Oracle数据库RMAN的一次不完全恢复案例

前2天因为一次错误的truncate 操作，删除了hz用户下的几个表，费了1天多时间才恢复出来，在这里记录下恢复过程和这个深刻的教训。
使用坏境： oracle11.2+rhce5.5 RAC,归档模式,数据库有1TB容量。
在配置流环境时，因为一个误操作，删除了hz用户下的几个表。在这个数据库上还部署了其它几个用户，停机恢复数据显然是不可能的，分析情况后可以使用下面两种方式来恢复数据。
1、把当前库的备份文件和归档日志传到测试机上进行时间点不完全恢复，然后exp导出需要误删除的数据。
2、使用duplicate表空间的方式生成辅助数据库，在新库中导出需要的数据。

这次的数据恢复操作我我使用第一种方式，以下是操作步骤：
1、拷贝数据文件和2节点的归档日志文件。由于故障库的备份是nfs挂载到备份机上的，省去了拷贝数据文件的时间，只需要把2节点下的archive日志文件拷贝到测试库上就行。
2、创建新库的pfile参数文件，拷贝故障库的spfile文件，修改参数并删除有关rac的不需要参数信息。
3、创建新库的密码文件.
orapwd file=/u02/app/oracle/product/11.2.0/db1/dbs/orapwklir passord=oracle entries=10 force=y
4、创建新库的dump目录。
5、启动新库到nomount下。
[oracle@kms2 dbs]$ export ORACLE_SID=klir
[oracle@kms2 dbs]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.1.0 Production on Thu Nov 18 09:56:35 2010
Copyright (c) 1982, 2009, Oracle. All rights reserved.
Connected to an idle instance.
SQL> startup nomount pfile='/u02/pfile_klir.ora'
6、从备份集中恢复控制文件,并启动到mount状态下。
RMAN> restore controlfile from /orabk/ctl_c-949039848-20101116-00.ctl;
RMAN> alter database mount;

7、恢复表空间。因为只是一个表空间下的一些表被删除了，朋友建议我只恢复有问题的表空间就可以了，当然system，sysaux,undo这些基本的表空间是要恢复的。

RMAN> restore tablespace system;
Starting restore at 16-NOV-10
Starting implicit crosscheck backup at 16-NOV-10
allocated channel: ORA_DISK_1
Crosschecked 198 objects
Finished implicit crosscheck backup at 16-NOV-10
Starting implicit crosscheck copy at 16-NOV-10
using channel ORA_DISK_1
Finished implicit crosscheck copy at 16-NOV-10
searching for all files in the recovery area
cataloging files...
no files cataloged
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00001 to +DG1/system01.dbf
channel ORA_DISK_1: reading from backup piece /orabk/db_KLIR_13771_1_1_735012733.dbf
channel ORA_DISK_1: piece handle=/orabk/db_KLIR_13771_1_1_735012733.dbf tag=TAG20101114T021212
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:09:15
Finished restore at 16-NOV-10
RMAN> restore tablespace sysaux;
Starting restore at 16-NOV-10
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00002 to +DG1/sysaux01.dbf
channel ORA_DISK_1: reading from backup piece /orabk/db_KLIR_13770_1_1_735012528.dbf
channel ORA_DISK_1: piece handle=/orabk/db_KLIR_13770_1_1_735012528.dbf tag=TAG20101114T021212
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:08:50
Finished restore at 16-NOV-10
RMAN> restore tablespace HZDATATBS;
Starting restore at 16-NOV-10
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00011 to +DG1/hzdata01.dbf
channel ORA_DISK_1: reading from backup piece /orabk/db_KLIR_13770_1_1_735012528.dbf
channel ORA_DISK_1: piece handle=/orabk/db_KLIR_13770_1_1_735012528.dbf tag=TAG20101114T021212
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:16:35
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00154 to +DG1/hzdata02.dbf
channel ORA_DISK_1: reading from backup piece /orabk/db_KLIR_13938_1_1_735186686.dbf
channel ORA_DISK_1: piece handle=/orabk/db_KLIR_13938_1_1_735186686.dbf tag=TAG20101116T023123
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:02:05
Finished restore at 16-NOV-10
RMAN> restore tablespace hzindtbs;
Starting restore at 16-NOV-10
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00017 to +DG1/hzind02.dbf
channel ORA_DISK_1: restoring datafile 00021 to +DG1/hzind03.dbf
channel ORA_DISK_1: reading from backup piece /orabk/db_KLIR_13773_1_1_735012734.dbf
channel ORA_DISK_1: piece handle=/orabk/db_KLIR_13773_1_1_735012734.dbf tag=TAG20101114T021212
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:13:45
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00012 to +DG1/hzind01.dbf
channel ORA_DISK_1: reading from backup piece /orabk/db_KLIR_13771_1_1_735012733.dbf
channel ORA_DISK_1: piece handle=/orabk/db_KLIR_13771_1_1_735012733.dbf tag=TAG20101114T021212
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:14:46
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00153 to +DG1/hzind04.dbf
channel ORA_DISK_1: reading from backup piece /orabk/db_KLIR_13937_1_1_735186478.dbf
channel ORA_DISK_1: piece handle=/orabk/db_KLIR_13937_1_1_735186478.dbf tag=TAG20101116T023123
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:03:15
Finished restore at 17-NOV-10

RMAN> restore tablespace undotbs2;
Starting restore at 17-NOV-10
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00006 to +DG1/undotbs201.dbf
channel ORA_DISK_1: reading from backup piece /orabk/db_KLIR_13770_1_1_735012528.dbf
channel ORA_DISK_1: piece handle=/orabk/db_KLIR_13770_1_1_735012528.dbf tag=TAG20101114T021212
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:04:46
Finished restore at 17-NOV-10
RMAN> restore tablespace undotbs1;
Starting restore at 17-NOV-10
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00003 to +DG1/undotbs101.dbf
channel ORA_DISK_1: reading from backup piece /orabk/db_KLIR_13773_1_1_735012734.dbf
channel ORA_DISK_1: piece handle=/orabk/db_KLIR_13773_1_1_735012734.dbf tag=TAG20101114T021212
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:11:05
Finished restore at 17-NOV-10

8.Offline 其它不用的表空间。
SQL> alter database datafile 5,7 offline drop;
Database altered.
SQL> alter database datafile 13,14,15,16 offline drop;
Database altered.

