linux命令的oracle运用（一）
1 dd
模拟物理坏块
SQL> select * from v$database_block_corruption;no rows selectedSQL> select * from v$backup_corruption;no rows selectedSQL> select * from v$copy_corruption;no rows selectedSQL> create tablespace physical_corrupt datafile size 10m autoextend on;Tablespace created.SQL> conn hr/hrConnected.SQL> create table tt (i int) tablespace physical_corrupt;Table created.SQL> insert into tt values(1);1 row created.SQL> insert into tt values(2);1 row created.SQL> commit;Commit complete.SQL> select tt.*,dbms_rowid.rowid_block_number(rowid) from tt; I DBMS_ROWID.ROWID_BLOCK_NUMBER(ROWID)---------- ------------------------------------ 1 15 2 15SQL> select tablespace_name,file_name from dba_data_files where tablespace_name='PHYSICAL_CORRUPT';TABLESPACE_NAME------------------------------FILE_NAME--------------------------------------------PHYSICAL_CORRUPT/u01/app/oracle/oradata/ORCL/datafile/o1_mf_physical_88gkb4l9_.dbf[oracle@Think ~]$ dd of=/u01/app/oracle/oradata/ORCL/datafile/o1_mf_physical_88gkb4l9_.dbf bs=8192 conv=notrunc seek=15 <<!> my name is think> !0+1 records in0+1 records out17 bytes (17 B) copied, 9.9193e-05 s, 171 kB/s用rman向oracle注册，报告填充v$database_block_corruptionRMAN> backup validate tablespace physical_corrupt;Starting backup at 24-OCT-12using target database control file instead of recovery catalogallocated channel: ORA_DISK_1channel ORA_DISK_1: sid=140 devtype=DISKchannel ORA_DISK_1: starting full datafile backupsetchannel ORA_DISK_1: specifying datafile(s) in backupsetinput datafile fno=00006 name=/u01/app/oracle/oradata/ORCL/datafile/o1_mf_physical_88gkb4l9_.dbfchannel ORA_DISK_1: backup set complete, elapsed time: 00:00:01Finished backup at 24-OCT-12SQL> select * from v$database_block_corruption; FILE# BLOCK# BLOCKS CORRUPTION_CHANGE# CORRUPTIO---------- ---------- ---------- ------------------ --------- 6 15 1 0 CORRUPTSQL> select * from tt;select * from tt*ERROR at line 1:ORA-01578: ORACLE data block corrupted (file # 6, block # 15)ORA-01110: data file 6:'/u01/app/oracle/oradata/ORCL/datafile/o1_mf_physical_88gkb4l9_.dbf'
逐行比较两个文本文件，列出差异行
通常的输出格式：
n1 a n2
n1 c n2
n1 d n2
字母(a、d和c)之前的行号(n1)是针对file1的，其后面的行号(n2)是针对file2的。
字母a、d和c分别表示附加、删除和修改操作。
以”<”打头的行属于第一个文件，以”>”打头的行属于第二个文件。
例如：
[oracle@Think udump]$ diff orcl_ora_18900.trc orcl_ora_19032.trc1c1< /u01/app/oracle/admin/orcl/udump/orcl_ora_18900.trc---> /u01/app/oracle/admin/orcl/udump/orcl_ora_19032.trc11,13c11,13< Redo thread mounted by this instance: 0 <none>< Oracle process number: 0< Unix process pid: 18900, image: oracle@Think---> Redo thread mounted by this instance: 1> Oracle process number: 17> Unix process pid: 19032, image: oracle@Think (TNS V1-V3)15,17c15,19< KCCDEBUG_LEVEL = 0< Dynamic strand is set to TRUE< Running with 2 shared and 18 private strand(s). Zero-copy redo is FALSE---> *** 2012-10-23 20:37:12.033> *** SERVICE NAME:(SYS$USERS) 2012-10-23 20:37:12.033> *** SESSION ID:(155.108) 2012-10-23 20:37:12.033> kwqalqt2: Warning !! queue table IX.ORDERS_QUEUETABLE exits, entry in system.aq$_queue_tables missing > kwqalqt2: Warning !! queue table IX.STREAMS_QUEUE_TABLE exits, entry in system.aq$_queue_tables missing
3 watch
监测一个命令的运行结果，并且动态刷新
-n：指定时间间隔，缺省是2秒
-d：高亮显示变化的区域
ctrl+c退出
例如：
[oracle@Think udump]$ watch -n 1 -d du -sh /u01/app/oracle/admin
4 tail
tail -f filename
文件最尾部的内容显示在屏幕上，并且不断刷新，是你看到最新的内容。
未完待续......
