JP6920235B2 - Failure impact investigation device and failure impact investigation method - Google Patents

Description

The present invention relates to a failure impact investigation device and a failure impact investigation method. Specifically, the failure impact investigation in a batch job is comprehensively and efficiently performed, and the investigation results are obtained according to needs. The present invention relates to a technique that can be preferably presented.

In large-scale systems such as core systems in financial institutions, there are many cases where a series of jobs are appropriately combined and batch-processed to perform information processing necessary for business execution. If a failure occurs in any of the jobs batch-processed in this way, the influence on other related jobs may spread and the execution of the relevant work may be hindered.

Therefore, there is a technology that responds to the occurrence of such a failure, collects necessary information, and connects it to recovery. As such a technique, for example, a job management system in which a server and a client are connected so as to be able to communicate with each other via a network, and the server is arbitrarily set for each job net and job. The element (successor jobnet and / or job), the recovery priority level, the jobnet job database that stores the end deadline time and the scheduled start time, and the successor element of the jobnet or job in which the error occurred are abnormal. An estimated recovery time database that stores the estimated recovery time from the occurrence to the start of normal operation, an abnormality detection means for detecting the occurrence of an error in a job net or job, and a job net or job in which an error has occurred. A recovery priority level determination means that determines whether or not a recovery priority level is set for each succeeding element, and if so, compares the recovery priority levels among the succeeding elements, and a job net in which an error has occurred. Alternatively, it is determined whether or not the end deadline time is set for each succeeding element of the job, and if it is set, the end deadline time determination means for comparing the end deadline time among the succeeding elements and an abnormality are found. A means for determining the scheduled start time, which determines whether or not the scheduled start time is set for each succeeding element of the generated job net or job, and if so, compares the scheduled start time among the succeeding elements. And, it is determined whether or not the estimated recovery time is memorized for each succeeding element of the job net or job in which the abnormality occurred, and if it is memorized, the estimated recovery time is compared between the succeeding elements. A job comprising: an estimated time determining means, a recovery order determining means for determining a recovery order of a succeeding element based on the determination comparison result, and a notification means for notifying the user of the determined recovery order. A management system (see Patent Document 1) and the like have been proposed.

In addition, the job definition database that can define the job definition and the predecessor / successor wait conditions and the job definition information stored in this job definition database are read, and the predecessor job of the defined job and the predecessor / successor relationship of the succeeding job are displayed. In a job operation management system provided with a search means for searching, a preceding job is specified as a search condition on the search screen provided by the search means, a job search having a successor relationship to the preceding job is executed, and search result information is retained. A job operation management system, which comprises means for displaying a succeeding job having a succeeding relationship to the preceding job on the search screen by redefining one of the search result information and performing a recursive search (the job operation management system). (See Patent Document 2) and the like have also been proposed.

Japanese Unexamined Patent Publication No. 2015-64723 Japanese Unexamined Patent Publication No. 2013-257803

By the way, the above-mentioned job is a component of a job net composed of a series of job groups. The job net is also a component of a series of job nets. Furthermore, the job net group can also be a unit element of the higher-level configuration. In such a job inclusion relationship, the above-mentioned job forms the minimum processing unit related to the business, and the inclusion relationship expands as the group of jobs gradually increases to the job net including the job net and the job net group including the job net. .. In other words, for each business division of each business, the inclusion relationship related to the job is defined in the form of a hierarchical structure according to the size of the processing unit.

In the above-mentioned hierarchical structure, one of the jobs fails, and an attempt is made to specify the range of influence. However, in the prior art, etc., the range of influence extends beyond the affiliation configuration and hierarchy of the job, or spans other business divisions and other businesses, and further spans jobs derived from external systems. Could not be identified.

Therefore, in order to identify the extent of such influence, it is necessary to conduct a human-assisted survey of a reasonable scale each time. Such a survey requires a huge number of jobs to be thoroughly surveyed, which is extremely difficult in itself, and also has a strong personality that depends on the skills and experience of the responding personnel. The accuracy and efficiency of the survey are also likely to be unstable.

Further, even if the above-mentioned investigation is conducted, the function of comprehensively displaying the information of the influence range as the investigation result and dynamically controlling the content according to the needs of the user is not provided.

Therefore, an object of the present invention is to provide a technique for comprehensively and efficiently conducting a failure impact survey in a batch job and for appropriately presenting the survey results according to needs.

The failure impact investigating apparatus of the present invention that solves the above problems is the job or the job as information regarding the hierarchical structure of a batch job including at least each layer of a job related to each job and a job net composed of the series of the jobs. A storage device that stores at least the information of the connection rule that defines at least the processing status of one of the nets or the trigger of waiting or execution of the other according to the processing result, and the operation log of the job or the job net. Based on the input / output relationship between jobs and job nets, which is indicated by the information on the hierarchical structure, the input / output relationship is established between the jobs and the jobs starting from the job for which an abend occurs in a predetermined job. A process for specifying a series of jobs and job nets to have as a failure influence range, and a directed graph or an undirected graph in which at least each of the jobs and job nets belonging to the failure influence range is set as a node and each of the nodes is connected by an edge. , A list of jobs and job nets belonging to the failure impact range and a graph in which the number of jobs or job nets belonging to the failure impact range are aggregated and displayed for each business unit are generated and output to a predetermined device. Jobs that execute processing and are included in the sequence with the passage of time with the occurrence of the abend as the reference time, starting from the job that was the target of the abend in the failure impact range. And as a job net, the operation timing indicated by the operation log is included between the reference time and the user-specified time, and the operation is estimated according to the standby or execution trigger indicated by the information of the connection regulation. A process of specifying at least one of the directed graph, the undirected graph, the list, and the graph and outputting the specified job and job net to a predetermined device is further executed. It is characterized by including a certain arithmetic device.

