========================RMAN数据库全库备份=========================================

[oracle@localhost ~]$ rman target /
RMAN> backup incremental level =0 database format='/home/oracle/db_bak/%u';
Starting backup at 17-MAR-11
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=146 devtype=DISK
channel ORA_DISK_1: starting incremental level 0 datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
input datafile fno=00001 name=/oradata/TEST/system01.dbf
input datafile fno=00003 name=/oradata/TEST/sysaux01.dbf
input datafile fno=00005 name=/oradata/TEST/t_recover.dbf
input datafile fno=00002 name=/oradata/TEST/undotbs01.dbf
input datafile fno=00004 name=/oradata/TEST/users01.dbf
channel ORA_DISK_1: finished piece 1 at 17-MAR-11
piece handle=/home/oracle/db_bak/02m7e1h1 tag=TAG20110317T053737 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:01:55
channel ORA_DISK_1: starting incremental level 0 datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
including current control file in backupset
including current SPFILE in backupset
channel ORA_DISK_1: starting piece 1 at 17-MAR-11
channel ORA_DISK_1: finished piece 1 at 17-MAR-11
piece handle=/home/oracle/db_bak/03m7e1kl tag=TAG20110317T053737 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:00:03
Finished backup at 17-MAR-11
或者使用脚本
#!/bin/bash
source ~/.bash_profile
$ORACLE_HOME/bin/rman target / <<EOF
run {
        allocate channel C1 type disk FORMAT '/home/oracle/db_bak/%u';
        allocate channel C2 type disk FORMAT '/home/oracle/db_bak/%u';
        backup incremental level=0 database format='/home/oracle/db_bak/%u';
     }
exit;
EOF
========================RMAN数据库增量备份=========================================
上一步进行了数据库0级的全备，我们接下来进行数据库的增量备份
RMAN> backup incremental level =1 database format='/home/oracle/db_bak/%u';
Starting backup at 17-MAR-11
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=154 devtype=DISK
channel ORA_DISK_1: starting incremental level 1 datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
input datafile fno=00001 name=/oradata/TEST/system01.dbf
input datafile fno=00003 name=/oradata/TEST/sysaux01.dbf
input datafile fno=00005 name=/oradata/TEST/t_recover.dbf
input datafile fno=00002 name=/oradata/TEST/undotbs01.dbf
input datafile fno=00004 name=/oradata/TEST/users01.dbf
channel ORA_DISK_1: starting piece 1 at 17-MAR-11
channel ORA_DISK_1: finished piece 1 at 17-MAR-11
piece handle=/home/oracle/db_bak/04m7ed0a tag=TAG20110317T085329 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:01:25
channel ORA_DISK_1: starting incremental level 1 datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
including current control file in backupset
including current SPFILE in backupset
channel ORA_DISK_1: starting piece 1 at 17-MAR-11
channel ORA_DISK_1: finished piece 1 at 17-MAR-11
piece handle=/home/oracle/db_bak/05m7ed30 tag=TAG20110317T085329 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:00:03
Finished backup at 17-MAR-11
========================数据库重建=========================================
将原来出现问题的数据库删除，重新创建一个数据库
①进入/oradata目录下，将所有的目录和文件删除
②将/etc下的oratab中的最后一句“TEST:/oracle/app/oracle/product/10.2.0/db 1:Y”删除，否则在接下来创建TEST数据库时会报错
③然后运行dbca进行数据库的创建
========================不完全恢复数据库=========================================
SQL>alter session set nls_date_format='yyyy-mm-dd hh24:mi:ss';
SQL>select sysdate from dual;
