When using the database recovery utility (pdrstr) to restore a database to the most recent synchronization point, if you execute the pdrstr command with a directory specified in the -d option, you need not be concerned about the sequence of unload log files. However, if you execute the pdrstr command with a file name specified in the -l option, you must specify the unload log files in order of time (according to generation). This section explains how to identify the unload log files, which constitute the information that is input during database recovery.

In the example shown in the figure below, to restore the database to the most recent synchronization point, backup file 1 and unload log files 2 to 4 are used. In this process, unload log files 2, 3, and 4 must be specified in that order in the -l option of the pdrstr command.

Figure 3-10 Example in which an error has occurred

To identify the first unload log file (unload log file 2 in the example) to be specified for the -l option of the pdrstr command, collect the information during backup acquisition. The last unload log file (unload log file 4 in the example) to be specified for the -l option is the unload log file created during database recovery. Once the first and the last files have been identified, rearrange them in order of time and specify them in the -l option of the pdrstr command. The following figure shows the procedure for restoring a database by identifying the unload log files to be used during database recovery.

Figure 3-11 Identifying the unload log files to be used during database recovery

Organization of this section

3.11.1 Acquiring information on the first unload log file to be specified (during backup acquisition)

3.11.2 Acquiring information on the last unload log file to be specified (during database recovery)

3.11.3 Rearranging the unload log files (creating a time-ordered list)