JPH11120105A - Method for collecting diagnostic information of distributed processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily collect information that is needed to diagnose a distributed processing system even when a person who diagnoses does not know the structure of an object system very well in order to know the state of the system when diagnosing abnormality that has occurred in the system. SOLUTION: A design information extracting processing 241 extracts design information that is used to operate system and stores it in a data catalog definition table 231 and an inter-data relation table 232 as the relation between data in the system and a system component and the relation between data. A related data collection processing 221 refers to the tables 231 and 232 and retrieves data that are designated by user interface processing 223, data that are related to a system component or data which are related to an event that occurs in the system. Related data are presented to the processing 223 and also are collected by using data input-output management processing 211, and collection result data are notified to the processing 223.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for collecting diagnostic information of a distributed processing system using a plurality of programs, and more particularly to a method for collecting diagnostic information using design information existing in the system and required for operation of the system. The present invention relates to a method of diagnosing a distributed processing system which can be suitably adapted to a case where important information is collected and an abnormality is detected in an operation state of a system or an output result.

[0002]

2. Description of the Related Art A method of diagnosing a distributed processing system is described in, for example, Network Computing (August 1993).
Monthly) As described in PP63-79 "Possibilities and Limitations of SNMP", there are some that use the network monitoring protocols SNMP and CMIP. According to this method, it is possible for the monitoring processing device to monitor and diagnose a predefined state of each device with respect to a computer, a printer, a router, and the like connected to the network.

In order to diagnose a business, for example, Japanese Patent Laid-Open No. 9-2236 discloses a method for diagnosing an operator and instructing a corresponding operation.
There is a diagnostic expert system based on a knowledge base such as that described in Section 0, which holds knowledge of abnormal conditions and coping methods so that an operator can cope with abnormalities during system operation, obtains the status of the target system, and An abnormality was diagnosed. This diagnosis expert system diagnoses the state of a plant controlled by a processing device, and stores knowledge about a target plant, assumed abnormalities, and knowledge on how to deal with the abnormalities in a knowledge base. As information required for diagnosis, information specified in the knowledge base is obtained from the target plant.

As a method of collecting information for diagnosis, there has been a method of acquiring specified information with a specified accuracy as disclosed in Japanese Patent Application No. 9-75529, for example.

[0005]

The abnormal state of the system is not limited to an abnormal operating state of the computer or the equipment, but is also various, such as an error caused by outputting erroneous data or an irregular control timing of the equipment. In particular, abnormalities that are not initially anticipated often occur due to inputs from facilities and operators outside the computer system, and expansion or modification of the computer system.

In this case, diagnosis cannot be performed only by processing incorporated in advance for abnormality diagnosis and abnormality notification. Therefore, the situation of the entire system is recognized to identify the cause of the abnormality, and a situation diagnosis that assumes an abnormal location is performed. There is a need to do. Usually, when such an abnormality diagnosis is performed, the diagnosis is performed by referring to data in the system in order to know the state of the system. In particular, when a diagnosis is made from a remote place, data is important because the physical situation is not known.

When diagnosing an unexpected abnormality as described above, it is rare to specify which data should be referred to. The system status is determined by referring to a plurality of pieces of information related to the abnormality, and the normal state is determined. It is necessary to localize the problem by isolating the abnormal part from the abnormal part. For example, when an abnormality occurs in a computer, the status of the computer is determined based on data held by the computer, and whether an abnormality has occurred in the computer due to input / output data of the computer or an abnormality caused externally. Judge. In addition, in the case of an abnormality of the equipment, referring to the data related to the equipment, which part or device has the abnormality is cut out. However, in the related art, as for the data to be obtained, the requester has to specify necessary data in advance, or it is necessary to specify all data to be obtained when necessary, which is very troublesome.

Further, since which data is referred to depends on the knowledge and know-how of the person making the diagnosis, it is necessary to refer to the data many times.

Further, in a system developed in the past or a system developed by another person, since a diagnostician does not memorize detailed implementation methods, information necessary for designating information to be collected is contained in any file or message. It is necessary to start by examining how to implement. For this reason, it is necessary to search a design document or to ask a designer of a system to be diagnosed, which is very troublesome.