In addition, the failure impact investigation method of the present invention provides information on the hierarchical structure of a batch job including at least each layer of a job related to each job and a job net composed of a series of the jobs, and each of the jobs or the job net. An information processing device including a storage device that stores at least the information of the connection rule that specifies at least the processing status of one process or the trigger of standby or execution of the other according to the processing result, and the operation log of the job or the job net. However, based on the input / output relationship between jobs and job nets, which is indicated by the information on the hierarchical structure, input / output from the job that is the target of the abend in the predetermined business is started from the job, and the data is input / output within the business and between the business. A process for specifying a series of related jobs and job nets as a failure impact range, and a directed graph or undirected graph in which at least each of the jobs and job nets belonging to the failure impact range is a node and each of the nodes is connected by an edge. Generate at least one of a graph, a list of jobs and job nets belonging to the failure influence range, and a graph in which the number of jobs or job nets belonging to the failure influence range is aggregated and displayed for each business unit, and the predetermined device is used. Jobs and job nets that will be included in the sequence with the elapse of time with the output process and the job that was the target of the abend as the reference time in the failure impact range. As the operation timing indicated by the operation log is included between the reference time and the user-specified time, and the operation is presumed according to the standby or execution trigger indicated by the information of the connection regulation. Then, with respect to the specified job and job net, a process of generating at least one of the directed graph, the undirected graph, the list, and the graph and outputting the graph to a predetermined device is executed. do.

According to the present invention, it is possible to make the failure effect investigation in a batch job comprehensive and efficient, and to appropriately present the investigation result according to the needs.

It is a network configuration diagram which includes the failure influence investigation apparatus of this embodiment. It is a figure which shows the hardware configuration example of the failure influence investigation apparatus in this embodiment. It is a figure which shows the concept of the hierarchical structure example of a job in this embodiment. It is a figure which shows the job hierarchy structure example 1 of this embodiment. It is a figure which shows the job hierarchy structure example 2 of this embodiment. It is a figure which shows the data structure example of the hierarchical structure information of this embodiment. It is a figure which shows the data structure example of the connection information of this embodiment. It is a figure which shows the data structure example of the hold / hold release information of this embodiment. It is a figure which shows the data structure example of the operation log of this embodiment. It is a figure which shows the data structure example of the event transmission file of this embodiment. It is a figure which shows the data structure example of the event reception file of this embodiment. It is a figure which shows the flow example 1 of the failure influence investigation method in this embodiment. It is a figure which shows the flow example 2 of the failure influence investigation method in this embodiment. It is a figure which shows the flow example 3 of the failure influence investigation method in this embodiment. It is a figure which shows the flow example 4 of the failure influence investigation method in this embodiment. It is a figure which shows the flow example 5 of the failure influence investigation method in this embodiment. It is a figure which shows the screen example 1 in this embodiment. It is a figure which shows the screen example 2 in this embodiment. It is a figure which shows the screen example 3 in this embodiment. It is a figure which shows the screen example 4 in this embodiment. It is a figure which shows the screen example 5 in this embodiment.

--- Network configuration ---
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a network configuration diagram including the failure impact investigation device 100 of the present embodiment.

The failure impact investigation device 100 shown in FIG. 1 is a computer device that makes failure impact investigation in a batch job comprehensive and efficient, and can appropriately present the investigation results according to needs.

The failure impact investigation device 100 can be assumed to be, for example, a server device operated by a system management department of a financial institution, a vendor of its outsourcing contractor, or the like. The failure impact investigation device 100 is communicably connected to the user terminal 200 and the investigation target system 300 via the secure network 10 in the operating organization.

Of these, the user terminal 200 is an information processing device operated by a person in charge of failure impact investigation work in the above-mentioned system department, vendor, or the like. Further, the survey target system 300 assumes a business system in a financial institution or the like. The business system that is the survey target system 300 is not particularly limited.
--- Hardware configuration ---

Further, the hardware configuration of the failure impact investigation device 100 of the present embodiment is as follows. FIG. 2 is a diagram showing a hardware configuration example of the failure impact investigation device 100 according to the present embodiment.

The failure impact investigation device 100 is held in a storage device 101 composed of an appropriate non-volatile storage element such as an SSD (Solid State Drive) or a hard disk drive, a memory 103 composed of a volatile storage element such as a RAM, and a storage device 101. The program 102 is executed by reading it into the memory 103 to perform overall control of the device itself, and is connected to a computing device 104 such as a CPU that performs various determinations, calculations, and control processes, and a user terminal 200 or a survey by connecting to the network 10. A communication device 105, which is responsible for communication processing with other devices such as the target system 300, is provided.

If necessary, the failure response support device 100 of the present embodiment may further include an input device that accepts key input and voice input from the user, and an output device such as a display that displays processing data.

Further, in the storage device 101, in addition to the program 102 for implementing the function required as the failure impact investigation device 100 of the present embodiment, the hierarchical structure information 125, the connection information 126, the hold / hold release information 127, and the operation At least the log 128, the event transmission file 129, and the event reception file 130 are stored. Details of each of these pieces of information will be described later. Further, the failure impact investigation device 100 of the present embodiment holds the information 150 of the failure impact range acquired by executing the failure impact investigation method in the memory 103, and serves as a source of display processing to be performed in response to a subsequent user instruction. do.
--- About the hierarchical structure of jobs ---
Here, the concept of the hierarchical structure of jobs in the present embodiment will be described. 3 to 5 show the concept of an example of a hierarchical structure of jobs in the present embodiment, respectively.

First, as illustrated in FIG. 3, in the hierarchical structure, for example, the root job group corresponding to the scheduler service (example: server unit) of the business is set as the highest hierarchy.

Further, this root job group includes a plurality of job groups 1 in the subordinate hierarchy. This job group 1 is provided for each business system such as foreign exchange and deposit, and its value is, for example, system ID + business system name.

Further, the job group 1 includes a plurality of job groups 2 in the subordinate hierarchy. This job group 2 is provided for each calendar type of batch job execution, and its value is, for example, T-regular, U-irregular, or the like.

