RMAN 实现数据块恢复
试用Rman可以实现数据块级的数据恢复，在传统恢复手段中即某个数据文件的一个数据块被损坏，就造成整个数据文件无法试用，此时必须通过备份恢复整个数据文件。显然这样的方法会会时间较长，而RMAN实现块级恢复，如果某个数据文件的数据损坏，通过数据文件的完整备份就可以恢复数据块。
案例：
数据库是一个单实例Oracle数据库，该库的总大小有700G。
存储设备使用华为存储，备份设备使用希捷3T的移动硬盘。该数据库无DG无OGG。备份策略为每周六0点全库备份，周三0点1级差异备份其余时间为每天0点做2级差异备份。全库备份大小为500G左右。
2.2 故障情况
本次故障原因是INSPUROA用户在查询EDOC_BASE_WORKFLOW表出现报错。提示故障坏块为datafile 5。报错信息取至alter日志如下：
Reading datafile '/oradata/datafiles/oadb/oa01.dbf' for corruption at rdba: 0x016d4dd5 (file 5, block 2969045)
Reread (file 5, block 2969045) found same corrupt data (no logical check)
Tue Aug 18 10:53:51 2015
Corrupt Block Found
TSN = 6, TSNAME = OA
RFN = 5, BLK = 2969045, RDBA = 23940565
OBJN = 95690, OBJD = 95690, OBJECT = EDOC_BASE_WORKFLOW, SUBOBJECT =
SEGMENT OWNER = INSPUROA, SEGMENT TYPE = Table Segment
Tue Aug 18 10:55:03 2015
Hex dump of (file 5, block 2969045) in trace file /u01/app/oracle/diag/rdbms/oadb/oadb/trace/oadb_ora_4565.trc
Corrupt block relative dba: 0x016d4dd5 (file 5, block 2969045)
Bad header found during buffer read
Data in bad block:
type: 117 format: 0 rdba: 0x20206b73
last change scn: 0x2020.20202020 seq: 0x20 flg: 0x20
spare1: 0x64 spare2: 0x69 spare3: 0x0
consistency value in tail: 0x4d240601
check value in block header: 0x5f49
block checksum disabled
Reading datafile '/oradata/datafiles/oadb/oa01.dbf' for corruption at rdba: 0x016d4dd5 (file 5, block 2969045)
Reread (file 5, block 2969045) found same corrupt data (no logical check)
Tue Aug 18 10:55:03 2015
Corrupt Block Found
TSN = 6, TSNAME = OA
RFN = 5, BLK = 2969045, RDBA = 23940565
OBJN = 95690, OBJD = 95690, OBJECT = EDOC_BASE_WORKFLOW, SUBOBJECT =
SEGMENT OWNER = INSPUROA, SEGMENT TYPE = Table Segment
Tue Aug 18 10:57:29 2015
Hex dump of (file 5, block 2969045) in trace file /u01/app/oracle/diag/rdbms/oadb/oadb/trace/oadb_ora_21708.trc
Corrupt block relative dba: 0x016d4dd5 (file 5, block 2969045)
Bad header found during buffer read
Data in bad block:
type: 117 format: 0 rdba: 0x20206b73
last change scn: 0x2020.20202020 seq: 0x20 flg: 0x20
spare1: 0x64 spare2: 0x69 spare3: 0x0
consistency value in tail: 0x4d240601
check value in block header: 0x5f49
block checksum disabled
分析原因
观察存储，无报错警告，初步怀疑逻辑坏块
执行修复
根据报错信息
Reading datafile '/oradata/datafiles/oadb/oa01.dbf' for corruption at rdba: 0x016d4dd5 (file 5, block 2969045)
Reread (file 5, block 2969045) found same corrupt data (no logical check)
Corrupt Block Found
TSN = 6, TSNAME = OA
RFN = 5, BLK = 2969045, RDBA = 23940565
OBJN = 95690, OBJD = 95690, OBJECT = EDOC_BASE_WORKFLOW, SUBOBJECT =
SEGMENT OWNER = INSPUROA, SEGMENT TYPE = Table Segment
确定数据文件 datafile 5，oa01.dbf出现坏块现象
查看坏块信息：
SQL> select * from v$database_block_corruption;
FILE# BLOCK# BLOCKS CORRUPTION_CHANGE# CORRUPTIO
---------- ---------- ---------- ------------------ ---------
5 2969045 1 0 CORRUPT
确定坏块为2969045号
检查备份日志（增量，全量）是否完整备份
检查备份datafile 5 是否完整
RMAN> backup validate datafile 5;
Starting backup at 18-AUG-15
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=982 device type=DISK
channel ORA_DISK_1: starting full datafile backup set
channel ORA_DISK_1: specifying datafile(s) in backup set
input datafile file number=00005 name=/oradata/datafiles/oadb/oa01.dbf
channel ORA_DISK_1: backup set complete, elapsed time: 00:05:35
List of Datafiles
=================
File Status Marked Corrupt Empty Blocks Blocks Examined High SCN
---- ------ -------------- ------------ --------------- ----------
5 FAILED 0 1840 4190720 9484751217293
File Name: /oradata/datafiles/oadb/oa01.dbf
Block Type Blocks Failing Blocks Processed
---------- -------------- ----------------
Data 0 2842014
Index 0 182983
Other 1 1163883
validate found one or more corrupt blocks
See trace file /u01/app/oracle/diag/rdbms/oadb/oadb/trace/oadb_ora_13513.trc for details
Finished backup at 18-AUG-15
执行修复
使用RMAN工具
RMAN> blockrecover datafile 5 block 2969045;
Starting recover at 18-AUG-15
using channel ORA_DISK_1