[0010] The present invention solves the above-mentioned problems, and enables not only collecting directly specified data but also collecting related information necessary for diagnosis based on the specifications of an installed system. .

[0011]

SUMMARY OF THE INVENTION A method for collecting diagnostic information of a distributed processing system according to the present invention comprises: a means for extracting design information from an implementation of the distributed processing system; a system component described in the design information; Means for retrieving information related to the specified component or information using the relationship between the information, so that the information for diagnosis is not individually specified, but related to the specified information. It is also possible to collect and provide information. Here, the design information refers to information essential for the operation of the system.

Further, in the present invention, before collecting other information related to the designated information, the related information is presented to a designated person so that necessary information can be selected so that the collected information can be selected. Facilitates the specification of

Further, in the present invention, design information for detecting related data is obtained from actual operation information such as control information of data used in the system and a program configuration in a processing device, and data is collected. , Making it easy to grasp the system status.

[0014]

Embodiments of the present invention will be described below in detail. In particular, an example using the following relationship will be described for the components and data of the system.

(1) Data generation relation using input / output relation by application program (2) Relation between data by relation database (3) Data extracted from actual operation information managed by operating system and the like and application program and Note that the application program refers to a program that performs business processing and is to be diagnosed.

(Embodiment 1) In a first embodiment of the present invention, among the relationships between data in the system, the relationship between the processing device and the application program for performing business processing and the data, and the data An example using the generation relation will be described.

FIG. 1 shows a configuration example of a system to which the present invention is applied. It comprises processing devices 111 to 113 for exchanging data with each other via the transmission medium 101 and external input / output devices 121 to 123 having an input / output interface with the outside. Terminals 121 to 123 are connected to the processing devices 111 to 113. Here, the external input / output device is used to control a program executed on the processing device, refer to the output of the program via a man-machine interface such as a display, a keyboard, a touch panel, or the like, and to communicate with a controlled machine. Has a function of inputting / outputting a control signal. However, this is not essential, and some processing devices do not have an external input / output device.

The processing device 111 exchanges data with the processing device 114 via the transmission medium 102. In this way, each processing device can connect a plurality of transmission media, and the processing device 111
The processing device 114 and the processing devices 112 and 113 can also transmit and receive data by mediating.

FIG. 2 is a diagram showing a software configuration of a diagnostic information collection process in any of the processing devices 111 to 114 in the first embodiment of the diagnostic method of the present invention. The data input / output management processing 211 is a program that manages access to data stored in its own processing device or another processing device, and data transmitted / received via a transmission medium. Using the data input / output management processing 211, data stored in the own processing device and other processing devices in the system,
And message data transmitted and received between the processing devices. As a means for acquiring, for example, a method of transferring data in which data input / output management processing of each processing device is designated as disclosed in Japanese Patent Application No. 9-75529 may be used. Alternatively, a broadcast method as shown in FIG. The data definition table 230 is a data catalog definition table 231
And a data relation table 232. The data catalog definition table 231 is a table that stores a list of data in the system. Data relation table 23
2 stores interrelated information about the data in the system. The design information extraction process 241 is a process of extracting a data list in the system and a mutual relationship regarding data from information input in system development. The extracted mutual relationship is stored in the data catalog definition table 231 and the data relation table 232. Alternatively, it may be obtained from another processing device.

The related data collection process 221 searches for data related to the data specified via the user interface process 223 by referring to the data relationship table 232,
Data is collected through the data input / output management processing 211.
The collected data is transmitted to another processing device, stored in its own processing device, or stored in the user interface 223.
Pass to The user interface process 223 is a process of passing the collected data specified by the user to the related data collecting process 221 and presenting the collected data received from the related data collecting process 221 to the user.

FIG. 3 is a diagram showing the data structure of the data definition table 230. FIG. 3 (i) is a configuration example of the data catalog definition table 231. Each record of the data catalog definition table 231 includes an index 311, a data name 312, a type 313, and a storage processing device 314. The index 311 is a unique number for identifying each record. The data name 312 is an item representing the name of each data. The type 313 indicates a type difference such as whether the data indicated by the record is a message or data stored in a memory or a disk. Where data name 312
The description of the type 313 depends on the data access method of the data input / output management processing 211, and a unique name may be assigned to both items. The storage processing device 314 is an item indicating the processing device in which the data indicated by the record exists, and indicates the stored processing device when the type 313 is data stored in a memory or a disk.