①将数据库设置为归档模式
        $mkdir -p /home/oracle/archiver01
        $mkdir -p /home/oracle/archiver02
        SQL> shutdown immediate
        SQL> startup mount
        SQL> alter database archivelog;
        SQL> alter system set log_archive_dest_1='location=/home/oracle/archive01' scope=both;
        SQL> alter system set log_archive_dest_2='location=/home/oracle/archive02' scope=both;
        SQL> show parameter archive;
        SQL> alter database open;
②对数据库进行全备
        $ rman target /
        RMAN> backup incremental level =0 database format='/home/oracle/db_bak/%u';
③删除一个表用于接下来的恢复测试
        SQL> drop table emp;
④使用日志挖掘，确定表被误删除的时间
        $mkdir -p /home/oracle/logmin
        SQL> alter system set utl_file_dir='/home/oracle/logmin' scope=spfile;
        SQL> shutdown immediate
        SQL> startup
        SQL> execute dbms_logmnr_d.build('shwdict.ora','/home/oracle/logmin');
        SQL> execute dbms_logmnr.add_logfile('/home/oracle/archive01/1_3_746014228.dbf',dbms_logmnr.new);
        SQL> exec dbms_logmnr.start_logmnr(dictfilename=>'/home/oracle/logmin/shwdict.ora');
        SQL> alter session set nls_date_format='yyyy-mm-dd hh24:mi:ss';
        SQL> select SCN,TIMESTAMP,SQL_REDO,SQL_UNDO from V$LOGMNR_CONTENTS where SQL_REDO like'%drop%';
        记录下误删除表的时间2011-03-17 10:50:51
⑤进行RMAN恢复
        SQL>shutdown immediate
        SQL>startup mount
        RMAN>run{
                set until time "to_date('2011-03-17 10:50:50','yyyy-mm-dd hh24:mi:ss')";
                restore database;
                recover database;
                alter database open resetlogs;
                }
⑥重新备份全部的数据库
        RMAN> delete noprompt backup;
        RMAN> backup database format='/home/oracle/db_bak/%u';
========================参数文件、数据文件、控制文件和日志文件丢失后数据库恢复============================
注：在恢复数据库之前，要在RMAN里查看参数文件和控制文件（RMAN>list backup）的备份文件在备份目录是哪个。
①创建最简初始化参数文件
        $echo "db_name=TEST">initnew.ora
②使用最简参数文件启动数据库到nomount状态
        SQL> startup nomount pfile=/oracle/app/oracle/product/10.2.0/db_1/dbs/initnew.ora;
③恢复参数文件
        重新连接到RMAN
        RMAN> restore spfile from '/home/oracle/db_bak/04m7elfd';
④使用数据库参数文件启动到数据库nomount状态进行恢复控制文件
        SQL> startup nomount force
        RMAN> restore controlfile from '/home/oracle/db_bak/04m7elfd';
⑤启动数据库到mount状态进行数据库的恢复
        RMAN> alter database mount;
        RMAN> restore database;执行这一步会报错，因为日志文件都被删除所以无法恢复，但虽然有错误信息不过没关系，因为不执行这一步的话就无法继续往下执行打开数据库
        RMAN> alter database open resetlogs;到此数据库已恢复
⑥重新备份全部的数据库
        RMAN> delete noprompt backup;
        RMAN> backup database format='/home/oracle/db_bak/%u';
========================数据库备份恢复测试============================
为了保证备份文件可以正常使用，要经常测试备份文件是否完好，但运行的数据库又不能进行数据库的恢复（因为rman是在数据库mount状态下进行数据恢复的），所以用到了恢复测试的方法，即在数据库打开的状态下进行测试：
        RMAN> restore database validate;
        Starting restore at 17-MAR-11
        using channel ORA_DISK_1
        channel ORA_DISK_1: starting validation of datafile backupset
        channel ORA_DISK_1: reading from backup piece /home/oracle/db_bak/06m7enaq
        channel ORA_DISK_1: restored backup piece 1
        piece handle=/home/oracle/db_bak/06m7enaq tag=TAG20110317T114946
        channel ORA_DISK_1: validation complete, elapsed time: 00:00:36
        Finished restore at 17-MAR-11
        注：此命令测试备份文件是否完好
        RMAN> report need backup;
        注：此命令查看那些文件需要备份

