JP2908442B1 - Trace information collection method - Google Patents

Abstract

Abstract: A minimum required trace information can be reliably obtained according to a system construction status, a failure status during operation, and the like without making any change to the processing itself of a communication management program. Provides a trace information collection method. SOLUTION: Initial setting processing means 11 creates a terminal resource management table 4 based on system configuration definition information in a system configuration definition (SG) information file 2. The trace level setting processing means 12 determines a trace level for each terminal resource based on the trace level information in the trace level information file 5 in the terminal resource management table 4.
Set to. The trace information collection processing means 14 collects information specified at specified points based on the trace level set in the terminal resource management table 4 and the collection points and collection information set in the collection information table 16. Then, the collected trace information is stored in the trace information storage file 3.

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 trace information in a communication management program.
The present invention relates to a trace information collection method which enables a communication information management program to arbitrarily set a trace information collection level and a trace information collection point for each terminal resource managed by the communication management program.

[0002]

2. Description of the Related Art Japanese Unexamined Patent Publication (Kokai) No. Sho 59-77553 discloses that a trace of only a specific program is minimized by adding a small amount of hardware to a data collection device to minimize the load on a computer. A data collection method is described, which makes it possible to increase the efficiency of the trace memory and reduce the memory capacity. This data collection method is configured as follows. The arithmetic processing unit, the main storage unit, and the I / O control unit are connected to a common bus, and a data collection device is connected to the common bus to form a data collection system. This device is provided with a data collection control circuit, a data collection memory, a program level register, a comparator, a trace level designation register, an OR gate, and the like. The operating program level signal is stored in the program level register, the value of the register set by the input device is compared with the level signal by the comparator, and when they match, a match signal is added to the gate. The output of the gate is input to a data collection circuit, which enables tracing of a specific program, and displays the data on a display lamp during data collection, thereby improving the efficiency of the memory.

Japanese Unexamined Patent Publication No. 1-231136 discloses a trace necessary for confirming the processing flow of a program by dividing a passing derail point from which trace information is to be output into levels 1 and 2. A trace information output method is described in which the amount of trace information to be output can be reduced to a minimum while covering the trace information. This trace information output method is configured as follows. The trace information edit output routine registration means registers the trace information edit output means for outputting trace information at the start of execution of the program in the program. Further, the trace information storage unit temporarily stores a certain number of latest trace information. When the trace information stacking means reaches a passing derail point preliminarily divided into level 1 and level 2 during execution of the program, a passing derail point name is recorded at level 1 as trace information. At level 2, the passing derail point name and the processed information are stored in the trace information storage unit.

Japanese Unexamined Patent Publication No. Hei 1-231137 discloses that a passing derail point from which trace information is to be output is divided into level 1 and level 2, and trace information of a fixed size corresponding to each level. Is temporarily stored in a trace information storage unit, so that necessary trace information is covered and the amount of trace information to be output can be reduced to a minimum. The method is described. This trace information output method is configured as follows. The trace information storage unit temporarily stores a certain number of latest trace information. The trace information stack means includes:
During the execution of the program, the level of the passing derail point is automatically determined when the passing derail point is divided into the level 1 and the level 2 in advance.
In level 1, the passing derail point name is stored in the trace information storage unit, and in level 2, the passing derail point name and processed information are stored in the trace information storage unit.

Japanese Patent Application Laid-Open No. 3-282833 discloses that trace information can be collected for assisting effective analysis of the running state of a program by storing trace information on a stack for each level. Describes the trace information collection method. This trace information collection method is configured as follows. The trace information storage means for each level is called at a timing when the trace information is collected from the program for performing the trace, and generates the trace information to store the trace information of the corresponding level among the upper level stack and the lower level stack. Store it on the stack. Then, the trace information time-series combination editing means is started at the timing of editing and outputting the trace information, and the trace information stored on the stack of each level is output.
Combine and edit source information and output it to a printer. As a result, trace information of a large acquisition unit with respect to the running state of the program can be collected at a higher level, and the resolution for analyzing the timing in the vicinity of a necessary portion is guaranteed by the lower level trace information. be able to.

