先假设一种情况,假设有一个数据库,数据量是2T。 如果要对这个数据库进行迁移,在不考虑磁盘空间的前提下,要求在最短的时间内完成切换操作。
在不考虑磁盘空间前提下,备份时间,网络传输时间,恢复时间。 这个时间对于7*24*365的系统来说,肯定是没办法承受的。
如果在考虑下备份空间+迁移后的空间。 这个对磁盘的开销也是非常大。
在我的前一篇Blog里,提到了RMAN的异机恢复,这种方法可以缩短我们DB的迁移时间。 但是空间问题还没有解决。在这一篇,我们主要从迁移时间的角度来考虑这个问题。
RMAN 备份异机恢复 并创建新DBID
http://blog.csdn.net/tianlesoftware/archive/2011/03/11/6240983.aspx
在异机恢复中,我们最后对db 是用open resetlogs的方式打开的。
我们先看open resetlogs的影响:
(1)如果current online redo可以访问,那么对其进行归档操作。 然后清空所有online redo 的内容,并将日志的序列设置为1.
(2)如果online redo 不存在,就创建online redo。 这个位置是在控制文件里记录的。 我们在做RMAN 异机恢复的时候,验证过这点。
(3)开始一个新的incarnation,初始化redo thread和 online redo 在控制文件中的记录。
比如:将redo thread 的状态设置为closed,在redo thread的记录中设置当前的log sequence 为1. 设置每个redo thread 的checkpoint 到log sequence 1。
从每个thread 选择一个redo log,并设置其sequence 为1.
(4)在控制文件中更新当前所有的datafile 和online redo logs,及以后的归档文件用新的resetlog SCN 和 time stamp。
注意: open resetlogs 不会充值SCN.
一般用open resetlogs 打开数据库之后,建议立即对数据库进行一次备份。
下面是一个迁移的测试:
1. 在Target 库rman 备份数据库,并将备份copy到Auxiliary库的相同位置
RMAN 的恢复对备份位置有讲究,对于nocatalog模式下,这些备份的位置是写在控制文件里的。所以目录结构要相同。
关于RMAN 的shell 备份脚本,参考:
Nocatalog 下的RMAN 增量备份 shell脚本
http://blog.csdn.net/tianlesoftware/archive/2011/01/26/6164931.aspx
[oracle@qs-dmm-rh1 backup]$ scp * 192.168.2.43:/u01/backup
oracle@192.168.2.43's password:
arch_23m709vi_1_1_20110312 100% 6431KB 6.3MB/s 00:00
ctl_file_24m709vn_1_1_20110312 100% 9568KB 9.3MB/s 00:01
dave_lev0_1rm709u9_1_1_20110312 100% 610MB 12.0MB/s 00:51
dave_lev0_1sm709u9_1_1_20110312 100% 399MB 11.7MB/s 00:34
dave_lev0_1tm709u9_1_1_20110312 100% 4728KB 4.6MB/s 00:00
dave_lev0_1vm709uk_1_1_20110312 100% 1344KB 1.3MB/s 00:01
dave_spfile_25m709vp_1_1_20110312 100% 96KB 96.0KB/s 00:00
2. 在Auxiliary库创建相关的目录,口令文件,和初始化文件。
初始化文件可以直接从Target copy过来。如果目录结构相同的话,都不用修改参数。
3. 用初始化文件将数据库启动到noumout状态。
SQL> startup nomount pfile=?/dbs/initdave.ora
ORACLE instance started.
Total System Global Area 196681728 bytes
Fixed Size 1335444 bytes
Variable Size 83890028 bytes
Database Buffers 109051904 bytes
Redo Buffers 2404352 bytes
4. 恢复控制文件
RMAN> restore controlfile from '/u01/backup/ctl_file_2qm72jcf_1_1_20110312';
Starting restore at 12-MAR-11
using channel ORA_DISK_1
channel ORA_DISK_1: restoring control file
channel ORA_DISK_1: restore complete, elapsed time: 00:00:02
output file name=/u01/oradata/control01.ctl
output file name=/u01/oradata/control02.ctl
output file name=/u01/oradata/control03.ctl
Finished restore at 12-MAR-11
5. 将数据库启动到mout状态
SQL> alter database mount;
Database altered.