Further, the job group 2 includes a plurality of root job nets in the subordinate hierarchy. This root job net is provided for each business category such as foreign remittance and deposit balance management, and its value is, for example, system ID + calendar type + business category + serial number.

In addition, the nested job net includes a plurality of jobs in the subordinate hierarchy. This nested job net is provided for each function to be fulfilled by executing a series of jobs such as execution of a predetermined program, classification and transmission of execution results, and the value thereof is, for example, the root job net. Value + work cycle + serial number.

Further, the job is located at the lowest layer in the hierarchical structure, that is, is a terminal component, and is provided for each unit function such as execution of a predetermined program, classification and transmission of execution results, and its value is For example, the value of the nested job net + serial number.

In such a hierarchical structure, when a visual example of the relationship between a job and a nested job net and a nested job net and a root job net is visually shown, the configuration is as shown in FIG.

In this way, a nested job net is composed of a series of jobs, which is the smallest processing unit. In addition, the nested job nets are configured to cooperate with each other by using the processing results as input / output of each other. Furthermore, the root jobnet includes the nested jobnet that cooperates in this way.

As shown in FIG. 4, the jobs belonging to the nested job net are defined as a series of jobs within the nested job net, and the connection relationship according to the input / output relationship between the jobs is also defined. Further, when the nested job nets also have an input / output relationship with each other, a connection relationship between the nested job nets also exists. However, based only on the existing provisions of such a connection relationship, for example, "Job 5" of "Nested Job Net A" and "Job 6" of "Nested Job Net B" in FIG. 4 have a connection relationship. not exist.

On the other hand, when an abend occurs in "Job 5" of "Nested Job Net A", the effect naturally spreads to the processing result of "Nested Job Net A" to which it belongs, and it is a "Nested Job" having an input / output relationship. It also extends to "Job 6" of "Net B" (input of the output of nested job net A). In other words, it is not possible to comprehensively specify the range of failure effects that spread from the job for which the abend occurs, just by taking into account the existing connection-related regulations.

Therefore, in the present embodiment, as illustrated in FIG. 5, a prenode corresponding to a virtual job that receives data input from another nested job net in the nested job net and a virtual data output to the other nested job net are performed. Each information of the post node corresponding to the job is set and retained for each nested job net.

In the example shown in FIG. 5, it is specified that each of "Job 1" to "Job 5" belonging to "Nested Job Net A" has a connection relationship with each of the prenode 50 and the post node 51.

When such a concept is introduced, if an abend occurs in "Job 5" of "Nested Job Net A", the effect is the temporary post node 51 corresponding to the output processing of the "Nested Job Net A" to which it belongs. Through the provisions of the connection relationship, it can be specified that the temporary pre-node 50 corresponding to the input processing in the "nested job net B" is reached.

That is, by taking into account the provisions of the connection relationship to which the pre-node 50 and the post-node 51 are added, it is possible to comprehensively specify the range of failure influence even in the range that could have been specified in the past.
--- Data structure example ---
Subsequently, the tables used by the fault impact investigation device 100 of the present embodiment will be described. FIG. 6A shows an example of the hierarchical structure information 125 in this embodiment.
The hierarchical structure information 125 of the present embodiment is information that stores information related to the hierarchical structure already shown in FIGS. 3 to 5 described above.

The data structure is the value of job group 1 corresponding to the business system such as "forex" (for example, F: foreign exchange), and the value of the calendar type added to this for job group 2 (for example, for example).
F: Foreign exchange business / T: Regular), root job net (for example, F: Foreign exchange business / T: Regular / F-TA) with values such as business classification and its serial number added to this, and values such as work cycle Nested job net (for example, F: Forex business / T: Regular / F-TA1 / F-TA1) and jobs (for example, F: Forex business / T: Regular / F-TA1 / F-TA1- It is a collection of records in which each value such as W01 / F-TA1-W0101) is sequentially added with "/" as a hierarchical delimiter.

Further, FIG. 6B shows a data configuration example of the connection information 126 of the present embodiment. The connection information 126 of the present embodiment is information that defines the connection relationship of the components having an input / output relationship with each other in the above-mentioned hierarchical structure.

The data structure enumerates jobs having an input / output relationship and nested job nets separated by commas, and the value of the connection condition (for example, "c" in which a condition exists for input / output execution, that is, connection. : Conditional connection, input / output execution, that is, a collection of data to which a condition does not exist for connection "s": forward connection) is added at the end.

Further, FIG. 6C shows a data configuration example of the hold / hold release information 127 of the present embodiment. The hold / hold release information 127 of the present embodiment is information that defines the conditions of the components such as the job and the nested job net that are conditional connections in the connection information 126 described above.

The rule is, for example, "hold" when waiting for execution until the processing or operation status or result of a job or nested jobnet that becomes a resource of input data or interferes with execution becomes a predetermined value, such a job. Or "release hold" when the execution is started when the processing or operation status or result of the nested jobnet becomes a predetermined value. Then, the data to which such a rule is given to the job or the nested job net having the relation concerned becomes the hold / hold release information 127.

Further, FIG. 6D shows an example of data configuration of the operation log 128 of this embodiment. The operation log 128 of this embodiment is a table that stores information on the date and time when each job was actually executed.
The data structure is a collection of data in which the execution date and time are used as keys and the values of the components such as the job to be executed and the nested job net at that time are associated with each other.

Further, FIG. 6E shows an example of data configuration of the event transmission file 129 of the present embodiment. The event transmission file 129 of the present embodiment is a file in which information about an event in which a job or a nested job net executes a predetermined process and transmits a processing result to another job or a nested job net is stored. The data structure is a collection of data in which the values of each job or nested jobnet of the source and the destination are associated with an ID that uniquely identifies an event.

