JP2006190153A - Method and program for failure restoration - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To make processing for a business system normally function in a minimum restoration time in time of a failure of a computer wherein the business system operates, by a failure restoration method and a failure restoration program. <P>SOLUTION: In this method for failure restoration, a failure situation is confirmed when the failure occurs, a failure dissolution method for dissolving the failure is extracted on the basis of the confirmed failure situation, a processing time necessary to dissolve the failure is extracted in each the extracted failure dissolution method, and the extracted processing times are compared to determine the failure dissolution method. In the determination of the failure dissolution method, the failure dissolution method having the small extracted processing time is selected in each the extracted method. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a failure recovery method and a failure recovery program, and more particularly, to a failure recovery method and a failure recovery program capable of recovering a failure location according to a failure state of a computer.

  Conventionally, automatic recovery from a failure of application software by a computer system has been performed. For example, using the monitoring software, monitor the status of the running application software. If an error occurs, restart the application software, or restart the OS if it does not recover after a predetermined number of restarts. In other words, the entire system is recovered by executing a reboot of the computer (see Patent Document 1).

JP 2002-149437 A

  However, in the above prior art configuration, there is still a point to be improved in order to perform quicker and more reliable failure recovery with respect to computer failure recovery.

  In other words, if a failure occurs in the computer, it cannot be solved by restarting the application, restarting the OS, or resetting the hardware. Restoration using backup data may be necessary. It is desirable to apply to the place.

  In addition, since processing may be delayed due to a failure, it may be necessary to recover the failure by changing the processing path to a normal computer.

  Furthermore, it is necessary to consider the recovery time until normal processing is performed for “data recovery to the fault location” and “recovery by changing the processing path”. The actual situation was that the recovery method was selected accordingly.

  Therefore, there is a problem that the recovery of the failure may depend on the experience of the administrator, and the failure is not always solved in a short time.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to perform quicker and more reliable recovery from a failure in a computer.

  It is another object of the present invention to extract an optimal failure recovery method based on the failure status of a computer by constantly monitoring the operation of the computer.

  Another object of the present invention is to extract a recovery method with a short time by comparing the time required for eliminating the failure for each extracted method.

  Furthermore, the failure recovery method of the present invention applies backup processing to a failure location when a failure occurs in the computer and in addition to restarting the application, restarting the OS, or resetting the hardware, but still cannot be resolved. The purpose is to let you.

  Another object of the present invention is to enable recovery from a failure by changing a processing route to a normal computer when processing is delayed due to a failure.

  The failure recovery method of the present invention that solves the above problems is to check a failure status when a failure occurs, extract a failure elimination method for solving the failure based on the confirmed failure status, Extracting a processing time necessary for solving the failure for each failure solving method, and determining the failure solving method by comparing the extracted processing times.

  In the failure recovery method of the present invention, the failure resolution method may be determined so that the extracted processing time with a short processing time is selected for each extracted method. It is possible to include at least one selected from the group consisting of restarting an application in the computer, rebooting the computer, restoring data in the computer with backup data, and changing a processing path.

  In addition, the failure recovery program of the present invention that solves the above-mentioned problems is a function for checking a failure status in a computer when a failure occurs, and a function for extracting a failure resolution method for solving the failure based on the failure status. And a program that realizes a function of extracting a processing time necessary for solving the failure for each extracted method and a function of determining the failure solving method by comparing the extracted processing times. .

  In this program, the determination of the failure solving method can be set so that the extracted processing time with a short time is selected, and the failure solving method includes restarting the application in the computer, rebooting the computer, It may be possible to select from the group consisting of restoration of in-computer data by backup data and change of processing path.

  According to the present invention, the following effects can be obtained.

  It is possible to recover from a failure of the computer more quickly and reliably.

  When a failure occurs in the computer, in addition to restarting the application, restarting the OS, and resetting the hardware, it is possible to apply the restore processing from the backup data to the failure location.

  In addition, since the processing may be delayed due to a failure, a more reliable recovery is possible by attempting to recover the failure by changing the processing path to a normal computer.