6. restore 数据库
RMAN> restore database;
Starting restore at 12-MAR-11
released channel: ORA_DISK_1
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=20 device type=DISK
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 /u01/app/oracle/oradata/dave/undotbs01.dbf
channel ORA_DISK_1: reading from backup piece /u01/backup/dave_lev0_28m70ar0_1_1_20110312
channel ORA_DISK_1: piece handle=/u01/backup/dave_lev0_28m70ar0_1_1_20110312 tag=DAVE_LEV0
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:00:01
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 00004 to /u01/app/oracle/oradata/dave/users01.dbf
channel ORA_DISK_1: reading from backup piece /u01/backup/dave_lev0_2am70ara_1_1_20110312
channel ORA_DISK_1: piece handle=/u01/backup/dave_lev0_2am70ara_1_1_20110312 tag=DAVE_LEV0
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:00:01
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 /u01/app/oracle/oradata/dave/sysaux01.dbf
channel ORA_DISK_1: reading from backup piece /u01/backup/dave_lev0_27m70ar0_1_1_20110312
channel ORA_DISK_1: piece handle=/u01/backup/dave_lev0_27m70ar0_1_1_20110312 tag=DAVE_LEV0
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:00:25
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 /u01/app/oracle/oradata/dave/system01.dbf
channel ORA_DISK_1: reading from backup piece /u01/backup/dave_lev0_26m70ar0_1_1_20110312
channel ORA_DISK_1: piece handle=/u01/backup/dave_lev0_26m70ar0_1_1_20110312 tag=DAVE_LEV0
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:00:25
Finished restore at 12-MAR-11
7. recover 数据库
RMAN> recover database;
Starting recover at 12-MAR-11
using channel ORA_DISK_1
starting media recovery
channel ORA_DISK_1: starting archived log restore to default destination
channel ORA_DISK_1: restoring archived log
archived log thread=1 sequence=24
channel ORA_DISK_1: reading from backup piece /u01/backup/arch_2cm70as8_1_1_20110312
channel ORA_DISK_1: piece handle=/u01/backup/arch_2cm70as8_1_1_20110312 tag=ARC_BAK
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:00:01
archived log file name=/u01/app/oracle/product/11.2.0/dbhome_1/dbs/arch1_24_745352047.dbf thread=1 sequence=24
channel ORA_DISK_1: starting archived log restore to default destination
channel ORA_DISK_1: restoring archived log
archived log thread=1 sequence=25
channel ORA_DISK_1: reading from backup piece /u01/backup/arch_2dm70as8_1_1_20110312
channel ORA_DISK_1: piece handle=/u01/backup/arch_2dm70as8_1_1_20110312 tag=ARC_BAK
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:00:01
archived log file name=/u01/app/oracle/product/11.2.0/dbhome_1/dbs/arch1_25_745352047.dbf thread=1 sequence=25
unable to find archived log
archived log thread=1 sequence=26
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of recover command at 03/12/2011 00:58:04
RMAN-06054: media recovery requesting unknown archived log for thread 1 with sequence 26 and starting SCN of 1055401
8. 在主库进行一些事务操作
SQL> create table anhui as select * from all_tables;
Table created.
SQL> commit;
Commit complete.
SQL> alter system switch logfile;
System altered.
9. 关闭主库,迅速把主库的归档文件和online redo 复制到Auxiliary库
[oracle@qs-dmm-rh1 archivelog]$ scp * 192.168.2.43:/u01/archivelog
oracle@192.168.2.43's password:
1_26_745352047.dbf 100% 1509KB 1.5MB/s 00:00
[oracle@qs-dmm-rh1 dave]$ scp redo* 192.168.2.43:/u01/app/oracle/oradata/dave/
oracle@192.168.2.43's password:
redo01.log 100% 50MB 16.7MB/s 00:03
redo02.log 100% 50MB 12.5MB/s 00:04
redo03.log 100% 50MB 12.5MB/s 00:04
[oracle@qs-dmm-rh1 dave]$
10. 再次recover database
RMAN> recover database;
Starting recover at 12-MAR-11
using channel ORA_DISK_1
starting media recovery
archived log for thread 1 with sequence 26 is already on disk as file /u01/app/oracle/oradata/dave/redo02.log
archived log for thread 1 with sequence 27 is already on disk as file /u01/app/oracle/oradata/dave/redo03.log
archived log file name=/u01/app/oracle/oradata/dave/redo02.log thread=1 sequence=26
archived log file name=/u01/app/oracle/oradata/dave/redo03.log thread=1 sequence=27
media recovery complete, elapsed time: 00:00:05
Finished recover at 12-MAR-11
11. 打开数据库
SQL>alter database open resetlogs;
SQL> select count(*) from dave;
COUNT(*)
----------
71157
12. 重建Temp 表空间
在恢复的时候,会自动创建TEMP 表空间。但是和布豆聊到这个问题的时候,他说这个temp 文件好像没有起到作用。 尤其是数据量大的时候,所以他建议重建添加temp 数据文件。
SQL> select name from v$tempfile;
NAME
--------------------------------------------------------------------------------
/u01/app/oracle/oradata/dave/temp01.dbf
SQL> alter tablespace temp add tempfile '/u01/app/oracle/oradata/dave/temp02.dbf ' size 8192M;
Tablespace altered.
SQL> select name from v$tempfile;
NAME
--------------------------------------------------------------------------------
/u01/app/oracle/oradata/dave/temp01.dbf
/u01/app/oracle/oradata/dave/temp02.dbf
SQL> alter database tempfile '/u01/app/oracle/oradata/dave/temp01.dbf' offline;
Database altered.
SQL> alter database tempfile '/u01/app/oracle/oradata/dave/temp01.dbf' drop including datafiles;
Database altered.
SQL> select name from v$tempfile;
NAME
--------------------------------------------------------------------------------
/u01/app/oracle/oradata/dave/temp02.dbf