========================RMAN数据库全库备份=========================================

[oracle@localhost ~]$ rman target /
RMAN> backup incremental level =0 database format='/home/oracle/db_bak/%u';
Starting backup at 17-MAR-11
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=146 devtype=DISK
channel ORA_DISK_1: starting incremental level 0 datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
input datafile fno=00001 name=/oradata/TEST/system01.dbf
input datafile fno=00003 name=/oradata/TEST/sysaux01.dbf
input datafile fno=00005 name=/oradata/TEST/t_recover.dbf
input datafile fno=00002 name=/oradata/TEST/undotbs01.dbf
input datafile fno=00004 name=/oradata/TEST/users01.dbf
channel ORA_DISK_1: finished piece 1 at 17-MAR-11
piece handle=/home/oracle/db_bak/02m7e1h1 tag=TAG20110317T053737 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:01:55
channel ORA_DISK_1: starting incremental level 0 datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
including current control file in backupset
including current SPFILE in backupset
channel ORA_DISK_1: starting piece 1 at 17-MAR-11
channel ORA_DISK_1: finished piece 1 at 17-MAR-11
piece handle=/home/oracle/db_bak/03m7e1kl tag=TAG20110317T053737 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:00:03
Finished backup at 17-MAR-11
或者使用脚本
#!/bin/bash
source ~/.bash_profile
$ORACLE_HOME/bin/rman target / <<EOF
run {
        allocate channel C1 type disk FORMAT '/home/oracle/db_bak/%u';
        allocate channel C2 type disk FORMAT '/home/oracle/db_bak/%u';
        backup incremental level=0 database format='/home/oracle/db_bak/%u';
     }
exit;
EOF
========================RMAN数据库增量备份=========================================
上一步进行了数据库0级的全备，我们接下来进行数据库的增量备份
RMAN> backup incremental level =1 database format='/home/oracle/db_bak/%u';
Starting backup at 17-MAR-11
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=154 devtype=DISK
channel ORA_DISK_1: starting incremental level 1 datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
input datafile fno=00001 name=/oradata/TEST/system01.dbf
input datafile fno=00003 name=/oradata/TEST/sysaux01.dbf
input datafile fno=00005 name=/oradata/TEST/t_recover.dbf
input datafile fno=00002 name=/oradata/TEST/undotbs01.dbf
input datafile fno=00004 name=/oradata/TEST/users01.dbf
channel ORA_DISK_1: starting piece 1 at 17-MAR-11
channel ORA_DISK_1: finished piece 1 at 17-MAR-11
piece handle=/home/oracle/db_bak/04m7ed0a tag=TAG20110317T085329 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:01:25
channel ORA_DISK_1: starting incremental level 1 datafile backupset
channel ORA_DISK_1: specifying datafile(s) in backupset
including current control file in backupset
including current SPFILE in backupset
channel ORA_DISK_1: starting piece 1 at 17-MAR-11
channel ORA_DISK_1: finished piece 1 at 17-MAR-11
piece handle=/home/oracle/db_bak/05m7ed30 tag=TAG20110317T085329 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:00:03
Finished backup at 17-MAR-11
========================数据库重建=========================================
将原来出现问题的数据库删除，重新创建一个数据库
①进入/oradata目录下，将所有的目录和文件删除
②将/etc下的oratab中的最后一句“TEST:/oracle/app/oracle/product/10.2.0/db 1:Y”删除，否则在接下来创建TEST数据库时会报错
③然后运行dbca进行数据库的创建
========================不完全恢复数据库=========================================
SQL>alter session set nls_date_format='yyyy-mm-dd hh24:mi:ss';
