DBA之RMAN备份

13：00

backup database

 

backup db :3h

3h: 产生了10 archive log file

 

16：00 finish

restore database;

13:00 : datafile不一致（因为没有备份archive log,controlfile)

 

job调用方式

vi /tmp/1.rcv

CONFIGURE CONTROLFILE AUTOBACKUP ON;

 

方式一  :   rman target / @'/tmp/1.rcv'

 

方式二 ：rman target / cmdfile=/tmp/1.rcv log=/tmp/rman.log

 

方式三  : sh script

内容如下

#!/bin/sh

rman target  log=/tmp/rman.log append <<EOF

CONFIGURE CONTROLFILE AUTOBACKUP ON;

EOF; 

 

---------------------------------------------------------------------------------------

备份策略

CONFIGURE RETENTION POLICY TO REDUNDANCY 1; # default

一般的我们这样设置

configure retention policy to recovery window of 30 days;

 

最简单的可恢复的备份

run{

     backup database including archivelog format '/backup/db_%d_%s_%p_%T.bkp';

     backup current controlfile format '/backup/cf_%d_%s_%p_%T.bkp';

}

 

标准备份

online 

run{

     allocate channel c1 type disk maxpiecesize=20G;

     allocate channel c2 type disk maxpiecesize=20G;

     sql 'alter system archive log current';     

     backup database format '/backup/db_%d_%s_%p_%T.bkp' tag 'db_backup_full';

     backup archivelog all delete input format '/backup/log_%d_%s_%p_%T.bkp' tag 'log_backup_full';  

     backup current controlfile format '/backup/cf_%d_%s_%p_%T.bkp' tag 'cf_backup_full';

     backup spfile format '/backup/pr_%d_%s_%p_%T.bkp' tag 'pr_backup_full';

}

备份顺序不能返（1.database->2.archivelog->3.controlfile->4.spfile)

 

 

offline

run{

     allocate channel c1 type disk maxpiecesize=20G;

     allocate channel c2 type disk maxpiecesize=20G;

     sql 'alter system archive log current';     

     backup database format '/backup/db_%d_%s_%p_%T.bkp' tag 'db_backup_full'; 

     backup current controlfile format '/backup/cf_%d_%s_%p_%T.bkp' tag 'cf_backup_full';

     backup spfile format '/backup/pr_%d_%s_%p_%T.bkp' tag 'pr_backup_full';

}

少了archivelog的备份

 

 

备份完后查看备份情况

list backup;

 

offline的恢复(冷备)

startup force nomount;

restore controlfile;

alter database mount;

restore database;

recover database noredo;

alter database open resetlogs;

 

------------------------------------------------------------------

report schema;列出当前数据库的文件分布

 

 

数据文件丢失恢复:

数据库mount模式下：

restore datafile 5;

recovery datafile 5;

alter database open;

 

控制文件丢失恢复：

shutdown abort;

startup mount;

show parameter control

cp file1 file2

 

ls -ltr

restore controlfile from autobackup;

或者 restore controlfile from '/back/....';

alter database mount;

recover database;

alter database open resetlogs;

 

1.good : backup database

2.file 5: restore recover

3.good

4.controlfile error

5.restore controlfile

 

--------------------------------------------------------------

删除备份

run{

     delete force noprompt backup;

     delete force noprompt copy;

     delete force noprompt archivelog all;

}

 

delete backupset 13(BS Key);

delete obsolete;

---

 

 

模拟数据库全库删掉整库恢复

查看数据库是否状态正常

select open_mode from v$database;

 

 

report schema;

rm -r /orcl/datafile

 

restore database;

恢复过程中查看IO

iostat 1

asm的好处数据库使用盘读写很均匀 普通文件系统没办法做到

recover database;通过归档日志做前滚

alter database open resetlogs;

 

数据库server 重启

srvctl stop database -d orcl

srvctl start database -d orcl

 

 

---

数据库恢复至少需要的文件

1. controlfile * 1
2. system *1

          sysaux * 1

1. redolog * 2

          undo * 1

 

将临时文件temp datafile删掉。 数据库能正常起来

这些场景会用到temp表空间order by , distinct , union, group by , join （PGA不够时会用到temp表空间）

 

alter tablespace temp add tempfile '***/' size **M autoextend on next 10M maxsize 8192M;

恢复temp(建temp2并切换，再建temp再切换回temp)

1.create temporary tablespace temp2 tempfile '***/' size **M autoextend on next 10M maxsize 8192M;

2.alter database default temporary tablespace temp2;

3.drop tablespace temp including contents and datafiles;

4.create temporary tablespace temp tempfile '***/' size **M autoextend on next 10M maxsize 8192M;

5.alter database default temporary tablespace temp;

6.drop tablespace temp2 including contents and datafiles;