ITPub博客

发现把oracle用户放在dba组里也不行了，放在oinstall组里解决问题。[@more@]

This message is strictly informational. The "user requested cancel of current operation on rowid AAAAAAAAAx" error is just evidence of excessive rollback activity on the source database. This is when a user issues a ctrl c during execution of a script or someone's session was killed before a commit was issued. This is not a fatal error.[@more@]

SET verify OFF 取消参数传入时对sql的校验set pagesize 0 取消返回记录之间的空行[@more@]

克隆ORACLE之后用SYSDBA登陆出现1031错误，原因是新建的ORACLE用户没有放在DBA组里。[@more@]

把脚本COPY出来，删掉最后的"/;"，再执行就可以了。[@more@]

SQL> create pfile='initsid.ora' from spfile;[@more@]

re-execute $ORACLE_HOME/rdbms/admin/catexp.sql[@more@]

CREATE undo TABLESPACE UNDOTBS1 DATAFILE '/disk1/oradata/test/undotbs1.dbf' SIZE 2000M reuse AUTOEXTEND ON NEXT 1280K MAXSIZE UNLIMITED;alter system set undo_tablespace=undotbs1 scope=both;前提：启用spfile[@more@]

CREATE SPFILE='$ORACLE_HOME/dbs/spfilesid.ora'FROM PFILE='$ORACLE_HOME/dbs/initsid.ora'; 重启。[@more@]

start up数据库的时候报错：ORA-01157: cannot identify/lock data file 5 - see DBWR trace file ORA-01110: data file 5: '/disk3/oradata/test.dbf' 因为不需要恢复此数据文件，所以将它删除即可。解决方法：在数据库mount的状态下，alter database datafile '/disk3/oradata/test.dbf' offline drop;然后打开数据库：alter database open;删除表空间：select name from v$tablespace where ts# in(select ts# from v$datafile where name = '/disk3/oradata/test.dbf');drop tablespace NAME including contents;[@more@]

新建的用户oracle，组定义为oinstall，用sqlplus '/as sysdba'登陆报权限不够，用usermod将组改为dba后解决。[@more@]

每次取出cache_size个值，放在cache里。last_number每次增加的值＝INCREMENT_BY*CACHE_SIZE[@more@]

Problem:SQL> alter tablespace xxxxxxx begin backup;alter tablespace xxxxxxxx begin backupORA-01123: cannot start online backup; media recovery not enabledSQL> archive log listDatabase log mode No Archive ModeAutomatic archival EnabledArchive destination /disk01/archive/xxxxxOldest online log sequence 84Current log sequence 86[@more@]Solution:SQL> shutdown immediateDatabase closed.Database dismounted.ORACLE instance shut down.SQL> star

按日期升序排列：ls -l|sort +5按日期降序排列：ls -lt按名称升序排列：ls -ln[@more@]

需要建立一个独立的INSTANCE，源文件是从一个RAC的其中一个节点上CLONE出来的。在目标主机上解包、relink（这一步缺了几个lib，从源主机上获得后解决）后，启动oracle到nomount状态，报如下错误：[@more@]SQL> startup nomount pfile="$ORACLE_HOME/dbs/init<sid>.ora";ORA-27504: IPC error creating OSD context在metalink上获知此问题是因源主机是RAC的节点造成的，故需在新主机上关闭RAC功能。步骤如下：cd $ORACLE_HOME/rdbms/libmake -f ins_rdbms.mk rac_offmake -f ins_rdbms.mk ioracle如果要重新开启racORAC

查看oracle位数：file $ORACLE_HOME/bin/oracle 查看solaris的补丁安装情况：pkginfo -i SUNWxwfnt[@more@]

在指定了shareplex用到的表空间后开始建立相关对象，报错：ORA-01552 cannot use system rollback segment for non-system tablespace [@more@]与相关人员交流后了解这个库是从其他库CLONE过来的，版本是9.2.0.4。检查其表空间，发现没有UNDOTBS，有一个RBS，在INIT文件中，有 rollback_segments =(rbs01......) ，据此怀疑原库是从8i升级而来。解决步骤：1、建立UNDOcreate undo tablespace undotbsdatafile '/×××××/undotbs1.dbf' size 200M REUSE AUTOEXTEND ON NEXT 5120K MAXSIZE UNLIMITED;2、修改INIT注释rollback_segments添

This command create a compress file by tar|gzip.tar cvf - directory_name(or file_name) | gzip > file_name.tar.gzOS: Solaris 9[@more@]