Japanese Patent Laid-Open Publication No. Hei 7-36745 describes a trace data collecting method which enables data to be collected for each protocol control structure at a protocol level. This trace data collection method is configured as follows. The selection criterion of the protocol control structure corresponding to the data to be collected is stored in the registration table via the selection criterion setting means and the registration means. When a protocol control structure is generated in the protocol, the judging means judges whether or not the protocol control structure is within the selection criteria registered in the registration table. If it is determined that the data is within the standard, the trace control information is embedded in the protocol control structure. Data for each protocol
When the data is generated, the trace data collecting means outputs the data.
It is determined whether or not trace control information is embedded in a protocol control structure for controlling the transfer of data. Only when embedded, data is stored in a trace data collection file.

[0007]

The level of the trace information of the communication management program to be collected differs depending on the situation. In a normal system construction, an operation check (referred to as a continuity test) in the sense of a connection check is performed for each line type and procedure at an initial stage, then a connection test for all terminals is performed, and then actual operation is started. In particular, in the continuity test stage during the construction of the system, various failures often occur due to a malfunction of the application, an error in the setting information of the system, and the like.
In order to ascertain these causes reliably, it is necessary to always collect detailed trace information. On the other hand, since the program quality becomes stable after the system shifts to the operation stage, the system is modified so that detailed information is collected only on the failure processing route. This is to prevent the necessary information from being overwritten and lost when a large amount of information is acquired because the size of the trace information storage file is finite and is used cyclically. .

Further, when a failure occurs in a specific terminal or line during the operation of the system, trace information is acquired for analyzing the cause of the failure. At this time, necessary information is surely acquired. It is necessary to create and provide a property that incorporates trace information collection processing aimed at terminal names and the like so that users and developers are increasing the work for collecting information.

For this reason, it is possible to reliably acquire the minimum necessary trace information according to the system construction status, the failure status during operation, etc., without making any changes to the processing itself of the communication management program. It was desired.

In the trace information output method described in Japanese Patent Application Laid-Open No. 1-231136, when a derail point prepared in a program is reached, a level and contents designated in advance are stored as trace information. it can. In the trace information output method described in Japanese Patent Application Laid-Open No. 1-231137, when a derail point prepared in a program is reached, the level is automatically determined, and the contents designated in advance are replaced with the trace information. Can be stored as Therefore, trace information can be stored in a separate file for each level.

However, according to the trace information output method described in Japanese Patent Application Laid-Open Nos. 1-231136 and 1-231137, an arbitrary point for collecting trace information is set in a program, and the point is set. Cannot be stored as trace information in accordance with the level specified in the file for each resource. For this reason, it has been desired that the level can be specified finely for each resource and can be specified from the outside.

The present invention has been made to solve such a problem, and does not change the processing of the communication management program itself according to the system construction situation or the failure situation during operation. An object of the present invention is to provide a trace information collecting method capable of reliably acquiring minimum trace information and thereby reducing the number of steps required for troubleshooting.

[0013]

According to a first aspect of the present invention, there is provided a trace information collecting method for creating a terminal resource management table based on system configuration definition information stored in a system configuration information definition file. Initial setting processing means, trace level setting processing means for setting a trace level specified for each terminal resource in a terminal resource management table based on trace level information stored in a trace level information file, and collecting trace information A collection information table in which points (collection points) and information to be collected for each trace level (collection information) are set, a trace level set in the terminal resource management table, a collection point set in the collection information table, and collection information. Based on the specified point at the specified point It was taken broadcast, and a trace information collection processing means for storing the trace information storage file the collected trace information.

According to a second aspect of the present invention, there is provided a trace information collecting method.
2. A fault count processing means for counting the number of fault occurrences for each terminal resource in the trace information collecting method according to claim 1, and a trace level updating process for automatically changing a trace level according to the fault occurrence frequency. And means.

In the automatic trace collection level changing method according to the first aspect, a point (collection point) for collecting trace information can be arbitrarily set, and the set collection information corresponding to the trace level can be collected as trace information. it can. Furthermore, the collection point of trace information and the information to be collected can be specified in detail for each terminal resource without changing the processing of the communication management program itself. it can.

According to a second aspect of the present invention, there is provided a trace information collecting method.
When a failure occurs frequently in a specific terminal or line, the trace collection level can be automatically changed so that more detailed information can be obtained. As a result, a more detailed level of trace information is automatically collected. As a result, the information collected later becomes the most useful information. Therefore, even if the trace information storage file 3 becomes full and the information is overwritten, information necessary for analysis is lost. It does not happen.

