时间:2022-03-12 08:49:27 | 栏目:Oracle | 点击:次
介绍
控制文件(control file)是一个相当小的文件(最多能增长到64M左右),其中包含Oracle需要的其他文件的一个目录。参数文件告知实例控制文件的位置,控制文件则告知示例数据库和在线重做日志文件的位置。控制文件还告知了Oracle其他一些事情,如已发生检查点的有关信息、数据库名(必须和db_name参数匹配)、创建数据库的时间戳、归档重做日志的历史(有时这会让控制文件变大)、RMAN信息等。
控制文件应该通过硬件(RAID)多路保存,如果不支持镜像,则要通过Oracle多路保存。应该有不止一个副本,而且它们应该保存在不同的磁盘上,以防止万一出现磁盘故障而丢失控制文件。丢失控制文件并不是致命的,但是会使恢复变得困难很多。
如果丢失了所有的控制文件并且没有任何的备份,我们可以通过重建控制文件来打开数据库。其中,重建控制文件至少需要以下信息:
1.数据库名
2.字符集
3.数据文件名称
4.初始化参数,包括MAXLOGFILES、MAXLOGMEMBERS、MAXDATAFILES、MAXINSTANCES、MAXLOGHISTORY等;
一、环境准备
数据库版本
我们在Oracle11g中进行测试。
SQL> SQL> select * from v$version; BANNER -------------------------------------------------------------------------------- Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - Production PL/SQL Release 11.2.0.3.0 - Production CORE 11.2.0.3.0 Production TNS for Linux: Version 11.2.0.3.0 - Production NLSRTL Version 11.2.0.3.0 - Production SQL>
删除控制文件
1.通过查询control_files初始化参数,获取控制文件路径;
SQL> SQL> show parameter control_files NAME TYPE VALUE ------------------------------------ ----------- ------------------------------ control_files string /u01/app/oracle/oradata/HOEGH/ control01.ctl, /u01/app/oracle /oradata/HOEGH/control02.ctl SQL>
2.然后,使用rm命令删除控制文件;
[oracle@HOEGH ~]$ rm /u01/app/oracle/oradata/HOEGH/control01.ctl [oracle@HOEGH ~]$ rm /u01/app/oracle/oradata/HOEGH/control02.ctl [oracle@HOEGH ~]$
3.此时,强制关闭数据库,然后重启数据库,报ORA-00205错误。需要注意的是,此时执行shutdown immediate命令,数据库无法正常关闭,只能关闭到mounted状态;需要使用shutdown abort命令强制关闭数据库。
SQL> SQL> shutdown immediate Database closed. ORA-00210: cannot open the specified control file ORA-00202: control file: \'/u01/app/oracle/oradata/HOEGH/control01.ctl\' ORA-27041: unable to open file Linux Error: 2: No such file or directory Additional information: 3 SQL> select status from v$instance; STATUS ------------ MOUNTED SQL> SQL> shutdown abort ORACLE instance shut down. SQL> SQL> SQL> startup ORACLE instance started. Total System Global Area 941600768 bytes Fixed Size 1348860 bytes Variable Size 515902212 bytes Database Buffers 419430400 bytes Redo Buffers 4919296 bytes ORA-00205: error in identifying control file, check alert log for more info SQL>
二、获取数据库名
首先生成文本格式的参数文件;
SQL> SQL> create pfile from spfile; File created. SQL>
打开参数文件,查看db_name参数值,即为数据库名称。
[oracle@hoegh dbs]$ cat initHOEGH.ora HOEGH.__db_cache_size=419430400 HOEGH.__java_pool_size=4194304 HOEGH.__large_pool_size=4194304 HOEGH.__oracle_base=\'/u01/app/oracle\'#ORACLE_BASE set from environment HOEGH.__pga_aggregate_target=377487360 HOEGH.__sga_target=566231040 HOEGH.__shared_io_pool_size=0 HOEGH.__shared_pool_size=130023424 HOEGH.__streams_pool_size=0 *.audit_file_dest=\'/u01/app/oracle/admin/HOEGH/adump\' *.audit_trail=\'db\' *.compatible=\'11.2.0.0.0\' *.control_files=\'/u01/app/oracle/oradata/HOEGH/control01.ctl\',\'/u01/app/oracle/oradata/HOEGH/control02.ctl\' *.db_block_size=8192 *.db_domain=\'\' *.db_name=\'HOEGH\' *.diagnostic_dest=\'/u01/app/oracle\' *.dispatchers=\'(PROTOCOL=TCP) (SERVICE=HOEGHXDB)\' *.memory_max_target=943718400 *.memory_target=943718400 *.open_cursors=300 *.processes=150 *.remote_login_passwordfile=\'EXCLUSIVE\' *.undo_tablespace=\'UNDOTBS1\' [oracle@hoegh dbs]$
三、启动到nomount状态,获取字符集
由于需要执行查询语句select userenv('language') from dual;来获取字符集,因此需要将数据库启动到nomount状态。