FIG. 3 (ii) shows the relation table 232 between data.
This is an example of the configuration. The data relationship table 232 indicates the relationship between the application program and the input / output data.
22, an input data index 323 and an output data index 324. The processing device 321 indicates a processing device in which the application program 322 of the record is stored. The application program 322 indicates an application program that performs business processing. The input data index 323 is an index 311 of a record described in the data catalog definition table 231 of data input by the application program of the record.
Is stored. The output data index 324 stores the index 311 of the record described in the data catalog definition table 231 of the data output by the application program of the record.

The input / output relationship via the data list and the application program described in FIG.
It can be extracted by various methods. For example, application program data input / output is
In the case of management as shown in No. 529, it can be created by collecting tables used by the management program. Even when such a management program does not exist, the relationship between input and output data can be extracted by analyzing the source code created when developing the application program using, for example, a compiler.

FIG. 4 is a diagram showing a processing flow of the related data collection processing 221. The related data collection process 221 includes a process of searching for collected data (Step 401), a process of acquiring collected data (Step 416), and a process of storing or notifying collected data (Step 417). The collected data search processing 401 is a processing device, application program,
Alternatively, data is obtained, and if data is specified, an index is obtained by referring to the data catalog definition table 231 (step 411), and it is determined whether the key is a processing device or an application program (step 412).

In this case, the input / output data of the designated processing device or application program is searched with reference to the data relationship table 232 (step 413). If data is specified as a key, a data-related search process (step 402) is executed. In step 402, a record in which the data is specified as output data is searched for by referring to the data relation table (step 414). Next, input data specified in the searched record is detected (step 415). In this example, the search is limited to the search of the input data. However, the search of the output data can be performed.

The data related to the designated key searched in step 401 is collected via the data input / output management processing 211 (step 416). The collected data is notified to a user interface or another processing device specified by the user, or stored in its own processing device (step 41).
7).

FIG. 5 shows an example of collecting diagnostic information to which the first embodiment of the present invention is applied. FIG. 5 (i) is an example of an application program. Application programs 521 to 523 are stored in the processing devices 511 to 513, respectively. The application program 522 detects the entry of the workpiece 551 by the sensor 541, and notifies the application program 523 by a message 532. The application program 523 receives the processing conditions from the application program 521 in advance (531), processes the processed product, and writes the result to the file 534.

FIG. 5 (ii) shows a setting example of the data catalog definition table 231. The data in this system include the message “processing condition” (531) indicated by index 1,
Message “Indexed product entry” indicated by index 2 (53
2) and the file “processing result” (533) in the processing device “Node3” indicated by the index 3 is shown. FIG. 14 (iii) is an example of setting of the inter-data association table 232. The application program “AP1” of the processing device “Node1” outputs the data indicated by index 1 in the data catalog definition table 231 (record 561), and the application program “AP2” of the processing device “Node2” similarly outputs index 2 (Record 562), the application program "AP3" of the processing device "Node3" inputs the data of the indexes 1 and 2, and outputs the data of the index 3 (record 563). It is shown.

By using such design information, it is possible to efficiently collect system diagnostic information.

For example, when the processing of the processed product is not performed well, there are several causes of the abnormality such as when the recognition of the processed product is abnormal (AP2 or the sensor 541) or when the processing is abnormal (AP3). It is difficult to specify as it is. In such a case, the AP3 (523) to be processed is designated as a key, and the related information 531, 532, and 533 are collected, and whether the processing condition is abnormal, the workpiece entry message is abnormal, or these messages themselves are used. It is possible to diagnose whether or not the message is transmitted properly and narrow down the cause of the abnormality. Alternatively, when the data of the processing result 533 is abnormal, the message 531 is used as data related to the processing result 533.
By searching and collecting 532 and 532, the diagnostician can grasp the situation without knowing the application structure.