[0017]

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram showing one embodiment of a trace information collecting method according to the present invention. The trace information collection method shown in FIG. 1 uses a communication management program 1 and system configuration definition information (SG information) such as terminal information which is a resource to be managed by the communication management program 1.
Information file (system configuration definition information file) 2 that stores therein, a trace information storage file 3 that stores trace information, and terminal resource management created on the main storage device based on the system configuration definition information (SG information) It comprises a table 4 and a trace level information file 5.

The communication management program 1 includes an initial setting processing means 11, a trace level setting processing means 12, a communication processing means 13, a trace information collection processing means 14, a termination processing means 15, and a collection information which is an internal table. And a table 16.

The initialization processing means 11 reads system configuration definition information (SG information) necessary for the operation of the communication management program 1 from the SG information file 2 and
The terminal resource management table 4 is created on the main storage device based on the contents read from the information file 2.

The trace level setting processing means 12 reads the trace level information from the trace level information file 5 and sets the trace information collection level in the terminal resource management table 4.

The communication processing means 13 generates a communication path, transmits and receives data, disconnects the communication path, and the like based on a request from a terminal or a host. The communication processing means 13
Performs status management of various terminal resources according to the contents of the terminal resource management table 4.

The trace information collection processing means 14 is called from the communication processing means 13 as necessary, and refers to the contents set in the collection information table 16 according to the trace level set in the terminal resource management table 4. The trace information is collected and collected, and stored in the collected and collected trace information storage file 3.

In the collection information table 16, for each place where the trace information collection processing means 14 is called, the information to be collected is set according to the trace level at that time.

The termination processing means 15 performs post-processing for terminating the communication management program 1 based on a stop instruction from a user or the like.

FIG. 2 is an explanatory diagram showing the contents of the initial setting process, the terminal resource management table creation process, and the trace level setting process. The initial setting (S
In the (G information reading) process, terminal information (SG information) is read from the SG information file 2.

In the terminal resource management table creation processing shown in step S2, the terminal resource management table 4 is created based on the terminal information (SG information) read in step S1.
Is created on the main storage device. In the terminal resource management table 4, information relating to terminal resources to be managed by the communication management program 1 is set. In FIG. 2, specific information on terminal resources includes a terminal name, a terminal type,
An example in which a line name is set is shown.

In the trace level information reading process shown in step S3, trace level information is read from the trace level information file 5. As the trace level information, a trace level (level) for one or a plurality of terminals can be set for each category such as a terminal name, a terminal type, and a line name.

In the trace level setting process shown in step S4, the trace level (level) is set in the corresponding entry of the terminal resource management table 4 with reference to the trace level information read in step S3.

In FIG. 2, trace level information (level)
Is divided into three stages from level 1 (summary level) to level 3 (detail level).
In the trace level setting process (step S4),
When a plurality of levels are specified for the same terminal, a more detailed level is set in the terminal resource management table 4.

In the example shown in FIG. 2, since the level “2” is set for “line name = Line_1” in the trace level information, the terminal names “Term_1, Term_2, Term_3” in the terminal resource management table 4 are set. , The trace level (level) is set to level 2 first. Next, in the trace level information, “terminal name = T
Since the level “3” which is a more detailed level is set for “erm_2”, the trace level (level) of the terminal name Term_2 in the terminal resource management table 4
To level “3”. In this way, by setting all the levels specified by the trace level information in the terminal resource management table 4, the terminal management table 4 illustrated in FIG. 2 is created.

Note that no trace level is set for a terminal for which no trace level information has been set. In this case, the trace level (level) on the terminal resource management table 4 is the initial level “0”. The initial value “0” means that no trace information is collected. In the terminal resource management table illustrated in FIG.
“Terminal name = Term — 6” corresponds to this.

FIG. 3 is an explanatory diagram showing a specific example of the collection information table. The collection information table 16 creates one entry for each call point (information collection point) of the trace information collection processing. Each entry stores an information name to be collected for each trace information collection level. In the example of FIG. 3, four points of point A, point B, point C, and point D are registered as information collection points. Also,
Level 1, level 2, as trace level (level)
Three levels of level 3 are set, and the information name to be collected at that time is set in each column. For example, point B
If the trace level of the specified terminal resource is "3" when the trace information collection processing is called in step 3, the transmission processing result code (result code), transmission data and transmission destination information at that time are collected as trace information. You. If the trace level is “1”, no trace information is collected because the corresponding column is blank (not collected).