Further, FIG. 6F shows an example of data configuration of the event reception file 130 of the present embodiment. The event reception file 130 of the present embodiment is a file that stores information about an event in which a job or a nested job net receives a processing result from another job or a nested job net that has executed a predetermined process. The data structure is a collection of data in which the values of each job or nested jobnet of the source and the destination are associated with the ID that uniquely identifies the event.
--- Flow example 1 ---

Hereinafter, the actual procedure of the failure impact investigation method in the present embodiment will be described with reference to the figure. Various operations corresponding to the failure impact investigation method described below are realized by a program read by the failure impact investigation device 100 into a memory or the like and executed. Then, this program is composed of the code for performing various operations described below.

FIG. 7 is a diagram showing a flow example 1 of the failure impact investigation method in the present embodiment. Here, first, the failure impact investigation device 100 receives a designation regarding a job (or a nested job net or a root job net) to be generated from the user terminal 200 (s100). Specifically, for example, the IDs of jobs, nested job nets, and root job nets in the designated node column 1001 on the tool screen 1000 (FIG. 11) distributed and displayed by the failure impact investigation device 100 on the user terminal 200. Can be assumed.

In the example of the tool screen 1000 of FIG. 11, in addition to the above-mentioned designated node field 1001, the ID of the component is set in each layer of the job group 1, the job group 2, the root job net, the nested job net, and the job. The interface 1004 selected from the list accepts the designation of the job to be generated from the user terminal 200.

Further, when the designated node received in s100 is a root job net or a nested job net, the failure impact investigation device 100 indicates "Pre" indicating a pre-node with respect to the values of the root job net and the nested job net in the connection information 126. Give a value (s101). Here, for convenience of explanation, a job, a nested job net, or the like that has been designated by s100 is referred to as a "node" corresponding to the case where the job and its connection relationship are expressed by graph theory.

Subsequently, the failure impact investigation device 100 stores information such as jobs and nested job nets included in the failure impact range, starting from the job or the like to which the abend occurs, which is specified in s100 described above, for example, a job. A list array (example: a file for storing data having a plurality of value storage fields) is created in the memory 103 (s102). Further, in this job list array, information such as a job to be generated as an abend specified in s100 is stored.

Further, the failure impact investigation device 100 searches for the connection information 126 related to the abend generation target specified in s100 above by using the ID of the job or the like as a key, and the component identified by the search, that is, Information on components such as jobs that have an input / output relationship with the abend generation target is stored in the above-mentioned job list array as a job of a succeeding job (s104).

In this case, according to the connection information 126 of the present embodiment, the connection relationship between the series of jobs belonging to the nested jobnet and other nested jobnets having an input / output relationship with the nested jobnet is also the prenode 50 and the postnode. It can be identified via 51.

Since each job belonging to the nested job net is defined to be all connected to the prenode 50 and the post node 51 in the nested job net, the connection regulation between each component is traced in the connection information 126. So, you can naturally identify the relationship that straddles the nested job net. These points also apply to cases where the business system is crossed, that is, the job group is crossed, or between other vendors and other systems, as long as there is a provision in the connection information 126. Subsequent jobs can be identified.
Further, the failure impact investigation device 100 determines whether or not the above-mentioned succeeding job already exists in the job list array (s105).

As a result of the above determination, when the above-mentioned succeeding job already exists in the job list array (s105: Yes), the failure impact investigation device 100 does not add the succeeding job to the job list array (s106), and performs the process in s108. To transition to.

On the other hand, as a result of the above determination, when the above-mentioned succeeding job does not exist in the job list array (s).
105: No), the failure impact investigation device 100 adds the succeeding job to the job list array (s107), and shifts the process to s108.

Further, the failure impact investigation device 100 determines whether the job list array ends by determining whether s104 to s107 have been executed for all the records in the connection information 126 (s108).
As a result of the above determination, if it is determined that it is not the end of the job list array (s108: No), the failure impact investigation device 100 returns the process to s104.

On the other hand, when it is determined as the end of the job list array as a result of the above determination (s108: Yes), the failure impact investigation device 100 stores the job list array as information 150 of the failure impact range, for example, in the memory 103. (S109), the process is terminated.

The failure influence range information 150 is defined in the connection information 126 starting from the abend generation target among the jobs, nested job nets, etc. defined in the hierarchical structure 125, that is, there is an input / output relationship. This is the result of sequentially identifying jobs, nested job nets, etc., and indicates a series of such jobs. The data structure is the same as that of the connection information 126.
--- Flow example 2 ---

Further, FIG. 8 is a diagram showing a flow example 2 of the failure impact investigation method in the present embodiment. Here, the above-mentioned generation processing of the prenode 50 and the post node 51 will be specifically described.

In this case, the failure impact investigation device 100 reads the job list, that is, information such as jobs included in the failure impact range from the job list array obtained in the above flow example 1 (s200), and reads one record from the job list (s200). ). This one record is an ID indicating a job or a nested job net in the job hierarchy.

Subsequently, the failure impact investigation device 100 counts the number of "/" in the record obtained in s201, that is, the depth of the hierarchy in the hierarchical structure of the job, and determines whether or not the value is 5, for example. It is determined whether the record indicates a job or a nested jobnet (s202).

As a result of the above determination, if the number of "/" is not "5", or if it is found that the record is a nested job net (s202: nested job), the failure impact investigation device 100 determines that the record is From the indicated value, that is, the full path (for example, / F: / T / F-TA1 / F-TA1-W01), the root jobnet ID (system ID + calendar type + business classification + serial number, for example, / F: / T / F-TA1) is acquired (s203).

In this case, the failure impact investigation device 100 generates a prenode 50A having "ID +/ Pre of the root job net" as an ID as an input of the nested job net shown in the above record, that is, as a preceding job. Further, the failure impact investigation device 100 serves as an output destination of the prenode 50, that is, as a succeeding job, "ID + / Pre of its own nested job net".
”Is the ID of the prenode 50B (s204). The information about the prenodes generated here is stored in the connection information 126.