As shown in the above example, by using the design information on the relationship between the processing device or the application program and the data and the generation relationship between the data using the same,
Relevant information for system diagnosis can be searched and collected without the user having to specify all based on the implementation of the system.

(Embodiment 2) In the second embodiment of the present invention, among the relations between data in the system, collection of related information when the relation between data is specified by a relation database as a data management platform. An example will be described.

FIG. 6 shows an example of the structure of the data relation table 232 in this embodiment. The data relationship table 232 indicates the relationship between table data, and each record has a related source data index 611, a related data index 612,
It is composed of related data items 613. Here, the table data includes tables constituting a relational database and a virtual table called a view or a query generated by joining the tables. The related source data index 611 and the related destination data index 612 store the index 311 of the corresponding record described in the data catalog definition table 231 of the data in which the respective data items are related to each other. The related data item 613 stores the relationship between data items that cause association among the data items constituting the data indicated by 611 and 612. The relation between the data described in FIG. 6 can be easily obtained by searching the construction information of the relation database.

FIG. 7 shows an inter-data relation search processing (step 402) in the relation data collection processing 221 in this embodiment.
It is a figure showing the flow of. In step 402 of the present embodiment, the record in which the designated data is designated as the related source data index is searched by referring to the data relationship table 232 (step 711). Next, related destination data specified in the searched record is detected (step 712). Similarly, the record in which the specified data is specified as the related data index is searched (step 713), and the related source data specified in the searched record is detected (step 714). The process of collecting, notifying, or storing data using the data detected in the above steps is the same as the process shown in FIG. 4 in the first embodiment.

FIG. 8 shows an example of collecting diagnostic information to which the second embodiment of the present invention is applied. FIG. 8I shows an example of data. The manufacturing condition table 801 is a table storing manufacturing conditions for each product. Each record has a product ID 811, a material type 1 (812), a material type 2 (813), and a chemical type 1 (814).
Etc. The product definition table 802 is a table for managing product types. Each record has a product ID 821 and a name 8
It consists of 22, etc. The equipment definition table 803 is a table for managing equipment types, and each record is composed of an equipment ID 831, a name 832, and the like. The production results table 804 is a table for managing the results of each product, and each record is Lot N
o.841, serial number 842, product ID 843, equipment ID 844, processing result 845
Etc. Here, in order to avoid duplication, the product ID 811 of the manufacturing condition table 801 and the manufacturing result table 804
The product ID 843 is defined in the relational database so as to refer to the product ID in the product definition table 802, and the equipment ID 844 in the production results table 804 is defined as referring to the equipment ID 831 in the equipment definition table 803.

FIG. 15 is a setting example of the data definition table 230. FIG. 15 (i) is a setting example of the data catalog definition table. The data in this system include the table “product definition” (851) in the DB “DB1” in the processing device “Node1” indicated by index 1 and the DB “DB1” in the processing device “Node1” indicated by index 2 Table "Production conditions"
(852) in the processing device “Node1” indicated by index 3
The table “equipment definition” (853) in the DB “DB1” and the table “manufacturing results” (854) in the DB “DB1” in the processing device “Node1” indicated by index 4 are shown. FIG.
4 (iii) is a setting example of the inter-data relation table 232. The product definition table indicated by index 1 and the manufacturing condition table indicated by index 2 are related to each other via a data item called product ID (record 86).
1) Similarly, the product definition table and the production results table are related via the data item “product ID” (record 86
2), it is shown that the equipment definition table and the production result table are related via the data item “equipment ID” (record 863).

By using such design information, it is possible to provide data related to data specified by the user for diagnosis. For example, if you want to investigate product defects, specify the product ID as collected data,
Manufacturing results and manufacturing conditions are collected and provided as related to the data. By using this, it is easy to diagnose a defect-producing product, such as the equipment used to produce the product and the raw materials used based on the production conditions, and to investigate the cause of the defect. In this way, even if you are unfamiliar with how to implement the system, or if you have forgotten the detailed implementation, you can easily recognize the system status by providing the relevant information by giving the opportunity to refer to the information you want to see. can do.