  In addition, for “data recovery to the fault location” and “recovery by changing the processing path”, a recovery method with a short recovery time until normal processing is appropriately selected, so that quicker recovery is applied smoothly. be able to.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. The present invention is not limited to the embodiments described below, and can be appropriately modified and combined within the scope of the gist of the present invention.

  FIG. 1 is a schematic configuration diagram for explaining a public example of a computer system when a failure recovery method and a failure recovery program of the present invention are applied.

  In FIG. 1, 1 is a backup system having an operation monitoring function, 5 is a virtual monitoring program, 6 is a backup program, 7 is a failure recovery plan creation program, 8 is an information management table, 9 is backup data, 30 is the Internet or an intranet. , 52 is a computer, 60 is a business system, 61 is an application, and 62 is data.

  As shown in the figure, the business system 60 is usually composed of a plurality of computers 52, and the backup system 1 having an operation monitoring function includes a virtual monitoring program 5, a backup program 6, a failure recovery plan creation program 7, and PC management information. And an information management table 8 and backup data 9 including processing management information, backup information, failure information, influence range information and the like as necessary.

  The operation monitoring program 5 monitors the operating state of the computer 52 connected through the network 30 including the Internet or an intranet as an information communication line, the operating state of the application 61, and the operating state of the business system 60. In the present embodiment, the operation monitoring program 5 has a function of comparing the monitoring information with the failure information in the information management table 8.

  The backup program 6 periodically acquires backup data of the computer 52, application 61 and data 62, stores it as backup data 9, stores backup information in the information management table 8, and has a restore function. By using the same information management table for the operation monitoring result and the backup result, it is possible to save the redundant management effort. Further, when an abnormality is detected as a result of the operation monitoring, the backup data is restored and restored, so that it is possible to save manpower necessary for the failure recovery.

  FIG. 2 shows a preferred example of a table stored in the information management table.

  The process management master table 150 is used to clarify the relationship with the name of the application necessary for the business process contents. For example, Process No. S-001 is AAA-APP1, BBB-APP1, Process No. S-002 is stored as AAA-APP1 and AAA-APP2. As shown in the figure, by assigning the computer name to the beginning of the application name described in each processing item, it is possible to associate which application of which computer is required for which business processing. Convenient not only in time but also in daily management.

  In the backup management master table 200, the names of the computers that have been backed up regularly (in the figure, indicated as AAA or BBB in the PC column) and OS (operation system) information on the computer (in the OS column in the figure). AAA-OS, BBB-OS), application information (in the figure, AAA-APP1, APP2, BBB-APP1, APP2 in the application column), data information (in the figure, in the data column) AAA-DATA1 and BBB-DATA1) are stored.

  These data are used as restoration of backup data when a failure occurs.

  The backup management master table 200 can be used to manage which data corresponds to which PC and application. When a failure occurs, necessary processing is performed based on this information.

  In the failure information table 250, the name of the application (indicated as AAA-APP1 and AAA-APP2 in the figure) and the name of the data with respect to the failure content that may occur on the computer. 270 (denoted as AAA-DATA1 in the figure), recovery procedure (denoted as RD-001 or RD-002 in the figure), and restoration time 260 (60 minutes or 5 minutes) is stored.

  As a result of the operation monitoring, if the information matches the information described in “PC” 251, “Fault 1” 252, or “Fault 2” 253 in the fault information table, it is determined as a fault and the fault information is identified. The From the failure information table 250, an application, a recovery method, and a recovery time that are affected when a failure is detected are specified.

  The influence range information table 300 is provided to indicate an influence range in which other normal computer processing stops for the failure information (D-001, D-002). Here, the processing contents of the processing management master table 150 affected by the failure information, the recovery procedure, and the recovery time 310 due to the processing path change are stored.

  In other words, if the failure information D-001 is determined, the influence range is found to be the influence range E-001. It can be seen from the influence range information table 300 that S-001 and S-002 are affected. Further, it can be seen from the influence range information table 300 that the procedure for recovering the affected business process by changing the processing path is the recovery procedure RD-001, and the recovery time in that case is 5 minutes.

  As described above, the failure information is specified from the result of the operation monitoring, whereby the influence range of the influence range table is specified, and the process that stops at the failure point is specified.

