Databases Reference
In-Depth Information
If any of your archived redo log files or online redo log files are physically available, you can also view the
sequence number by dumping the file contents to a trace file with the dump logfile command as follows:
SQL> alter system dump logfile '<directory/log file>';
For example:
SQL> oradebug setmypid
SQL> alter system dump logfile
'/ora01/fra/O12C/archivelog/2012_08_08/o1_mf_1_10_825jth6l_.arc';
SQL> oradebug tracefile_name;
/ora01/app/oracle/diag/rdbms/o12c/o12c/trace/o12c_ora_15002.trc
You can search that trace file for the string Seq#, which shows the sequence number for the log file that you
dumped. Here's a sample of what you should see in the dump file:
descrip:"Thread 0001, Seq# 0000000010, SCN 0x000000102a7c-0x000000102a7f"
In this case, the sequence number is 10.
Performing Cancel-Based Recovery
Problem
You want to perform a cancel-based incomplete database recovery first using RMAN to restore the data files and then
using SQL*Plus to recover the data files.
Solution
The following example restores from the latest RMAN backup in preparation for an SQL*Plus cancel-based recovery:
$ rman target /
RMAN> startup mount;
RMAN> restore database; # restore database from last backup
Once your database is restored, you can start an SQL*Plus session and initiate a cancel-based recovery, as shown here:
$ sqlplus / as sysdba
SQL> recover database until cancel;
You will now be prompted by SQL*Plus to manually apply each archived redo log file. The following is the prompt
that you'll get for each log file:
Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
Hit the Return/Enter key until you arrive at the archived redo log file where you want to stop the recovery process.
When you want to stop the recovery process, type the CANCEL keyword, as shown:
CANCEL