FIG. 4 is an explanatory diagram showing the operation of collecting trace information. FIG. 4 shows a detailed processing operation example of the path generation processing executed by the communication processing unit 13 shown in FIG. Here, as events that occur during the path generation processing, “path generation preparation completed”, “path generation request transmission”,
It is assumed that there are four events of “reception of path generation response” and “success of path generation”, and point A, point B, point C, and point D are assigned as identifiers of respective points. In these location points A to D of the communication process, a procedure for calling the trace information collection process using the terminal resource name being processed and an identifier for uniquely identifying the calling location as parameters is described.

The trace information collection processing means 14 uses the terminal resource name passed as a parameter as key information,
The terminal resource management table 4 is searched. Then, when an entry that matches the terminal resource name is found, the trace information collection level set in the entry is obtained. Next, the trace information collection processing means 14 searches the collection information table 16 using the identifier of the call location passed as a parameter as key information. Then, when an entry that matches the call location is found, the information set in that column is collected as trace information by referring to the column corresponding to the previously obtained trace information collection level,
The collected trace information is stored in the trace information storage file 3.

When the terminal resource management table 4 and the collection information table 16 are set according to the contents illustrated in FIGS. 2 and 3, for example, the operation when the collection processing of the trace information is called from the point B will be described below. explain. The parameter passed when the trace information collection process is called from the point B is “terminal name = Term_
3, caller identifier = point B ". The trace information collection processing means 14 searches the terminal resource management table 4 using “terminal name = Term — 3” as a key. Then, “2” is obtained as the trace information collection level (level) set in the terminal resource management table 4. Next, the trace information collection processing means 14 searches the collection information table 16 using “caller identifier = point B” as a key, refers to the trace level 2 column of the found entry, and As the result code. Then, the trace information collection processing means 14
Collects a terminal name and a result code as information to be collected and stores them in the trace information storage file 3.

In the normal system construction, an operation check (continuity test) in the sense of connection check is performed for each line type or procedure at an initial stage, then a connection test for all terminals is performed, and then actual operation is started. . In the trace information collection method shown in FIG. 1, the terminal for acquiring the trace information and the collection level of the trace information can be freely set from the outside as a trace information file. It is possible to reliably acquire the minimum necessary trace information according to the condition. Further, since the terminal for acquiring the trace information and the trace information collection level can be freely set from the outside as a trace information file, it is necessary to change any processing program of the communication management program 1 when changing the trace conditions. Absent. Therefore, for developers and users,
There is a merit that the man-hour at the time of troubleshooting can be greatly reduced.

In the trace information collecting method shown in FIG. 1, a point at which trace information is collected (collection point)
In addition, since the information to be collected (collected information) can be set as the collected information table 16 for each trace level, the point at which the trace information is collected (collection point) can be arbitrarily set, and the trace level can be set. The collected information (for example, result code, parameter information, etc.) set as above can be collected and collected as trace information. That is, the collection point of the trace information and the contents of the collection (collection) information (for example, result code, parameter information, etc.) can be specified in detail from the outside for each terminal resource.

FIG. 5 is a block diagram showing another embodiment of the trace information collecting method according to the present invention. The trace information collection method shown in FIG. 5 is different from the trace information collection method shown in FIG. 1 in that a failure count counting means 17 and a trace level update processing means 18 are added, and a failure occurrence Number of times (number of failures)
Column has been added.

The communication processing means 13 calls the failure frequency counting processing means 17 when a failure occurs during the operation of the communication management program main body 1. When called by the communication processing means 13 when an operation failure of the communication management program body 1 occurs, the failure frequency counting processing means 17 identifies the terminal resource in which the failure has occurred, and counts the number of failure occurrences of the identified terminal resource. ), And updates the number of times of failure in the number of times of failure (number of times of failure) column of the terminal resource management table 4 corresponding to the specified terminal resource (+1 is added to the number of times of failure).