  In the recovery plan table 400, the failure information (D-001, D-002) and the affected range (E-001, E-002) are associated with the recovery plan (Y-001, Y-002). Stored.

  This makes it possible to compare the recovery time by data restoration to the fault location stored in the fault information table 250 and the recovery time by change of the processing path stored in the affected range information table using the fault information as a trigger. ing.

  FIG. 3 shows an abnormality detected after the operation monitoring of the backup system 1 having the operation monitoring function to the computer 52 of the business system 60 and the application 61 operating on the computer is performed, and the failure location and the affected range are identified. The flowchart of the rough process until failure recovery is shown. Next, the flowchart of FIG. 3 will be described.

  Operation monitoring by an operation monitoring program is performed on a business system composed of a plurality of computers, and failure information is searched from the failure information table (step 500). Here, if the operation information does not match “PC”, “failure 1”, or “failure 2” in the failure information table, it is regarded as normal operation and it is determined that there is no abnormality. If the failure information matches, an abnormality is detected. “Failure 1” and “Failure 2” describe the type of the failure caused by the PC, and if the failure information matches even one of them, the failure is detected.

  When the operation monitoring program searches the failure information table and the operation monitoring result matches “PC”, “failure 1”, or “failure 2”, the failure information table is searched for the failure target application and recovery data. The failure information and the recovery procedure are specified (step 510).

  Next, the influence range is searched from the influence range information table, and the influence range is specified from the influence range information table based on the failure information (step 520).

  Next, compare the recovery time from the restoration obtained from the failure information table with the recovery time from the processing range change obtained from the affected range table, and compare the "recovery time by restoration" with the "recovery time by change of the processing route". If it is short, the answer is Yes. That is, here, it is specified whether “restoring the backup data to the failure location” or “changing the processing path to a location other than the failure location” is applied as the recovery method with a short recovery time (step 520). Therefore, the comparison between large and small may be reversed, in which case Yes and No are interchanged.

  That is, in this process, it is only necessary to select and determine one with a short processing time. When the processing time is the same, it may be determined according to the character and situation required for the related business process. For example, for failures in processing that prioritize data consistency, recovery is performed by restoring backup data. As for the failure of the processing giving priority to the linkage speed to the next processing, a method of determination such as recovery by changing the processing path to a location other than the failure location can be considered.

  If the processing method is determined by comparing the recovery times, the backup data is restored to the failed location (step 540) or the processing path change processing to a location other than the failed location (step 550) is performed.

  FIG. 4 shows a flowchart of a rough process until the operation monitoring program generates a computer operation result as log (LOG) information and searches from the failure information.

  Next, the flowchart of FIG. 4 will be described.

  When the monitoring target computer is registered in the monitoring program (step 600), the monitoring target computer is polled (step 610), and text data to be LOG information is generated as the monitoring result (step 620).

  If there is no response as a result of polling, it is determined that there is an abnormality at this point.

  Next, the failure information in the failure information table is searched with the generated text data that is the monitoring result of the computer, and “PC”, “failure 1”, or “failure 2” is compared (step 630) and matched. In case of failure detection. If an abnormality is detected, the process proceeds to a recovery procedure.

  Therefore, if it does not match the item in the failure information, it is determined that there is no more and the process returns to step 610.

  FIG. 5 shows a rough procedure from failure detection to restoration by restoration of backup data. Next, the flowchart of FIG. 5 will be described.

  As described in the flowchart of FIG. 4, the failure recovery program searches for failure information in the failure information table using text data as a monitoring result and detects a failure (* 1 in FIG. 4 and * in FIG. 5). 1), in order to specify the backup data from the search result, the backup data of the backup management master table is specified in accordance with the recovery procedure in the failure information table (step 710).

  Next, in order to designate the failure target from the search result, the failure target is specified from the failure information table (step 720).

  In this flowchart, in order to instruct the failure recovery program to restore from the restoration procedure, the restoration procedure of the failure information table is instructed to be restored (step 730).

  The failure recovery program restores the backup data to the failure target (step 740).

