《只读数据文件损坏恢复实验记录讲义》由会员分享，可在线阅读，更多相关《只读数据文件损坏恢复实验记录讲义（44页珍藏版）》请在金锄头文库上搜索。
1、只读表空间的数据文件指为只读数据文件(read only)，当一个表空间从read write更改为read only时，该数据文件里已经产生的脏块儿都会由DBWn写到磁盘，完成一次不完整的完全检查点。从这一刻起，该数据文件数据块和文件头信息都不再更新，包括检查点。在以后open数据库时实例也忽略该只读数据文件文件头的检查点SCN与其他数据文件或联机REDO日志的同步。我们可以通过以下命令在read only和read write之间进行更改read write read only:alter tablespace tablespace_name read only;read only rea
2、d write:alter tablespace tablespace_name read write;查看系统里的只读表空间及其只读数据文件：sysMAA col TABLESPACE_NAME for a20sysMAA col FILE_NAME for a65sysMAA select t.tablespace_name, d.file_id, d.file_name from dba_tablespaces t, dba_data_files d 2 where t.tablespace_name=d.tablespace_name 3 and t.status=READ ONLY
3、4 /TABLESPACE_NAME FILE_ID FILE_NAME- - -LTB 12 /u01/app/oracle/oradata/ltb01.dbf下面我们针对以下几种场景进行恢复：1）控制文件无损，数据库运行时丢失了只读数据文件，访问其内数据报错2）控制文件无损，只读数据文件丢失，数据库无法OPEN3）控制文件无损，只读数据文件头部损坏4）控制文件无损，有读写状态时候的备份，后来更改为只读状态，只读状态时数据文件损坏5）控制文件无损，有只读状态时候的备份，后来更改为读写状态，读写状态时数据文件损坏6）控制文件损坏，只读数据文件损坏7）控制文件损坏，有只读数据文件状态时候的控制文
4、件备份，后来数据文件改为读写，读写状态时数据文件损坏8）控制文件损坏，在线日志文件没有损坏，归档日志丢失，有旧的跟踪控制文件trace，数据库里有只读数据文件前提条件，有相应的备份。操作环境： OS : Oracle Enterprise Linux 5.5 64Bit DB Type : Oracle Restart DB Version : 11.2.0.3准备一个实验对象只读表空间sysMAA create tablespace ltb datafile /u01/app/oracle/oradata/ltb01.dbf size 50M;sysMAA create user l ide
5、ntified by oracle default tablespace ltb;sysMAA grant resource, connect to l;sysMAA create table l.luocs(domain varchar2(100);sysMAA insert into l.luocs values();sysMAA commit;sysMAA alter tablespace ltb read only;场景1：控制文件无损，数据库运行时只读数据文件损坏，访问其内数据报错备份只读表空间ltbRMAN backup tablespace ltb;Starting backup
6、 at 23-DEC-2012 18:29:30using target database control file instead of recovery catalogallocated channel: ORA_DISK_1channel ORA_DISK_1: SID=44 device type=DISKchannel ORA_DISK_1: starting full datafile backup setchannel ORA_DISK_1: specifying datafile(s) in backup setinput datafile file number=00012
7、name=/u01/app/oracle/oradata/ltb01.dbfchannel ORA_DISK_1: starting piece 1 at 23-DEC-2012 18:29:31channel ORA_DISK_1: finished piece 1 at 23-DEC-2012 18:29:32piece handle=/u01/recovery/MAA/backupset/2012_12_23/o1_mf_nnndf_TAG20121223T182931_8ffq8cbz_.bkp tag=TAG20121223T182931 comment=NONEchannel OR
8、A_DISK_1: backup set complete, elapsed time: 00:00:01Finished backup at 23-DEC-2012 18:29:32数据库运行状态下，破坏只读数据文件以模拟损坏oraclemaa3 $ dd if=/dev/zero of=/u01/app/oracle/oradata/ltb01.dbf bs=10M count=55+0 records in5+0 records out52428800 bytes (52 MB) copied, 0.04918 seconds, 1.1 GB/s查看v$datafile内容时，返回正常s
9、ysMAA col NAME for a65sysMAA select checkpoint_change# from v$datafile where file#=12;CHECKPOINT_CHANGE# - 3567837 但日志里将报错：Hex dump of (file 12, block 1) in trace file /u01/app/oracle/diag/rdbms/maa/maa/trace/maa_m000_5343.trcCorrupt block relative dba: 0x03000001 (file 12, block 1)Completely zero b
10、lock found during kcvxfh v8Reading datafile /u01/app/oracle/oradata/ltb01.dbf for corruption at rdba: 0x03000001 (file 12, block 1)Reread (file 12, block 1) found different corrupt data (no logical check)Hex dump of (file 12, block 1) in trace file /u01/app/oracle/diag/rdbms/maa/maa/trace/maa_m000_5
11、343.trcCorrupt block relative dba: 0x03000001 (file 12, block 1)Completely zero block found during reread访问该数据文件里的数据：sysMAA select * from l.luocs;select * from l.luocs*ERROR at line 1:ORA-01578: ORACLE data block corrupted (file # 12, block # 138)ORA-01110: data file 12: /u01/app/oracle/oradata/ltb0
12、1.dbf告警日志也会抛出相应错误信息：Hex dump of (file 12, block 2) in trace file /u01/app/oracle/diag/rdbms/maa/maa/trace/maa_ora_5220.trcCorrupt block relative dba: 0x03000002 (file 12, block 2)Completely zero block found during buffer readReading datafile /u01/app/oracle/oradata/ltb01.dbf for corruption at rdba:
13、0x03000002 (file 12, block 2)Reread (file 12, block 2) found same corrupt data (no logical check)Sun Dec 23 19:56:00 2012Corrupt Block Found TSN = 7, TSNAME = LTB RFN = 12, BLK = 2, RDBA = 50331650 OBJN = -1, OBJD = -1, OBJECT = , SUBOBJECT = SEGMENT OWNER = , SEGMENT TYPE = Errors in file /u01/app/
14、oracle/diag/rdbms/maa/maa/trace/maa_ora_5220.trc (incident=31391):ORA-01578: ORACLE data block corrupted (file # 12, block # 2)ORA-01110: data file 12: /u01/app/oracle/oradata/ltb01.dbfHex dump of (file 12, block 1) in trace file /u01/app/oracle/diag/rdbms/maa/maa/trace/maa_ora_5220.trcCorrupt block
15、 relative dba: 0x00000001 (file 12, block 1)Completely zero block found during validating datafile for block rangeReread of blocknum=1, file=/u01/app/oracle/oradata/ltb01.dbf. found same corrupt dataReread of blocknum=1, file=/u01/app/oracle/oradata/ltb01.dbf. found same corrupt dataReread of blocknum=1, file=/u01/