Further, the failure impact investigation device 100 uses a post node 51A having "ID + / Post of its own nested job net" as an ID as a preceding job that is a post node that is an output end of the failure impact investigation device 100 and is a preceding job for the root job net. Generate. Further, the failure impact investigation device 100 generates a post node 51B having "ID +/ Post of the root job net" as an ID as an output destination of the post node 51A, that is, as a succeeding job (s205). The information about the post nodes generated here is stored in the connection information 126.

On the other hand, if the number of "/" is "5" as a result of the above determination, or if it is found that the record is a job (s202: job), the failure impact investigation device 100 indicates that the record indicates. From the value, that is, the full path (for example, / F: / T / F-TA1 / F-TA1-W0101), the ID of the nested job net (root job net ID + work cycle + serial number, for example, / F: / T / F- TA1-W01) is acquired (s206).

In this case, the failure impact investigation device 100 generates a prenode 50C whose ID is the nested job net "Nest job net ID +/ Pre" to which the job belongs as an input of the job, that is, as a preceding job. .. Further, the failure impact investigation device 100 generates a prenode 50D having "own job ID +/ Pre" as an ID as an output destination of the prenode 50C, that is, as a succeeding job (s207).

Further, the failure impact investigation device 100 generates a post node 51C having "own job ID +/ Post" as an ID as a preceding job that is a post node that is an output end of the failure impact investigation device 100 and is a preceding job for the nested job net. Further, the failure impact investigation device 100 generates a post node 51D having "ID +/ Post of the nested job net" as an ID as an output destination of the post node 51C, that is, as a succeeding job (s208). The information about the post nodes generated here is stored in the connection information 126.

Subsequently, the failure impact investigation device 100 determines the end of the job list by checking whether s201 to s208 have been executed for all the records of the job list obtained in s200 (s209).
If the result of the above determination is not the end of the job list (s209: No), the failure impact investigation device 100 returns the process to s201.
On the other hand, if the result of the above determination is the end of the job list (s209: Yes), the failure impact investigation device 100 ends the process.
--- Flow example 3 ---

Further, FIG. 9 is a diagram showing a flow example 3 of the failure impact investigation method in the present embodiment. Here, when specifying the failure impact range between systems (failure impact range across servers) such as between each root job group or between job group 1 in the above-mentioned hierarchical structure, an event between the systems is used. The process of extracting the connection relationship based on the transmission / reception history that has occurred will be described. By performing such processing to improve the accuracy of the connection information 126, it is possible to identify the fault impact range with good accuracy and comprehensiveness based on the connection relationship between the business system and the server device itself, that is, the system straddling the components. Become.

In this case, the failure impact investigation device 100 reads the event transmission file 129 (s300). Further, the failure impact investigation device 100 reads the event reception file 130 (s301).

These event transmission file 129 and event reception file 130 are processed from an event in which a job or a nested jobnet executes a predetermined process and sends a processing result to another job or a nested jobnet, or from another job or a nested jobnet. It is a file that stores information about the event that received the result (FIGS. 6E, 6F).

Subsequently, the failure impact investigation device 100 reads one record from the event transmission file (s302) and determines whether it is an event transmission job (s303). This determination assumes, for example, that when the event ID (for example, "iv01" in the case of FIG. 6E) includes a specific value (example: "01"), it is determined to be an event transmission job. Such a determination method is the same for determining whether or not the job is an event reception job.
As a result of the above determination, when the record is not an event transmission job (s303: No), the failure impact investigation device 100 shifts the process to s309.

On the other hand, as a result of the above determination, when the record is an event transmission job (s303: Yes), the failure impact investigation device 100 reads one record from the event reception file (s304) and determines whether it is an event reception job. (S305).
As a result of the above determination, when the record is not an event reception job (s305: No), the failure impact investigation device 100 shifts the process to s308.

On the other hand, as a result of the above determination, when the record is an event reception job (s305: Yes), the failure impact investigation device 100 indicates the records of the event transmission job and the event reception job read in s302 and s304, respectively. It is determined whether the event IDs match (s306).
If the event IDs do not match as a result of the above determination (s306: No), the failure impact investigation device 100 shifts the process to s308.

On the other hand, when the event IDs match as a result of the above determination (s306: Yes), the failure impact investigation device 100 sets the transmission subject (job, nested job net, etc.) indicated by the corresponding record of the event transmission file 129 as the preceding job. A record including a set of IDs of each of the preceding job and the succeeding job is generated with the receiving subject (job, nested job net, etc.) indicated by the record of the event receiving file 130 as the succeeding job, and stored in the connection information 126. (S307).
Subsequently, the failure impact investigation device 100 determines whether or not the above-mentioned s304 to s307 have been completed for all the records in the event reception file 130 (s308).

As a result of the above determination, if the above-mentioned s304 to s307 have not been completed for all the records in the event reception file 130 (s308: No), the failure impact investigation device 100 returns the processing to s304.

On the other hand, as a result of the above determination, when the above-mentioned s304 to s307 are completed for all the records in the event reception file 130 (s308: Yes), the failure impact investigation device 100 describes all the records in the event transmission file 129 as described above. It is determined that s302 to s307 have been completed (s309).

As a result of the above determination, if the above-mentioned s302 to s307 have not been completed for all the records in the event transmission file 129 (s309: No), the failure impact investigation device 100 returns the processing to s302.

On the other hand, as a result of the above determination, when all the records in the event transmission file 129 have completed the above-mentioned s302 to s307 (s309: Yes), the failure effect investigation device 100 ends the process.
--- Flow example 4, 5 ---

Further, FIG. 10A is a diagram showing a flow example 4 of the failure effect investigation method in the present embodiment, and FIG. 10B is a diagram showing a flow example 5 of the failure effect investigation method in the present embodiment.

The failure effect investigation device 100 of the present embodiment can display the information 150 of the failure effect range held in the memory 103 on the user terminal 200 in a form desired by the user.

In this case, the failure impact investigation device 100 distributes the tool screen 1000 (FIG. 11) to the user terminal 200, and accepts the user designation regarding the display form on the interface 1003 of the tool screen 1000 (s400).