9、不完全恢复数据库。

SQL> recover database until time '2010-11-16 16:25:00' using backup controlfile;
ORA-00279: change 2602636589 generated at 11/14/2010 13:16:46 needed for thread
1
ORA-00289: suggestion : +ARCHIVE
ORA-00280: change 2602636589 for thread 1 is in sequence #5709

Specify log: {=suggested | filename | AUTO | CANCEL}
thread_1_seq_5709.528.735293603
ORA-00308: cannot open archived log 'thread_1_seq_5709.528.735293603'
ORA-27037: unable to obtain file status
Linux-x86_64 Error: 2: No such file or directory
Additional information: 3

Specify log: {=suggested | filename | AUTO | CANCEL}
/orabk/aaa/thread_1_seq_5709.528.735293603
ORA-00279: change 2602636589 generated at 11/14/2010 06:43:25 needed for thread
2
ORA-00289: suggestion : +ARCHIVE
ORA-00280: change 2602636589 for thread 2 is in sequence #6914

ORA-00289: suggestion : +ARCHIVE
ORA-00280: change 2609694870 for thread 1 is in sequence #5824
ORA-00278: log file '/orabk/aaa/arch11/1_5823_719402218.dbf' no longer needed
for this recovery

Specify log: {=suggested | filename | AUTO | CANCEL}
/orabk/aaa/arch11/1_5824_719402218.dbf
ORA-00279: change 2609942602 generated at 11/16/2010 16:07:30 needed for thread
2
ORA-00289: suggestion : +ARCHIVE
ORA-00280: change 2609942602 for thread 2 is in sequence #6990
ORA-00278: log file '/orabk/aaa/arch22/2_6989_719402218.dbf' no longer needed
for this recovery
........................省略中间部分
Specify log: {=suggested | filename | AUTO | CANCEL}
/orabk/aaa/arch22/2_6990_719402218.dbf
ORA-00279: change 2609955345 generated at 11/16/2010 16:08:25 needed for thread
1
ORA-00289: suggestion : +ARCHIVE
ORA-00280: change 2609955345 for thread 1 is in sequence #5825
ORA-00278: log file '/orabk/aaa/arch11/1_5824_719402218.dbf' no longer needed
for this recovery

Specify log: {=suggested | filename | AUTO | CANCEL}
/orabk/aaa/arch11/1_5825_719402218.dbf
Log applied.
Media recovery complete.

10、打开数据库。
SQL> alter database open resetlogs;
经过几分钟的等待后，数据成功打开，登录数据库查询需要的文件都已经恢复出来了。谢天谢地！

这次数据恢复操作，用了近1天的时间，当时发生问题时我都快搞懵了，幸好在朋友的帮助下顺利完数据恢复，在这里要非常感谢他们！顺便记录下恢复期间遇到的几个问题。
<1>、有些日志文件在故障库的日志目录里是没有的，需要从备份中还原，不同节点的日志文件恢复时要写上thread=*命令，例如:

RMAN> restore archivelog from logseq 6914 until logseq 6915 thread=2;
Starting restore at 17-NOV-10
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=182 instance=klir2 device type=DISK
channel ORA_DISK_1: starting archived log restore to default destination
channel ORA_DISK_1: restoring archived log
archived log thread=2 sequence=6914
channel ORA_DISK_1: reading from backup piece /orabk/arc_KLIR_13864_1_1_735184831.dbf
channel ORA_DISK_1: piece handle=/orabk/arc_KLIR_13864_1_1_735184831.dbf tag=TAG20101116T020016
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:00:07
channel ORA_DISK_1: starting archived log restore to default destination
channel ORA_DISK_1: restoring archived log
archived log thread=2 sequence=6915
channel ORA_DISK_1: reading from backup piece /orabk/arc_KLIR_13866_1_1_735184956.dbf
channel ORA_DISK_1: piece handle=/orabk/arc_KLIR_13866_1_1_735184956.dbf tag=TAG20101116T020016
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:00:07
Finished restore at 17-NOV-10

否则会提示以下错误，
RMAN> restore archivelog from logseq 6914 until logseq 6915;
Starting restore at 17-NOV-10
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=182 instance=klir2 device type=DISK
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of restore command at 11/17/2010 08:15:07
RMAN-20242: specification does not match any archived log in the repository

<2>、故障库恢复出日志文件后，因为在asm磁盘组里，要进入asmcmd里拷贝出日志文件。
ASMCMD> cp +ARCHIVE/klir/archivelog/2010_11_17/thread_1_seq_5705.369.735274973 /orabk/aaa/

<3>、恢复命令要写对，刚开始写的几种命令都不正确。
SQL> recover database until time '2010-11-16 16:25:00'
ORA-00283: recovery session canceled due to errors
ORA-01610: recovery using the BACKUP CONTROLFILE option must be done

SQL> recover database until time to_date('2010-11-16 16:25:00','yyyy-mm-dd HH24:MI:SS');
ORA-00285: TIME not given as a string constant

<4>、对于这种只是丢失部分表的情况，就可以只还原需要的表空间来打开数据库，可以节省大量的恢复时间。