The trace level update processing means 18 periodically monitors the number of failures (the number of failures). If the frequency of failures increases, the trace collection level corresponding to the terminal resource is set to a more detailed level. Change to For example, when a failure event occurs a predetermined number of times or more within a predetermined time, the trace collection level is incremented by one. Note that the trace level update processing means 18 returns the trace collection level to the initial (before change) setting level if a failure does not occur even after a lapse of a predetermined time after performing the change for increasing the trace collection level. It may be configured.

With the above configuration, the trace information collection method shown in FIG. 5 automatically changes the trace collection level so that more detailed information can be obtained when a failure occurs frequently in a specific terminal or line. can do. This allows
More detailed levels of trace information are automatically collected. As a result, the information collected later becomes the most useful information. Therefore, even if the trace information storage file 3 becomes full and the information is overwritten, information necessary for analysis is lost. It does not happen.

[0042]

As described above, the trace information collecting method according to claim 1 is a trace level setting processing means for setting a trace level specified for each terminal resource in the terminal resource management table based on the trace level information. For collecting trace information (collection points)
And a collection information table in which information (collection information) to be collected for each trace level is set, and a trace level set in the terminal resource management table, a collection point set in the collection information table, and collection information.
The system is equipped with a trace information collection processing unit that collects the specified information at the specified point and stores it in the trace information storage file. Therefore, the point (collection point) where the trace information is collected can be set arbitrarily. Collection information (for example, a result code, parameter information, and the like) set corresponding to the trace level can be collected and collected as trace information. That is, the collection point of the trace information and the contents of the collection (collection) information (for example, result code, parameter information, etc.) can be specified in detail from the outside for each terminal resource. further,
The trace information collection point and the content of the collected information can be specified in detail for each terminal resource without changing the processing of the communication management program itself. Therefore, man-hours for troubleshooting can be significantly reduced.

The trace information collecting method according to claim 2 is
A failure frequency counting means for counting the number of failure occurrences for each terminal resource and a trace level update processing means for automatically changing a trace level according to the failure occurrence frequency are provided, so that a specific terminal or line can be used. If there are many obstacles,
The trace collection level can be automatically changed to obtain more detailed information. As a result, a more detailed level of trace information is automatically collected. As a result, information collected later is the most useful information,
Even if the capacity of the trace information storage file 3 becomes full and information is overwritten, information necessary for analysis is not lost.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a trace information collection method according to the present invention.

FIG. 2 is an explanatory diagram showing contents of an initial setting process, a terminal resource management table creation process, and a trace level setting process.

FIG. 3 is an explanatory diagram showing a specific example of a collection information table.

FIG. 4 is an explanatory diagram showing an operation of collecting trace information.

FIG. 5 is a block diagram showing another embodiment of the trace information collecting method according to the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Communication management program body (communication management program) 2 System configuration definition information file (SG information file) 3 Trace information storage file 4 Terminal resource management table 5 Trace level information file 11 Initial setting processing means 12 Trace level setting processing means 13 Communication processing Means 14 Trace information collection processing means 15 Termination processing means 16 Collection information table 17 Failure frequency counting processing means 18 Trace level update processing means

Claims (2)

(57) [Claims] 1. A system configuration definition information file for storing system configuration definition information, a trace level information file for storing trace level information, and an initial setting processing means for creating a terminal resource management table based on the system configuration definition information Trace level setting processing means for setting a trace level specified for each terminal resource in the terminal resource management table based on the trace level information, points for collecting trace information, and information to be collected for each trace level Is specified based on the trace information set in the terminal resource management table, the point at which the trace information set in the collected information table is collected, and the information to be collected for each trace level. Specified by points A trace information collection method for collecting information and storing the information in a trace information storage file. 2. A system configuration definition information file for storing system configuration definition information, a trace level information file for storing trace level information, and an initial setting processing unit for creating a terminal resource management table based on the system configuration definition information. Trace level setting processing means for setting a trace level specified for each terminal resource in the terminal resource management table based on the trace level information, points for collecting trace information, and information to be collected for each trace level Is specified based on the trace information set in the terminal resource management table, the point at which the trace information set in the collected information table is collected, and the information to be collected for each trace level. Specified by points Trace information collection processing means for collecting information and storing the information in a trace information storage file; failure count processing means for counting the number of times a failure has occurred for each terminal resource; and setting the trace level according to the frequency of occurrence of the failure. A trace information collecting method comprising a trace level update processing means for automatically changing the trace level.