rman全库备份，恢复过程

最近一直在忙于写shell脚本，忽略了oracle的基础的回顾，正好做了个rman的恢复，记录一下加深印象！

oracle全库备份的理想状态下在mount时做全备，这样不需要去做考虑一致性问题。仅仅是理想而已。

将数据库开为归档模式下，进行全备，脚本如下

log_file=/paic/hq/bk/restore/home/op11202/xionglang/investdw/full_backup_${ORACLE_SID}_`date '+%Y%m%d_%H%M%S'`.log
export log_file
rman target / << EOF > ${log_file} 2>&1
sql 'alter system switch logfile';
run
{
allocate channel c1 type disk;
allocate channel c2 type disk;
allocate channel c3 type disk;
allocate channel c4 type disk;
allocate channel c5 type disk;
allocate channel c6 type disk;
allocate channel c7 type disk;
allocate channel c8 type disk;
BACKUP as compressed backupset full DATABASE plus archivelog FORMAT '/paic/hq/bk/restore/home/op11202/xionglang/investdw/full_%T-%d-%u-%p'  filesperset=10;
backup as copy current controlfile format '/paic/hq/bk/restore/home/op11202/xionglang/investdw/ctrl_%t-%s-%p.f';
release channel c1;
release channel c2;
release channel c3;
release channel c4;
release channel c5;
release channel c6;
release channel c7;
release channel c8;
}
EOF
echo "Backup finished at `date '+%Y%m%d_%H%M%S'`" >> ${log_file}

1.生成一份原开发/测试库的pfile 
2.利用步骤1中的pfile，启动到nomount
3.rman target /
RMAN>  Restore controlfile from  ‘/xxxx/xxxx/*****.f’;   /xxxx/xxxx/*****.f为生产dg库的controlfile的备份
RMAN> Alter database mount;
RMAN>  Catalog start with ‘/xxxx/xxxx/’;  注册备份片到恢复出来的控制文件里

4.后台执行如下restore脚本。如果恢复出来了归档日志restore database之后需要recover database;
(注意incarnation相关报错)
5.恢复完成以后，shutdown实例，修改pfile里的db_name为原测试库SID  启动到nomount
6.使用create controlfile set database 命令重建控制文件
7.alter database open resetlogs;
8.添加临时文件

恢复脚本：

log_file=/paic/dev/oracle/11g/odsz11g/xionglang/xionglang/d1gccrep/restore_${ORACLE_SID}_`date '+%Y%m%d_%H%M%S'`.log  
export log_file
rman target / << EOF > ${log_file} 2>&1
run
{
allocate channel c1 type disk ;
allocate channel c2 type disk ;
allocate channel c3 type disk ;
allocate channel c4 type disk ;
allocate channel c5 type disk ;
allocate channel c6 type disk ;
allocate channel c7 type disk ;
allocate channel c8 type disk ;
set newname for datafile 1 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 10 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 11 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 12 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 13 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 14 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 15 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 16 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 17 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 18 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 19 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 2 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 20 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 21 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 22 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 23 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 24 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 25 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 26 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 27 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 28 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 29 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 3 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 30 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 31 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 32 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 33 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 34 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 35 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 36 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 37 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 38 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 39 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 4 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 40 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 41 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 42 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 43 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 44 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 45 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 46 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 47 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 48 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 49 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 5 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 50 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 51 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 52 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 53 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 54 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 55 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 56 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 57 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 58 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 59 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 6 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 60 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 61 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 62 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 63 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 64 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 65 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 66 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 67 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 68 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 69 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 7 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 70 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 71 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 72 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 73 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 74 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 75 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 8 to '+DATA_10G_D1GCCREP_MDG';
set newname for datafile 9 to '+DATA_10G_D1GCCREP_MDG';
set newname for tempfile 1 to '+DATA_10G_D1GCCREP_MDG';
set newname for tempfile 2 to '+DATA_10G_D1GCCREP_MDG';
set newname for tempfile 3 to '+DATA_10G_D1GCCREP_MDG';
restore database;
switch datafile all;
switch tempfile all;
release channel c1;
release channel c2;
release channel c3;
release channel c4;
release channel c5;
release channel c6;
release channel c7;
release channel c8;
}                                                                  
EOF                                                                
echo "Restore finished at `date '+%Y%m%d_%H%M%S'`" >> ${log_file}  

