JPS6151248A - Recovery processing system - Google Patents

Abstract

PURPOSE:To recover the corresponding data base or set by reading a log data from a log data file and the corresponding schema from a management information file, creating a control table, and executing an I/O processing with that table. CONSTITUTION:A log data input part 10 reads data from a history log file 6 and a backout file 7, and transfers it to a reload part 11 for a data set. This processing continus until all the data is input. The reload part 11 checks whethr the intercept from a input parr 10 is a termination intercept or not, and executes a termination processing if it is a termination intercept, otherwise searches the extension corresponding to the data intercepted. It checks whether the environment corresponding to the extension has been already created or not, and reads the corresponding schema from the management information file 1 to create the control table if the answer is No. It executes the I/O processing according to the control table, and recovers the specified extension using the log data.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention reads a schema from a management file based on control information on log data that is manually entered at the time of data input, and performs processing based on the read schema. This invention relates to a recovery processing method for restoring a database or data set based on log data by creating a control table and performing input/output processing according to the created control table.

[Conventional technology and problems]

Figure 2 explains the conventional recovery processing method from system failure, in which 1 is a management information file, 2
is the preparatory processing section, 3 is the schema on C3A, 4 is the job of the recovery main body, 5 is various control tables on the main unit, 6 is the history log file, 7 is the bankout file, 8
is a database or data set, 9 is an initial processing unit, 1
0 indicates a log data input section, 11 indicates a data set restoration section, and 12 indicates a post-processing section. A plurality of schemas are registered in the management information file 1. A schema is a member that stores definition information and operational information for a database or dataset. The pre-preparation processing unit 2 is a TRCV
It operates when the ()Ranxion/Recovery) command is input, and performs processing such as expanding the schema of the management file 1 onto the C3A. C3A means virtual space intercommunication area. The job 4 of the recovery main body is composed of an initial processing section 9, a log data input section 1o, a database restoration section 11, and a post-processing section 12. The initial processing unit 9 creates an extent control table 5 on the region based on the schema on the C3A. A region is an area where an application program is loaded and executed. The log data input unit 10 is a READ task that reads the contents of the history log file 6 and bankout file 7. The history log file 6 is for holding updated data, and the bankout file 7 is for holding pre-updated data.

The data set restoration section 11 is a wR*rE task, which restores the corresponding extent using the log data passed from the log data input section 10.

FIG. 3 shows the conventional processing of the preparatory processing section 2.

Read the schema from AIM directory 1 (synonymous with management information file), check whether the corresponding database or dataset is in 0PEN state, and if Yes, expand the read schema on C3A and run TRCV (+
- Ransomware/Recovery) Attach target indication. The above process is performed for all schemas in the AIM directory 1, and then the TRCV job is started.

FIG. 4 shows the conventional processing of the initial processing section 9. C
Based on the schema on 3A, various control tables (for example, DECB, DCB, etc.) are created for each extent. The above processing is performed for all schemas on C3A.

FIG. 5 is a diagram showing conventional processing of the log data input processing section 1o. Data is read from the history log data 6 and bankout file 7, and the data set restoration section (W
RITII! Task) Pass to 11. Repeat the above process until all data has been input, and then click 11? Notify the ITE task of completion.

FIG. 6 is a diagram showing conventional processing by the data set restoration unit 11. It is checked whether it is an end notification (END notification), and if NO, it searches for the extent corresponding to the notified data, restores the data in the 8th extent, and waits for the next notification. Upon receiving the termination notification, perform termination processing.

In the conventional recovery processing methods shown in Figures 2 to 6, all schemas in the system are read from the management information file, so the processing time increases as the system scale increases.C3A Disadvantages include the fact that C3A becomes larger because the schema that was in use immediately before the downtime is expanded above, and the region becomes larger because control tables corresponding to all schemas on C3A are expanded onto the region. have.

[Purpose of the invention]

SUMMARY OF THE INVENTION The present invention aims to provide a recovery processing method that eliminates the above-mentioned drawbacks, shortens the time required to start up a system after it is down, and reduces the amount of memory required. There is.

[Means to achieve the purpose]

Therefore, the recovery processing method of the present invention requires a database or a dataset, a management information file that stores a schema for the database or dataset, and a control that indicates which database or dataset it is for. The system includes a log data file for storing log data to which information has been added, and a recovery means, and the recovery means reads log data from the log data file and restores the read log data. The following log data is read by referring to the control information, reading the corresponding schema from the management information file, creating a control table based on the read schema, and performing input/output processing according to the created control table. The invention is characterized in that it is configured to restore the corresponding database or data cenode.

[Embodiments of the invention]

Embodiments of the present invention will be described below with reference to the drawings.

In summary, the recovery processing method of the present invention dynamically collects various definition information and allocation information based on control information on log data that is manually input when log data is input, rather than at the first time. It is something.

The functional blocks of the system for implementing the present invention are the second
Although similar to the figure, in the present invention, the preparatory processing unit 2
performs only TRCV job startup, the initial processing unit 9 does not create a control table for each extent, and only creates an environment for inputting log data, and the log data input unit 10 is basically the same as before. Perform processing.

Currency ``Figure'' *Shizan-9i y ``Shagumi''
It is a diagram showing the struggle. The data set restoration unit 11 performs the following processing. It is checked whether the notification from the log data input unit 10 is a termination notification (accumulated notification) or not, and if it is a termination notification, the termination process is performed, and if it is not a termination notification, the extent corresponding to the notified data is checked. Search for. Then, it is checked whether the environment for the corresponding extent has been created, and if NO, the corresponding schema is read from the management information file 1 and a control table is created. Note that the control table is queued in a recovery vector table (not shown). Furthermore, control information indicating what database or data set the log data corresponds to is added. Once the control table is created, input/output processing is performed according to this control table, the corresponding extent is restored using log data, and the next notification is waited for.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, a schema corresponding to log data is read when log data appears, so the amount of schema to be read can be minimized, and processing time can be shortened. Furthermore, since manual input of log data and reading of the schema can be performed in parallel, processing time can also be reduced. Furthermore, according to the present invention, it is possible to read one schema onto C3A, so C8
The space used on A can be reduced, and the control tables deployed on the region can be limited to those for the database or dataset currently being restored, so the space used on the region can also be reduced. I can do it.

[Brief explanation of drawings]

Fig. 1 is a diagram showing the processing of the data set restoration section in the present invention, Fig. 2 is a diagram explaining the conventional recovery processing method from system failure, and Fig. 3 is a diagram showing the conventional processing of the preparatory processing section. 4 is a diagram showing conventional processing of the initial processing section, FIG. 5 is a diagram showing conventional processing of the log data input processing section, and FIG. 6 is a diagram showing conventional processing of the data set restoration section. . 1... Management information file, 2... Preparation processing section, 3
... Schema of C3A, 4... Job of recovery main body, 5... Various control tables on the region, 6... History log file, 7... Backout file,
8... Database or data set, 9... Initial processing section, 10... Log/data input section, 11... Data set restoration section, 12... Post processing section.

Claims (1)

[Claims] A management information file that stores a database or dataset, a schema for the database or dataset, and log data that stores log data with control information indicating which database or dataset it is for.・It is equipped with a file and a recovery means, and the recovery means is configured to recover the above log data.
Read log data from a file, refer to the control information of the read log data, read the corresponding schema from the management information file, create a control table based on the read schema, and according to the created control table. A recovery processing method characterized by being configured to restore a corresponding database or data set using log data read by performing input/output processing.