As an example of this user specification, first, specify whether to relate to the search result of either forward search or backward search, and then make it in one of graph, list, and job flow forms. And at least include. In order to accept each designation, the screen 1000 includes interfaces 1003 and 10031.

In the forward search and backward search, starting from the job that was the target of the abend, the forward search is for the upstream side in the input / output relationship, and the backward search is for the downstream side in the input / output relationship. There is.

Subsequently, when the user designation received in s400 is a job flow (s401: job flow), the failure impact investigation device 100 sets each of the jobs and job nets belonging to at least the failure impact range information 150 as nodes. Regarding the nodes whose input / output relationship is defined by the connection information 126, a directed graph connected by an edge is generated based on the value indicated by the record of the connection information 126 (s402), and this is used as a user terminal. It returns to 200 and displays it (s405), and ends the process.

Examples of the directed graph generated in this way are shown in FIGS. 12 and 13. In the directed graph on the display screen 1100 of FIG. 12, each component of the job group 1 is set to nodes 1110 and 1120, and those having an input / output relationship between them are specified by the connection information 126 and connected by the edge 1130. There is. As a matter of course, the failure impact investigation device 100 holds in advance the data of the drawing objects corresponding to such nodes and edges in the storage device 101, and can appropriately read and use the data. Further, in this directed graph, it is shown that the job for which the abend occurs is included in the "foreign exchange" of the job graph 1. The extent to which the impact of disability has spread from this business system "foreign exchange" extends to five business systems such as "loans" and "public bonds", with "loans" directly connected at the edge as a relay point. Can be recognized.

Further, on the display screen 1100, for example, it is assumed that a predetermined operation such as double-clicking is performed on the user terminal 200 for the "foreign exchange" node to which the abend is generated. In this case, the failure impact investigation device 100 sets the nested job nets under the root job net "F-TA1" in the business system "foreign exchange" as nodes 1210, and connects the nodes 1210 of these nested job nets to the connection information 126. It is preferable that a directed graph 1202 connected by an edge 1220 is generated based on the value indicated by, and a screen (FIG. 13) including the directed graph 1202 is distributed to the user terminal 200.

In the directed graph 1202 shown in FIG. 13, information based on the hold / hold release information 127, which is a condition regarding the connection relationship between the nodes 1210, is also included in the display contents. The hold / hold release information 127 is information that defines the conditions of the components such as the job and the nested job net that are conditional connections in the connection information 126 described above.

Specifically, for example, "hold" when waiting for execution until the processing or operation status or result of a job or nested jobnet that becomes a resource of input data or interferes with execution becomes a predetermined value. This is "release of hold" when execution is started when the processing or operation status or result of a job or nested jobnet becomes a predetermined value.

In the case of the example shown in FIG. 13, for example, the node 1210 displays the information "F-TA1-W11 OL closed_public bond # 1". This shows the condition that the nested job net "F-TA1-W11" operates when "OL closed_public bond # 1", that is, "# 1" of job group 1 "public bond" is closed online. ing.

Further, in the directed graph 1202 shown in FIG. 13, information on the operation timing of the nested job net corresponding to the node 1210 is also included in the display contents. The information regarding the operation timing is information indicating, for example, the maximum operation frequency of the past operation periods already obtained for the nested job net in the operation log 128, and is defined by, for example, a 24-hour system. It is a time value (example: if it is "5.4", it is 5:24 am).

On the other hand, when the user designation received in the above-mentioned s400 is a list display (s401: list), at least each ID (full path) of the job belonging to the information 150 of the failure influence range is extracted, and this is used, for example, the job group 1. , Job group 2, root job net, nested job net, and job, a list 1400 (FIG. 15) classified by each unit is generated (s403), and this is returned to the user terminal 200 and displayed (s405). End the process.

As illustrated in FIG. 15, this list 1400 includes values such as "depth" and "importance count" in addition to the list of the number of nodes belonging to each unit described above. Of these, “depth” is a value counted by the failure impact investigation device 100 corresponding to the value specified by the user on the interface 1420. “Depth” indicates the distance (number of hops) from the job to be avenged. If it is “0”, only the job to be avenged is generated, and if it is “1”, the job and edge to be avenged are generated. It becomes a job directly connected with.
In addition, the "importance count" is a value obtained by counting the number of corresponding items with respect to the importance determined in advance by those who have knowledge for each job.

On the other hand, when the user designation received in the above-mentioned s400 is a graph display (s401: graph), a graph in which at least the number of jobs or job nets belonging to the failure influence range information 150 is aggregated and displayed for each business unit, for example. 1300 is generated (s404), this is returned to the user terminal 200 and displayed (s405), and the process ends.

An example of the display screen 1300 of the graph generated in this way is shown in FIG. The display screen 1300 includes a reference time 1301, an elapsed time 1302 from the reference time, the remaining 1303 up to the importance A, and a graph display area 1304.

Of these, in the graph display area 1304, among the jobs indicated by the failure influence range information 150, the number of jobs included in each business system is generated and displayed on the vertical axis, and the business system is arranged on the horizontal axis.

Further, the value of the reference time 1301 indicates the abend occurrence time. Further, the elapsed time 1302 from the reference time is an interface for designating the elapsed time at which the user wants to confirm how the failure influence range changes with the passage of time from the reference time. If the value specified in this interface is "0", it is the abend occurrence time, and if it is "1", it is one hour after the abend occurrence time.

Further, the remaining 1303 until the occurrence of the importance A indicates the remaining time from the job for which the abend occurs to the operation timing of the job of the importance A affected by the failure. Therefore, for the important jobs having at least "importance A" among the jobs, the information is given as appropriate information such as the hierarchical structure information 125 and the connection information 126. Further, the operation timing of each information is the time of the maximum operation frequency indicated by the operation log 128 for the job or the like.

