24
DBMS_LOGMNR

LogMiner allows you to make queries based on actual data values. For instance, you could issue a query to select all updates to the table scott.emp or all deletions performed by user scott. You could also perform a query to show all updates to scott.emp that increased sal more than a certain amount. Such data can be used to analyze system behavior and to perform auditing tasks.

The DBMS_LOGMNR package contains procedures used to initialize the LogMiner tool. You use these procedures to list the redo log files to be analyzed and to specify the SCN or time range of interest. After these procedures complete, the server is ready to process SQL SELECT statements against the V$LOGMNR_CONTENTS view.

The data in redo log files is especially important for recovery, because you can use it to pinpoint when a database became corrupted.You can then use this information to recover the database to the state just prior to corruption.

This chapter discusses the following topics:

DBMS_LOGMNR Constants

The following sections describe the constants for the DBMS_LOGMNR package.

Constants for ADD_LOGFILE Options Flag

`NEW`	`DBMS_LOGMNR`.`NEW` purges the existing list of redo log files, if any. Places the specified redo log file in the list of log files to be analyzed.
`ADDFILE`	`DBMS_LOGMNR`.`ADDFILE` adds the specified redo log file to the list of log files to be analyzed. Any attempts to add a duplicate file raise an exception (ORA-1289).
`REMOVEFILE`	`DBMS_LOGMNR`.`REMOVEFILE` removes the redo log file from the list of log files to be analyzed. Any attempts to remove a file that has not been previously added, raise an exception (ORA-1290).

Constants for START_LOGMNR Options Flag

`COMMITTED_DATA_ONLY`	`If set, only DMLs corresponding to committed transactions are returned. DMLs corresponding to a committed transaction are grouped together. Transactions are returned in their commit order. If this option is not set, all rows for all transactions (committed, rolled back and in-progress) are returned.`
`SKIP_CORRUPTION`	Directs a `SELECT` operation from `V$LOGMNR_CONTENTS` to skip any corruptions in the redo log file being analyzed and continue processing. This option works only when a block in the redo log file (and not the header of the redo log file) has been corrupted. Caller should check the `INFO` column in the `V$LOGMNR_CONTENTS` view to determine the corrupt blocks skipped by LogMiner.
`DDL_DICT_TRACKING`	If the dictionary in use is a flat file or in the redo log files, LogMiner ensures that its internal dictionary is updated if a DDL event occurs. This ensures that correct `SQL_REDO` and `SQL_UNDO` information is maintained for objects that are modified after the LogMiner dictionary is built. This option cannot be used in conjunction with the `DICT_FROM_ONLINE_CATALOG` option.
`NO_DICT_RESET_ONSELECT`	This option is only valid if the `DDL_DICT_TRACKING` option is also specified. It prevents LogMiner from reloading its internal dictionary at the beginning of each select operation on `V$LOGMNR_CONTENTS`. This can be an advantage because it can be time consuming to refresh the dictionary if a DDL operation has updated the dictionary. If you use this option, be aware that because the dictionary has not been refreshed for subsequent select operations, you may get incompletely reconstructed `SQL_REDO` and `SQL_UNDO` information for objects that are modified in the redo log files. Such incomplete reconstructions produce SQL that cannot be executed.
`DICT_FROM_ONLINE_CATALOG`	Directs LogMiner to use the current live database dictionary rather than a dictionary snapshot contained in a flat file or in a redo log file. This option cannot be used in conjunction with the `DDL_DICT_TRACKING` option.
`DICT_FROM_REDO_LOGS`	If set, LogMiner expects to find a dictionary in the redo log files that were specified with the `DBMS_LOGMNR.ADD_LOGFILE` procedure.

Extracting Data Values From Redo Log Files

LogMiner data extraction from redo log files is performed using two mine functions: DBMS_LOGMNR.MINE_VALUE and DBMS_LOGMNR.COLUMN_PRESENT, described later in this chapter.

Example of Using DBMS_LOGMNR

The following example shows how to use the DBMS_LOGMNR procedures to add redo log files to a LogMiner session, how to start LogMiner, how to perform a select operation from V$LOGMNR_CONTENTS, and how to end a LogMiner session. For complete descriptions of the DBMS_LOGMNR procedures, see Summary of DBMS_LOGMNR Subprograms .

EXECUTE DBMS_LOGMNR.ADD_LOGFILE( 
   LogFileName => '/oracle/logs/log1.f',
   Options => dbms_logmnr.NEW);

EXECUTE DBMS_LOGMNR.ADD_LOGFILE( 
   LogFileName => '/oracle/logs/log2.f', 
   Options => dbms_logmnr.ADDFILE);

EXECUTE DBMS_LOGMNR.START_LOGMNR(
   DictFileName =>'/oracle/dictionary.ora');

SELECT sql_redo 
FROM V$LOGMNR_CONTENTS
EXECUTE DBMS_LOGMNR.END_LOGMNR();

Summary of DBMS_LOGMNR Subprograms

Table 24-1 describes the procedures in the DBMS_LOGMNR supplied package.

Table 24-1 DBMS_LOGMNR Package Subprograms

Subprogram	Description
"ADD_LOGFILE Procedure"	Adds a file to the existing or newly created list of archive files to process.
"START_LOGMNR Procedure"	Initializes the LogMiner utility.
"END_LOGMNR Procedure"	Finishes a LogMiner session.
"MINE_VALUE Function"	This function may be called for any row returned from `V$LOGMNR_CONTENTS` to retrieve the undo or redo column value of the column specified by the `column_name` input parameter to this function.
"COLUMN_PRESENT Function"	This function may be called for any row returned from `V$LOGMNR_CONTENTS` to determine if undo or redo column values exist for the column specified by the `column_name` input parameter to this function.

24 DBMS_LOGMNR