  In addition, when “change processing route to other than faulty location” is selected, processing without affecting the faulty location is performed by changing the processing route to other than the faulty location according to the recovery procedure in the influence range information table 300. Flows.

  A flowchart in this case is shown in FIG. Here, the same step number as in FIG. 5 is for the same residence as that described in FIG. Only the procedure part for changing the processing path will be described. FIG. 6 shows a rough procedure from failure detection to recovery by changing the processing path. Next, the flowchart of FIG. 6 will be described.

  As described in the flowchart of FIG. 4, the failure recovery program searches for failure information in the failure information table using text data that is a polling monitoring result, and a failure is detected (* 2 in FIG. 4 and * in FIG. 6). 2 corresponds). Next, in order to specify the failure target server from the search result, the failure target server is identified from the failure information table (step 820). In this flowchart, the failure recovery program instructs the processing path change from the stop and recovery procedure to the failure target server (step 830).

  The failure recovery program changes the route to other than the failure location (step 840).

  Fault recovery by processing path change and restoration according to the present invention is performed by specifying fault information and an affected range from computer monitoring results such as polling to the monitoring target computer and monitoring result LOG in FIG. is doing. Further, according to the present invention, the failure recovery processing may be performed by combining the processing route change and the restore processing depending on how the failure information is set, instead of selecting either the processing route change or restoration.

  In the above embodiment, the comparison of the recovery time for determining the failure recovery is performed only for one failure. However, when a plurality of failures occur, what kind of failure recovery processing is performed depending on the total value of the failures. It goes without saying that it is possible to determine whether to do.

1 is a configuration diagram of a “backup system having an operation monitoring function”. FIG. It is a detailed view of an information management table. It is a recovery process flowchart from the abnormality detection by operation monitoring. It is an operation | movement monitoring process flowchart. It is a restore process flowchart. It is a processing path change flowchart.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Backup system provided with operation monitoring function 5 Operation monitoring program 6 Backup program 7 Fault recovery plan creation program 8 Information management table 9 Backup data 30 Internet / Intranet 52 Monitored computer 60 Business system 61 Application 62 Data 150 Processing management master table 200 Backup management master table 250 Failure information table 251 PC
252 Disability 1
253 Disability 2
300 Influence range information table 500 Operation monitoring 510 Retrieval of failure target and recovery data from failure information table 520 Retrieval of influence range from influence range information table 530 Comparison of recovery time 540 Restoration of backup data to failure location 550 Rest of failure location Processing path change 600 Registering the monitoring target computer in the monitoring program 610 Polling the monitoring target computer 620 Generating text data to be LOG information as a monitoring result 630 Searching for fault information using text data 710 Specifying backup data 720 Specifying a fault target 730 Restore instruction from recovery procedure 740 Restore backup data to failure control 810 Search failure information by text 820 Specify failure target server 830 In disaster recovery program Route change to other than the failure point instructions 840 processing of the processing path change process Ri

Claims (6)

  To confirm a failure status when a failure occurs, to extract a failure solving method for solving the failure based on the confirmed failure status, and to eliminate the failure for each of the extracted failure solving methods A failure recovery method characterized by extracting a necessary processing time and comparing the extracted processing times to determine the failure elimination method.   2. The failure recovery method according to claim 1, wherein the failure resolution method is determined by selecting a method having a short extracted processing time for each of the extracted methods.   2. The failure solving method includes at least one selected from the group consisting of restarting an application in a computer, rebooting a computer, restoring data in the computer with backup data, and changing a processing path. Or the failure recovery method of 2.   A function for checking a failure status when a failure occurs in a computer, a function for extracting a failure solving method for solving the failure based on the failure status, and for solving the failure for each extracted method A failure recovery program for realizing a function of extracting a necessary processing time and a function of comparing the extracted processing times and determining the failure solving method.   The failure recovery program according to claim 4, wherein the determination of the failure elimination method can be set so that the extracted processing time with a short time is selected.   6. The fault recovery program according to claim 4 or 5, wherein the fault elimination method can be selected from the group consisting of restarting an application in the computer, rebooting the computer, restoring data in the computer using backup data, and changing a processing path. .

Images