SQL>select sysdate from dual;
①将数据库设置为归档模式
        $mkdir -p /home/oracle/archiver01
        $mkdir -p /home/oracle/archiver02
        SQL> shutdown immediate
        SQL> startup mount
        SQL> alter database archivelog;
        SQL> alter system set log_archive_dest_1='location=/home/oracle/archive01' scope=both;
        SQL> alter system set log_archive_dest_2='location=/home/oracle/archive02' scope=both;
        SQL> show parameter archive;
        SQL> alter database open;
②对数据库进行全备
        $ rman target /
        RMAN> backup incremental level =0 database format='/home/oracle/db_bak/%u';
③删除一个表用于接下来的恢复测试
        SQL> drop table emp;
④使用日志挖掘，确定表被误删除的时间
        $mkdir -p /home/oracle/logmin
        SQL> alter system set utl_file_dir='/home/oracle/logmin' scope=spfile;
        SQL> shutdown immediate
        SQL> startup
        SQL> execute dbms_logmnr_d.build('shwdict.ora','/home/oracle/logmin');
        SQL> execute dbms_logmnr.add_logfile('/home/oracle/archive01/1_3_746014228.dbf',dbms_logmnr.new);
        SQL> exec dbms_logmnr.start_logmnr(dictfilename=>'/home/oracle/logmin/shwdict.ora');
        SQL> alter session set nls_date_format='yyyy-mm-dd hh24:mi:ss';
        SQL> select SCN,TIMESTAMP,SQL_REDO,SQL_UNDO from V$LOGMNR_CONTENTS where SQL_REDO like'%drop%';
        记录下误删除表的时间2011-03-17 10:50:51
⑤进行RMAN恢复
        SQL>shutdown immediate
        SQL>startup mount
        RMAN>run{
                set until time "to_date('2011-03-17 10:50:50','yyyy-mm-dd hh24:mi:ss')";
                restore database;
                recover database;
                alter database open resetlogs;
                }
⑥重新备份全部的数据库
        RMAN> delete noprompt backup;
        RMAN> backup database format='/home/oracle/db_bak/%u';
========================参数文件、数据文件、控制文件和日志文件丢失后数据库恢复============================
注：在恢复数据库之前，要在RMAN里查看参数文件和控制文件（RMAN>list backup）的备份文件在备份目录是哪个。
①创建最简初始化参数文件
        $echo "db_name=TEST">initnew.ora
②使用最简参数文件启动数据库到nomount状态
        SQL> startup nomount pfile=/oracle/app/oracle/product/10.2.0/db_1/dbs/initnew.ora;
③恢复参数文件
        重新连接到RMAN
        RMAN> restore spfile from '/home/oracle/db_bak/04m7elfd';
④使用数据库参数文件启动到数据库nomount状态进行恢复控制文件
        SQL> startup nomount force
        RMAN> restore controlfile from '/home/oracle/db_bak/04m7elfd';
⑤启动数据库到mount状态进行数据库的恢复
        RMAN> alter database mount;
        RMAN> restore database;执行这一步会报错，因为日志文件都被删除所以无法恢复，但虽然有错误信息不过没关系，因为不执行这一步的话就无法继续往下执行打开数据库
        RMAN> alter database open resetlogs;到此数据库已恢复
⑥重新备份全部的数据库
        RMAN> delete noprompt backup;
        RMAN> backup database format='/home/oracle/db_bak/%u';
========================数据库备份恢复测试============================
为了保证备份文件可以正常使用，要经常测试备份文件是否完好，但运行的数据库又不能进行数据库的恢复（因为rman是在数据库mount状态下进行数据恢复的），所以用到了恢复测试的方法，即在数据库打开的状态下进行测试：
        RMAN> restore database validate;
        Starting restore at 17-MAR-11
        using channel ORA_DISK_1
        channel ORA_DISK_1: starting validation of datafile backupset
        channel ORA_DISK_1: reading from backup piece /home/oracle/db_bak/06m7enaq
        channel ORA_DISK_1: restored backup piece 1
        piece handle=/home/oracle/db_bak/06m7enaq tag=TAG20110317T114946
        channel ORA_DISK_1: validation complete, elapsed time: 00:00:36
        Finished restore at 17-MAR-11
        注：此命令测试备份文件是否完好
        RMAN> report need backup;
        注：此命令查看那些文件需要备份