As an application of the first and second embodiments described above, for example, as described in Japanese Patent Application No. 8-42464, the relationship between message data and table data is managed as information necessary for system operation. If it is, it can be added to the data relation table and made a search target.

Further, in the above embodiment, an example of one-layer related retrieval based on the inter-data relation table has been described, but multi-layer related data retrieval is performed by repeatedly using the related data collection processing in this embodiment. It is also possible.

(Embodiment 3) In the first and second embodiments, a method of searching for relevant data, collecting the relevant data and notifying the user is described. Can also be presented.

FIG. 9 is a diagram showing the flow of the related data collection processing in this embodiment. User interface processing
When the data, the processing device, or the application program is designated from 223, the related data is searched in the same manner as in Step 401 in the first and second embodiments (Step 401). The detected relevant data is notified to the user interface processing 223 or another processing device (step 911).

FIG. 10 is a diagram showing an example of a user interface in this embodiment. In the present embodiment, a message “manufacturing result” (1011) is specified as key data on the user interface screen (1001). The specified data is searched for related data, a processing device or an application program, and their input / output data by a related data search process (step 401). In the present embodiment, the painting process (1031) is detected as the side that generates the data, that is, the process or processing device (1021) that outputs the data. Also painting process
As the data to be entered in 1031, the file "painting conditions"
(1032), and the message “entering the processed product” (1033).

Similarly, a process "result management" (1041) is detected as a process or a processing device (1022) using the message "manufacturing results" on the application side, and an output file "painting results" of the process "results management" is detected. (1042) has been detected. In this embodiment, the related data is also searched for the file “painting results”, and the generation side (1051),
As the directly related data as described in the application side (1052) and the second embodiment, the file “workpiece definition” (1053) and the file “equipment fixed period” (1054) are detected.

It is possible for the user to select data to be collected from the relevant data detected in this way, and then to collect the data to diagnose the state of the system.

(Embodiment 4) In the first, second, and third embodiments, a method for presenting or collecting data related to designated data using a relation between data extracted from design information is described. explained. In this embodiment,
A method for selecting related data using design information included in the data will be described.

FIG. 11 is a diagram showing a software configuration in this embodiment. Processing device as in the first embodiment
It is a figure showing the software composition of diagnostic information collection processing in arbitrary processing units among 111-114. The data input / output management processing 1111 is a program that manages access to data stored in its own processing device or another processing device by an application program, and data transmitted / received via a transmission medium. Using the data input / output management processing 1111, message data transmitted / received between the processing devices via the common transmission medium, and data input / output by the application program in its own processing device are acquired, and the output source device or , It is possible to check the application program. Means for obtaining is, for example, Japanese Patent Application No. 9-75529.
The data input / output management processing of each processing device may use a method of transferring designated data as described in the issue number, or the Nikkei Computer No. 1995.9 “Message Cooperation System” pp
A message sending and receiving platform such as those at 126-139 may be used. User interface processing 1122
The user is made to designate a processing device or an application program as a key, and the selected data is received and displayed.

The related data selection processing 1121 obtains data transmitted from the processing device specified by the user interface processing 1122 or data output from the specified application program.

FIG. 12 is a diagram showing the flow of the related data selection process 1121. The data input / output management processing 1111 receives the data received from the common transmission medium or the data output from the application program in its own processing device (step 1211), and the data is output from the designated processing device or application program. It is determined whether the data is data (step 1212). If the condition is satisfied, the data is output to the user interface or another processing device (step 1213). Data input / output management processing 1111
If the transfer destination is specified in the data received from the server, it is determined whether the data is output data to the specified processing device or application program (step 1214). If the condition is satisfied, the user interface processing or other processing is performed. Output to the device (step 1215).

As described above, the operation status of the system can be diagnosed by using the design information added to the actual data.

(Embodiment 5) As an application of the above-described fourth embodiment, it is possible to support an abnormality diagnosis when the configuration of an operating application program in a processing apparatus changes dynamically. When the application program configuration is not determined in advance, such as when the processing device starts the application program according to the situation or downloads and uses the application program via the network, the operation of the processing device according to the present embodiment is applied. Analysis becomes easy.