Therefore, as shown in the flow of FIG. 10B, if the user changes the designation of the elapsed time on the interface 1302 (s500), the failure impact investigation device 100 will specify the failure impact range again accordingly. In this case, the failure impact investigation device 100 starts from the job that is the target of the abend, sets the time when the abend occurs as the reference time, and is in the time zone until the elapsed time specified by the interface 1302 (example: midnight). ~ 2:00 am) includes the operation timing, and according to the waiting or execution trigger indicated by the connection information 126 and the hold / hold release information 127, the above-mentioned time zone (example: 0:00 to 2:00 am) ) Is expected to be in operation (for example, a certain job whose operation timing is included in the time zone is held, that is, is waiting for input, and its own operation timing is also included in the time zone. ) Is specified (s501).

Further, the failure impact investigation device 100 generates the above-mentioned graph (s502) for the job (and job net) identified in this way, and displays the graph on the user terminal 200 (s503). That is, each time the interface 1302 is specified, the fault influence range is specified, and the graph is regenerated and redisplayed.

Here, the case where the graph is generated in response to the designation of the elapsed time on the interface 1302 is shown, but such processing is the same for the generation and display of the directed graph and the list, which involves the identification of the failure influence range. To run.

Although the best mode for carrying out the present invention has been specifically described above, the present invention is not limited to this, and various modifications can be made without departing from the gist thereof.

According to this embodiment, when a failure occurs in a batch job, the range of influence is comprehensively specified for the entire system across business and vendors, and this is dynamically displayed according to the needs of the user. .. In this case, it is expected that the time and cost required for the impact investigation, which has been required in the past, will be reduced, and the user can recognize the fault impact range without omission, so that the fault response will be expedited. This will minimize the impact on related business for businesses, etc. that carry out the business, and will lead to improvement of business efficiency and success or failure of customer service.
That is, the failure impact investigation in the batch job can be comprehensively and efficiently performed, and the investigation result can be appropriately presented according to the needs.

The description herein reveals at least the following: That is, in the failure effect investigation device of the present embodiment, the storage device is a prenode corresponding to a virtual job that receives data input from another job net in the job net as information regarding the hierarchical structure, and the job. Each information of the post node corresponding to the virtual job that outputs data to another job net on the net is held, and each of the prenode and the post node and all the jobs constituting the job net are input. The information on the connection relationship having an output relationship is further stored, and the arithmetic unit further stores the input / output relationship between at least jobs and between job nets indicated by the information on the hierarchical structure in the process of specifying the failure influence range. Based on the above, when identifying a sequence of jobs and job nets that have an input / output relationship with each other within the job and between jobs, starting from the job that was the target of the abend in the predetermined job, regarding the sequence between job nets. , It may be specified based on the connection relationship via the prenode and the postnode.

According to this, since the pre-node and the post node are set in the above-mentioned hierarchical structure of the batch job, for example, when a backward search is performed starting from the job for which the abend occurs, the jobs that are continuous after the job are displayed. Even if there is no case (however, the job net to which the job belongs outputs the processing result to another job net), the above-mentioned sequence identification is not interrupted there, and the post node of the job net to which the job belongs is used. It is possible to identify the connection with other job nets (and their affiliated jobs) that are in an input / output relationship. This also applies to the forward search, and the connection with other job nets (and their jobs) having an input / output relationship can be specified via the prenode of the job net to which the job belongs.
As a result, the failure impact investigation in the batch job can be made more comprehensive and efficient, and the investigation result can be appropriately presented according to the user's needs.

Further, in the failure impact investigation device of the present embodiment, the arithmetic unit is upstream in the input / output relationship starting from the job for which the abend is generated based on the user's designation in the process of specifying the failure impact range. Either a forward search toward the side or a backward search toward the downstream side in the input / output relationship may be performed to specify the sequence of the job and the job net.

According to this, it is possible to identify the sequence of the job to which the abend generation target gives the processing data and the job to further give the processing data to the job by the backward search, and the job to give the processing data to the abend generation target job. Furthermore, it is possible to identify a sequence such as a job net that gives processing data to the job by a forward search.
As a result, the failure impact investigation in the batch job can be made more comprehensive and efficient, and the investigation result can be appropriately presented according to the user's needs.

In the failure effect investigation device of the present embodiment, the storage device, as information regarding the hierarchical structure, triggers the other to wait or execute according to the processing status or processing result of one of the jobs or the job nets. The information of the connection rule specified at least for, and the operation log of the job or the job net are further stored, and the arithmetic unit is the job for which the abend is generated in the fault influence range. The operation timing indicated by the operation log is from the reference time to the user-specified time as the jobs and job nets that will be included in the sequence with the passage of time with the occurrence of the abend as the reference time. The directed graph or the undirected graph, said The process of generating at least one of the list and the graph and outputting it to a predetermined device may be further executed.

According to this, it is possible to identify the range of influence of the failure that will change with the passage of time from the time of occurrence of the abend, and to visually clarify this. As a result, the failure impact investigation in the batch job can be comprehensively and efficiently performed, and the investigation result can be presented more preferably according to the user's needs.

In the failure effect investigation method of the present embodiment, the information processing device is a prenode corresponding to a virtual job that receives data input from another job net in the job net as information regarding the hierarchical structure in the storage device. And, each information of the post node corresponding to the virtual job that outputs data to another job net in the job net is retained, and each of the prenode and the post node and all the post nodes constituting the job net are retained. In the process of further storing the information of the connection relationship in which the job has an input / output relationship and specifying the failure influence range, it is determined based on at least the input / output relationship between jobs and job nets indicated by the information on the hierarchical structure. When identifying a series of jobs and job nets that have an input / output relationship with each other within the business and between businesses, starting from the job that was the target of the abend in the business, the connection between the job nets is the above-mentioned prenode and It may be specified based on the connection relationship via the post node.

In the failure impact investigation method of the present embodiment, in the process of specifying the failure impact range, the information processing device starts from the job that is the target of the abend based on the user's designation, and is on the upstream side in the input / output relationship. Either the forward search toward the above and the backward search toward the downstream side in the input / output relationship may be performed to identify the sequence of the job and the job net.