SQL> SQL> startup nomount ORACLE instance started. Total System Global Area 941600768 bytes Fixed Size 1348860 bytes Variable Size 515902212 bytes Database Buffers 419430400 bytes Redo Buffers 4919296 bytes SQL> SQL> select userenv(\'language\') from dual; USERENV(\'LANGUAGE\') ---------------------------------------------------- AMERICAN_AMERICA.US7ASCII SQL> SQL>
四、获取数据文件名称
通过ls命令获取数据文件列表。
[oracle@hoegh HOEGH]$ ls -lh total 1.8G -rw-r----- 1 oracle oinstall 314M May 30 11:07 example01.dbf -rw-r----- 1 oracle oinstall 51M May 30 11:07 redo01.log -rw-r----- 1 oracle oinstall 51M May 30 11:07 redo02.log -rw-r----- 1 oracle oinstall 51M May 30 11:07 redo03.log -rw-r----- 1 oracle oinstall 541M May 30 11:07 sysaux01.dbf -rw-r----- 1 oracle oinstall 721M May 30 11:07 system01.dbf -rw-r----- 1 oracle oinstall 30M Oct 13 2014 temp01.dbf -rw-r----- 1 oracle oinstall 96M May 30 11:07 undotbs01.dbf -rw-r----- 1 oracle oinstall 5.1M May 30 11:07 users01.dbf [oracle@hoegh HOEGH]$
五、生成创建控制文件脚本
这样,创建控制文件所需的基本信息都已经有了,我们来生成创建控制文件脚本。
STARTUP NOMOUNT CREATE CONTROLFILE REUSE DATABASE \"HOEGH\" NORESETLOGS ARCHIVELOG MAXLOGFILES 5 MAXLOGMEMBERS 3 MAXDATAFILES 100 MAXINSTANCES 1 MAXLOGHISTORY 226 LOGFILE GROUP 1 \'/u01/app/oracle/oradata/HOEGH/redo01.log\' SIZE 50M, GROUP 2 \'/u01/app/oracle/oradata/HOEGH/redo02.log\' SIZE 50M, GROUP 3 \'/u01/app/oracle/oradata/HOEGH/redo03.log\' SIZE 50M DATAFILE \'/u01/app/oracle/oradata/HOEGH/system01.dbf\', \'/u01/app/oracle/oradata/HOEGH/sysaux01.dbf\', \'/u01/app/oracle/oradata/HOEGH/undotbs01.dbf\', \'/u01/app/oracle/oradata/HOEGH/users01.dbf\', \'/u01/app/oracle/oradata/HOEGH/example01.dbf\', \'/u01/app/oracle/oradata/HOEGH/temp01.dbf\' CHARACTER SET US7ASCII ;
六、重建控制文件
需要注意的是,在执行上述创建脚本时会报错,系统提示临时文件不属于数据文件,如下所示:
SQL> @/u01/app/oracle/oradata/HOEGH/CreateControlFile.sql ORA-01081: cannot start already-running ORACLE - shut it down first CREATE CONTROLFILE REUSE DATABASE \"HOEGH\" NORESETLOGS ARCHIVELOG * ERROR at line 1: ORA-01503: CREATE CONTROLFILE failed ORA-01160: file is not a data file ORA-01110: data file : \'/u01/app/oracle/oradata/HOEGH/temp01.dbf\' SQL>
修改脚本并重新执行,重建控制文件后,数据库会打开到mount状态。
SQL> SQL> @/u01/app/oracle/oradata/HOEGH/CreateControlFile.sql ORACLE instance started. Total System Global Area 941600768 bytes Fixed Size 1348860 bytes Variable Size 515902212 bytes Database Buffers 419430400 bytes Redo Buffers 4919296 bytes Control file created. SQL> SQL> select status from v$instance; STATUS ------------ MOUNTED SQL>
七、打开数据库
在打开数据库时,会报错,提示system01数据文件需要执行介质恢复,我们执行recover database即可。
SQL> SQL> alater database open; SP2-0734: unknown command beginning \"alater dat...\" - rest of line ignored. SQL> SQL> alter database open; alter database open * ERROR at line 1: ORA-01113: file 1 needs media recovery ORA-01110: data file 1: \'/u01/app/oracle/oradata/HOEGH/system01.dbf\' SQL> SQL> recover database; Media recovery complete. SQL> SQL> alter database open; Database altered. SQL> SQL> select * from v$version; BANNER -------------------------------------------------------------------------------- Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - Production PL/SQL Release 11.2.0.3.0 - Production CORE 11.2.0.3.0 Production TNS for Linux: Version 11.2.0.3.0 - Production NLSRTL Version 11.2.0.3.0 - Production SQL> SQL> select tablespace_name from dba_tablespaces; TABLESPACE_NAME ------------------------------ SYSTEM SYSAUX UNDOTBS1 TEMP USERS EXAMPLE 6 rows selected. SQL>
下面总结一下重建控制文件的步骤:
1.获取数据库名;
2.获取字符集名;
3.获取数据文件名;
4.重建控制文件;
5.执行介质恢复;
6.打开数据库。
总结