executes a process of editing the input function code and the function module information in the software error notification data 7 or the software error notification data 8 input in the process S91 into the internal memory as display information. Processing S93
2 obtains function module connection information using a function code output from the function module as an input function code by a known “software / module relation recognition method”, and FIG.
A process for editing and creating as displayable information in the data flow diagram format shown in FIG. 4 in the internal memory is executed. Here, FIG.
In the information that can be displayed in the form of a data flow diagram displayed in the, the pictorial information that can identify the function module in which the abnormality has occurred and the function module in the affected area is displayed, and whether the function module is abnormal due to the output of an invalid function code, Or,
It is configured to be displayed as picture information that can identify whether a function module processing abnormality is detected by timer monitoring. Processing S9
In step 4, the display information in the internal memory created in step S93 is transmitted to the CRT 63, and as shown in FIG. 24, the influence relation caused by the abnormality occurrence function module is displayed as picture information.

The abnormality monitoring system implemented by the above-described processing will be described again by taking the contents displayed on the CRT monitor 63 as an example. FIG. 24 shows that, when a function module abnormality is detected in the process control computer system that performs information management of the rolling plant equipment, the abnormal part and the affected area of the APP in the computer system are displayed together with the input / output message. I have. The function module 631 stores the input function code FC1 that has caused the abnormality detection,
This is displayed together with the picture information indicating the input function code that caused the abnormality detection. The display line 632 is displayed as a line indicating the connection with the input function code FC1 input by the function module 633 which is the abnormal part, as line display information different from the connection line indicating the other influence-related range. After inputting the function code FC1, the function module 633 displays the abnormal part function module M1 in which the processing error has occurred, together with the picture information indicating the function module. The function module 635 displays the function code FC2 whose output has been disabled due to the abnormality of the function module 633, together with the picture information indicating the function code related to the influence. Display line 63
Reference numeral 4 denotes a line indicating the connection with the function module 635 which is the output destination of the function module 633 which is the abnormal part, and is displayed as line display information different from the connection line indicating the cause of the abnormality. The display line 636 indicates the function module 635
Is displayed as line display information different from the connection line indicating the cause of occurrence of the abnormality as a line indicating that is input to the downstream APP. In the function module 637, a function module M3, which is an influence range in which the function code FC2 output from the function module 635 is used as an input function code, is displayed together with picture information indicating the function module. Further, in the function module 639, the function code FC3 whose output is disabled due to the abnormality of the function module 633 is displayed together with the picture information indicating the function code related to the influence. Display line 63
Reference numeral 8 denotes a line indicating the connection with the function code FC3 of the function module 639 which is the output destination of the function module 633 which is the abnormal part, and is displayed as line display information different from the connection line indicating the cause of the abnormality. The display line 640 is displayed as a line indicating that the function code FC3 has been input to the downstream APP as line display information different from the connection line indicating the cause of the abnormality. The function module 641 includes a function code FC3 output by the function module 639.
Is displayed, together with the picture information indicating the function module.

The display color and line thickness of each piece of picture information on the monitor screen are changed according to the abnormality occurrence status of the abnormal part. To be considered. In addition, a process control computer that manages information on rolling plant equipment by incorporating a display processing method so that the pictorial information display color and line thickness of a function code that has not been output or a function code that has been output instead can be understood. You can easily monitor the abnormal location and the affected area of the system software in real time. Further, the above-described software abnormal part display information creation processing 61
In the past, abnormal status of the software of the computer system running in real time by adding the monitor information storage function and the retroactive playback display function using the stored information by the storage function Can be easily confirmed. Further, since the abnormal parts and the influence relation of a series of programs in the computer system can be monitored in real time in the form of a data flow diagram, software troubleshooting of the computer system can be easily performed, and the reliability of the system is improved.

[0053]

As described above, according to the present invention,
The provision of the comparison / comparison means and the provision of the abnormal part detection means make it possible to detect an abnormal part of software in a distributed processing computer system operating in real time.

Further, since a message relating to the detected abnormality is not output based on the detected abnormality, the system can be protected and the abnormality can be localized.

Further, since the alternative message output means is provided, the localization, autonomy and coordination of the abnormality are made possible by complementing the detected abnormality.

Further, since a display control means for displaying the detected abnormal part and the affected area on the display device is provided, in the computer system operating in real time, the abnormal part and the affected area of the software are displayed as visual information. It can be easily confirmed.

[Brief description of the drawings]

FIG. 1 is an overall system configuration diagram of an embodiment of the present invention.

FIG. 2 is a diagram showing a general configuration of a message with a content code.

FIG. 3 is a diagram showing a general configuration of abnormality detection management information.

FIG. 4 is a diagram showing a specific example of abnormality detection management information for the control computer 10.

FIG. 5 is a diagram showing a specific example of abnormality detection management information for the control computer 20.

FIG. 6 is a diagram showing a specific example of abnormality detection management information for the control computer 30.

FIG. 7 is a diagram showing a specific example of abnormality detection management information for the control computer 40.

FIG. 8 is an example diagram for explaining a relationship between a function module and a function code;

FIG. 9 is a diagram illustrating a processing procedure of an output message monitoring unit.

FIG. 10 is a diagram illustrating a processing procedure of a timer monitoring unit.