In the failure effect investigation method of the present embodiment, the information processing device responds to one of the processing statuses or processing results between the jobs or the job nets as information regarding the hierarchical structure in the storage device. The information of the connection regulation that at least defines the waiting or execution trigger of the above and the operation log of the job or the job net are further stored, and the job that is the target of the abend is the starting point in the failure influence range. In addition, as the time elapses with the occurrence of the abend as the reference time, the operation timing indicated by the operation log as the job and the job net to be included in the sequence is between the reference time and the user-specified time. The directed graph, the undirected graph, the list, etc. And, the process of generating at least one of the graphs and outputting it to a predetermined device may be further executed.

10 Network 100 Failure impact investigation device 101 Storage device 102 Program 103 Memory 104 Computing device 105 Input device 106 Output device 107 Communication device 125 Hierarchical structure information 126 Connection information 127 Hold / hold release information 128 Operation log 129 Event transmission file 130 Event reception file 200 User terminal 300 Survey target system

Claims (6)

As information on the hierarchical structure of a batch job including at least each layer of a job related to each job and a job net composed of a series of the above jobs , the processing status or processing result of one of the jobs or the job nets can be used. A storage device that stores the operation log of the job or the job net, and the information of the connection rule that at least defines the other standby or execution trigger that responds.
Based on the input / output relationship between jobs and job nets, which is indicated by the information on the hierarchical structure, the input / output relationship is established between the jobs and the jobs starting from the job for which an abend occurs in a predetermined job. A process for specifying a series of jobs and job nets to have as a failure impact range, and a directed or undirected graph in which at least each of the jobs and job nets belonging to the failure impact range is a node and each of the nodes is connected by an edge. , A list of jobs and job nets belonging to the failure impact range and a graph in which the number of jobs or job nets belonging to the failure impact range are aggregated and displayed for each business unit are generated and output to a predetermined device. Jobs that execute processing and are included in the sequence with the passage of time with the occurrence of the abend as the reference time, starting from the job that was the target of the abend in the failure impact range. And as a job net, the operation timing indicated by the operation log is included between the reference time and the user-specified time, and the operation is estimated according to the standby or execution trigger indicated by the information of the connection regulation. A process of specifying an object, generating at least one of the directed graph, the undirected graph, the list, and the graph and outputting the specified job and job net to a predetermined device is further executed. With a certain arithmetic device
A failure impact investigation device characterized by being equipped with. The storage device is
As information related to the hierarchical structure, it corresponds to a prenode corresponding to a virtual job that receives data input from another job net in the job net, and a virtual job that outputs data to another job net in the job net. It holds each information of the post node, and further stores the connection relationship information in which each of the prenode and the post node and all the jobs constituting the job net have an input / output relationship.
The arithmetic unit
In the process of specifying the failure impact range, based on at least the input / output relationship between jobs and job nets indicated by the information on the hierarchical structure, starting from the job that was the target of the abend in the predetermined job, within the job. When specifying a series of jobs and job nets that have an input / output relationship with each other between jobs, the series of job nets is specified based on the connection relationship via the prenode and the postnode.
The failure impact investigation device according to claim 1. The arithmetic unit
In the process of specifying the fault influence range, based on the user's specification, starting from the job that was the target of the abend, forward search toward the upstream side in the input / output relationship and toward the downstream side in the input / output relationship. One of the backward searches is performed to identify the job and the sequence of the job nets.
The fault impact investigating apparatus according to claim 2. As information on the hierarchical structure of a batch job including at least each layer of a job related to each job and a job net composed of a series of the above jobs , the processing status or processing result of one of the jobs or the job nets can be used. An information processing device including a storage device that stores at least the information of the connection rule that defines the other standby or execution trigger that responds and the operation log of the job or the job net.
Based on the input / output relationship between jobs and job nets, which is indicated by the information on the hierarchical structure, the input / output relationship is established between the jobs and the jobs starting from the job for which an abend occurs in a predetermined job. Processing that identifies a series of jobs and job nets that it has as a failure impact range,
A directed graph or undirected graph in which at least each of the jobs and job nets belonging to the failure influence range is a node and each of the nodes is connected by an edge, a list of jobs and job nets belonging to the failure influence range, and the failure. A process that generates at least one of the graphs that aggregates and displays the number of jobs or job nets that belong to the affected area for each business unit and outputs it to a predetermined device .
In the failure impact range, the operation log is defined as a job and a job net that will be included in the sequence with the passage of time with the time of occurrence of the abend as the reference time, starting from the job that was the target of the abend. The operation timing indicated by is included between the reference time and the user-designated time, and the operation is presumed according to the standby or execution trigger indicated by the information of the connection regulation, and the identification is specified. A process of generating at least one of the directed graph, the undirected graph, the list, and the graph and outputting the job and the job net to a predetermined device.
A failure impact investigation method characterized by the execution of. The information processing device
In the storage device, as information related to the hierarchical structure, a prenode corresponding to a virtual job that receives data input from another job net in the job net, and a virtual that outputs data to the other job net in the job net. It holds each information of the post node corresponding to the specific job, and further stores the connection relationship information in which each of the prenode and the post node and all the jobs constituting the job net have an input / output relationship. ,
In the process of specifying the failure impact range, based on at least the input / output relationship between jobs and job nets indicated by the information on the hierarchical structure, starting from the job that was the target of the abend in the predetermined job, within the job. And when specifying a series of jobs and job nets that have an input / output relationship with each other between jobs, the series between job nets is specified based on the connection relationship via the prenode and the postnode.
The method for investigating the effects of a failure according to claim 4, wherein the method is characterized by the above. The information processing device
In the process of specifying the fault influence range, based on the user's specification, starting from the job that was the target of the abend, forward search toward the upstream side in the input / output relationship and toward the downstream side in the input / output relationship. Perform one of the backward searches to identify the job and the sequence of job nets.
The failure impact investigation method according to claim 5, wherein the method is characterized by the above.

Images