FIG. 13 is a diagram showing a software configuration in the fifth embodiment. Operation processing configuration change detection processing 13
11 is a process for detecting a change in the configuration of an application program running on the processing device. This can be implemented, for example, by monitoring processes and tasks managed by the operating system and detecting newly activated application programs. Alternatively, when an application program is downloaded via a network, the application program can be detected using a communication program used for downloading. In the case of a program on the WWW, an application program that newly operates using browser software or the like as an execution environment can be detected.

The detected application program is notified to the related data selection processing 1121, and the input / output data is monitored in the related data selection processing 1121 as described in the fourth embodiment.

As described above, the actual application program configuration in the processing device is used as the design information, and it is easy to diagnose the operation status of the newly activated application program when the operation failure or the malicious operation occurs. Become.

[0054]

According to the present invention, there is provided means for extracting and storing design information used for operating a system, and data designated using the design information and data related to the data, or a designated system. By presenting and collecting data related to the components, it is possible to facilitate the recognition of the system status when diagnosing the system. For this reason, even if you have little knowledge about the implementation of processing devices and application programs, or if you forget to implement,
It is possible to easily search and collect data necessary for diagnosis and recognize the situation of the system.

[Brief description of the drawings]

FIG. 1 is a hardware configuration diagram of a distributed processing system according to an embodiment.

FIG. 2 is a software configuration diagram of a diagnosis process in a first example of the embodiment.

FIG. 3 is a diagram illustrating a data configuration of a data definition table 230 according to the embodiment.

FIG. 4 is a diagram illustrating a flow of a related data collection process 221 according to the embodiment.

FIG. 5 is an example of collecting diagnostic information to which the first example of the embodiment is applied.

FIG. 6 is a configuration example of an inter-data relation table 232 in the second example of the embodiment.

FIG. 7 is a diagram showing a flow of an inter-data relation search process 402 in the second example of the embodiment.

FIG. 8 is an example of collecting diagnostic information to which the second example of the embodiment is applied.

FIG. 9 is a diagram showing a flow of a related data collection process in a third example of the embodiment.

FIG. 10 is a diagram illustrating an example of a user interface according to a third example of the embodiment.

FIG. 11 is a diagram illustrating a software configuration according to a fourth example of the embodiment.

FIG. 12 is a diagram showing a flow of a related data selection process 1121 in the third example of the embodiment.

FIG. 13 is a diagram illustrating a flow of a related data collection process in a third example of the embodiment.

FIG. 14 is a continuation of an example of collecting diagnostic information to which the first example of the embodiment is applied.

FIG. 15 is a continuation of an example of collecting diagnostic information to which the second example of the embodiment is applied.

[Explanation of symbols]

111 to 114: processing device, 211: data input / output management process, 22
1: related data collection processing, 223: user interface processing, 231: data catalog definition table, 232: data relation table, 241: design information extraction processing

Claims (5)

[Claims] 1. A distributed processing system for performing a series of processes in a distributed manner by a plurality of processing devices connected to a common transmission medium, wherein a program for executing each of the series of processes is stored in one of the plurality of processing devices. Alternatively, in a distributed processing system in which processing is performed by storing data in a plurality of processing devices, a relationship between data used in the system or a relationship between data and a system component is extracted from design information required for system operation. Using the design information, searching for specified data and data related to the data, or data related to a specified processing device or program, and collecting and providing the searched data Diagnostic information collecting method for a distributed processing system. 2. A method for collecting diagnostic information in a distributed processing system according to claim 1, wherein the data retrieved is presented to a user before the data is collected. 3. The design information according to claim 1, wherein the design information includes any one of data stored in the processing device, input / output data of the processing device or program, and a relation between data by a relational database. A method for collecting diagnostic information for a distributed processing system. 4. The distributed processing system according to claim 1, further comprising a function of detecting a processing device or a program at a transmission source or a transmission destination for data input / output by the program, and acquiring the data as design information. A method of collecting information relating to the transmission source or the transmission destination, and a step of extracting only data relating to a specified processing device or program. 5. The distributed processing system according to claim 4, further comprising a step of detecting that a program running on the processing device has been added, and a step of acquiring input / output data of the added program. A diagnostic information collection method for a distributed processing system, which is a feature.