11 is a diagram illustrating a detailed procedure of a shift wear abnormality monitoring information setting process 132A of FIG. 9;

12 is a message output matrix extraction process 1 of FIG.
It is a figure explaining the detailed procedure of 31B.

FIG. 13 is a diagram illustrating a detailed procedure of an output function code extraction process 131C in FIG. 9;

14 is a function code matching process 1 for each output pattern in FIG.
It is a figure explaining the detailed procedure of 31D.

FIG. 15 is a diagram illustrating a detailed procedure of software abnormality monitoring information reset processing 131F of FIG. 9;

FIG. 16 is a diagram illustrating a detailed procedure of a software abnormality notification 132F in FIG. 10;

FIG. 17 is a diagram illustrating a message configuration example of software abnormality notification data.

FIG. 18 is an alternative message output availability determination 132 of FIG. 10;
It is a figure explaining the detailed procedure of D.

FIG. 19 is a diagram illustrating a detailed procedure of an alternative message output 132E of FIG. 10;

FIG. 20 is a diagram illustrating a configuration example of an alternative message.

21 is a diagram illustrating a detailed procedure of a software abnormality notification 132F in FIG.

FIG. 22 is a diagram illustrating an example of a message configuration of software abnormality notification data.

23 is a software abnormal part display processing unit 62 of FIG. 1;
It is a figure showing the processing procedure of.

FIG. 24 is a diagram showing a specific example of software abnormal part display.

[Explanation of symbols]

1-5 Network control device 6 Network 7,8 Abnormality detection information 9 Message with content code 10,20,0,40 Process control computer 60 Monitoring personal computer 63 Monitoring CRT 64 Input means 11,21,31,41, 61 Operating system 12, 22, 32, 42, 45 Application program (APP) 13, 23, 33, 43 Software abnormality detection processing unit 14, 24, 34, 44 Error detection management information table 16 Software abnormal part display processing unit 121 Message input processing unit 122 Application processing unit 124 Message output processing unit 131 Message output monitoring unit 132 Timer monitoring processing unit

Claims (10)

[Claims] 1. A related output information table storing related output information defining output messages corresponding to processing conditions of an input message for each program that executes a series of processes of an information processing system by a plurality of programs; Selecting means for selecting related output information related to one program from the related output information table; comparing the output message of the execution result of the one program with the related output information selected by the selecting means to output the output message; A distributed processing computer system comprising a software abnormality processing device having a comparison / matching means for detecting the presence / absence of an abnormality. 2. The software abnormality processing device according to claim 1, wherein the software abnormality processing device stores input / output definition information defining attribute information regarding a program and a connection relationship between the program, the input message, and the output message. An output definition information table, comprising an abnormal part detecting means for specifying an abnormal part and an affected area related to the output message abnormality detected by the comparison / matching means with reference to the input / output definition information table. Distributed processing computer system. 3. The software abnormality detection device according to claim 1, wherein the software abnormality detection device refers to the related output information, and if an output message output from one program is not defined in the related output information, the output is performed. A distributed processing computer system comprising an abnormality processing means for preventing output of a message as an invalid message. 4. The software abnormality processing device according to claim 1, wherein the software abnormality processing device is configured such that the comparison / collation unit does not output an output message defined in the output related information and which should be output. And a substitute message output means for outputting a substitute message in place of the output message relating to the abnormality when detecting the abnormal condition. 5. A plurality of computers are connected via a signal transmission path, a series of processing is divided into a plurality of application programs and distributed and executed by the plurality of computers, and each of the application programs is transmitted from the signal transmission path. In a distributed processing computer system that is started based on a predetermined input message to be input and outputs an output message including a processing result and predetermined information to the signal transmission path, A software error processing device for performing detection and processing is provided, and the software error processing device includes, for each application program, related output information in which related output information defining an output message corresponding to a processing condition of an input message is stored. Table and one of the related output information tables. Selection means for selecting related output information relating to the application program, and comparing the output message of the execution result of the one application program with the related output information selected by the selection means to detect the presence or absence of an abnormality in the output message And a comparing and collating means. 6. The software abnormality processing device according to claim 5, wherein the software abnormality processing device stores input / output definition information defining attribute information regarding a program and a connection relationship between the program, the input message, and the output message. An output definition information table, comprising an abnormal part detecting means for specifying an abnormal part and an affected area related to the output message abnormality detected by the comparison / matching means with reference to the input / output definition information table. Distributed processing computer system. 7. A display control means according to claim 6, further comprising a display device provided on said network, for displaying a connection relationship of another application program operated by inputting an output message of one application program by picture information. A distributed processing computer system, wherein the display control means displays the abnormal part and the affected area as visual display information. 8. The method according to claim 5, wherein
The software abnormality detection device refers to the related output information, and, when an output message output from one program is not defined in the related output information, an abnormality that prevents the output message from being output as an invalid message. A distributed processing computer system having processing means. 9. The software error processing device according to claim 8, wherein the comparison and collation unit detects an abnormality of the program due to the fact that there is no output message defined in the output related information that should be output. And a substitute message output means for outputting a substitute message in place of the output message relating to the abnormality. 10. The distributed processing computer system according to claim 9, wherein said display control means displays an output of said alternative message as said visual display information.