前面提到了恢复过程可能因为不一直导致alter database open resetlogs 起库失败。这个过程我建议使用recover database using backup controlfile utile time '2015-12-17 23:01:22' 这种指定时间去做不完全恢复，这种方式比你去应用一个个的日志恢复到一致的几率要高。因为切换日志的时候可能会有长事务并没有结束。而你指定时间点，这个时间点可能有事务，但是你可以往后面添加几秒或是十几秒基本就可以处于一致状态，然后open数据库。

查看数据库是否一致：（虽然都处于一致状态，但是你open还是有可能open是否，这样就需要向后面追加几秒时间才能正常起库。）

SQL> select to_char( max( CHECKPOINT_TIME  ) ,'yyyy-mm-dd hh24:mi:ss' )  from v$datafile_header ;

 

TO_CHAR(MAX(CHECKPO

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

2015-12-16 23:01:08

 

SQL> select to_char( min( CHECKPOINT_TIME  ) ,'yyyy-mm-dd hh24:mi:ss' )  from  v$datafile;

 

TO_CHAR(MIN(CHECKPO

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

2015-12-16 23:01:08

 

SQL> select controlfile_time fromv$database;

 

CONTROLFILE_TIME

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

2015-12-16 23:01:08

 

 

SQL>  select  file#,checkpoint_change#  from  v$datafile_header  where fuzzy ='YES' ;

 

no rows selected

 

SQL> alter database open read only;

alter database open read only

*

ERROR at line 1:

ORA-16004: backup database requiresrecovery

ORA-01113: file 1 needs media recovery ifit was restored from backup, or END BACKUP if it was not

ORA-01110: data file 1:'+DATA_10G_D1GCCREP_MDG/d1gccrep/datafile/system.298.898716267'

 

SQL> recover database using backupcontrolfile;

ORA-00279: change 9910161874981 generatedat 12/16/2015 23:01:08 needed for thread 1

ORA-00289: suggestion : +FRA_10G_LARGE_MDG

ORA-00280: change 9910161874981 for thread1 is in sequence #3085

 

 

Specify log: {<RET>=suggested |filename | AUTO | CANCEL}

cancel

Media recovery cancelled.

另外在提一下，如果恢复不想使用set newname的方式可以通过db_file_name_convet的方式去转换文件名的路径。asm指定到dbname一级，文件系统指定到文件名前一级目录。

NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
db_file_name_convert                 string      +DATA_10G_T1GCCREP_MDG/t1gccrep, +DATA_10G_D1GCCREP_MDG/d1gccrep
log_file_name_convert                string      +DATA_10G_T1GCCREP_MDG/t1gccrep, +DATA_10G_D1GCCREP_MDG/d1gccrep, +FRA_10G_LARGE_MDG/t1gccrep, +FRA_10G_LARGE_MDG/d1gccrep

附上生成set newname的语句

select 'set newname for datafile ' || file# || ' to ''+${data_dg}'';'
  from v\$datafile
union
select 'set newname for tempfile ' || file# || ' to ''+${data_dg}'';'
  from v\$tempfile;

如果想要修改数据库的db_name可以通过nid的修改。需要数据库起到mount状态：

syspwd=`od -N 4 -t x4/dev/random | head -1 | awk '{print $2}'`

orapwdfile=$ORACLE_HOME/dbs/orapw${OLDSID} password=$syspwd entries=8 force=y

orapwdfile=$ORACLE_HOME/dbs/orapw${NEWSID} password=$syspwd entries=8 force=y

nidtarget=sys/$syspwd dbname=${NEWSID} >>$crt_log 2>&1 <<!

Y

!