ORACLE 如何检查找出损坏索引 ( Corrupt Indexes )

在Oracle数据库中如何找出损坏索引呢? 下面我们人为构造一个案例,将索引块损坏。如下案例所示:

SQL> create tablespace test_data 
  2  datafile  '/u01/app/oracle/oradata/gsp/test_data_01.dbf' 
  3  size 200M autoextend off
  4  logging
  5  segment space management auto
  6  extent management local;
 
Tablespace created.
 
SQL> create tablespace test_index 
  2  datafile  '/u01/app/oracle/oradata/gsp/test_idx_01.dbf' 
  3  size 200M autoextend off
  4  logging
  5  segment space management auto
  6  extent management local;
 
Tablespace created.
 
 
SQL> create user kerry
  2  identified by 123456
  3  default tablespace test_data;
 
User created.
 
SQL> grant connect to kerry;
SQL> grant resource to kerry;

上述脚本是创建表空间,创建用户kerry并授权,然后使用kerry账号登录数据库,构造测试数据,在TEST表上创建索引IX_TEST

SQL> show user;
USER is "KERRY"
SQL> 
SQL> CREATE TABLE TEST(ID  NUMBER(10), NAME VARCHAR2(64));
 
Table created.
 
SQL> DECLARE I  NUMBER;
  2  BEGIN
  3    FOR I IN 1..1000 LOOP
  4      INSERT INTO TEST VALUES(I, LPAD('T', 60));  
  5     END LOOP;
  6  COMMIT;
  7  END;
  8  /
 
PL/SQL procedure successfully completed.
 
 
SQL> CREATE INDEX IX_TEST ON KERRY.TEST(NAME) TABLESPACE TEST_INDEX;
 
Index created.

然后使用下面脚本找到索引段数据库文件ID,以及索引段的第一个块的块号。

SQL> show user;
USER is "SYS"
SQL> col segment_name for a32;                               
SQL> col header_file for 9999;                               
SQL> col header_block for 9999;                              
SQL> select segment_name                                     
  2   ,header_file                                      
  3   ,header_block                                     
  4   ,blocks                                           
  5  from dba_segments ds                                    
  6  where ds.owner='KERRY' and ds.segment_name='IX_TEST';   
 
SEGMENT_NAME                     HEADER_FILE HEADER_BLOCK     BLOCKS
-------------------------------- ----------- ------------ ----------
IX_TEST                                    8          130         16
 
SQL>

构造坏块的方法有不少(例如BBED等),这里我们使用RMAN下面的命令clear,可以标记数据块为corrupt,标记数据文件8中130号数据块为坏块。

[oracle@DB-Server ~]$ rman target /
 
Recovery Manager: Release 11.2.0.1.0 - Production on Thu Sep 13 17:41:05 2018
 
Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.
 
connected to target database: GSP (DBID=644393201)
 
RMAN> recover datafile 8 block 130 clear;
 
Starting recover at 13-SEP-18
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=12 device type=DISK
Finished recover at 13-SEP-18
 
RMAN>

那么我们先来看看使用那些方法验证索引损坏了,测试验证一下看看是否可行。 

1:使用ANALYZE分析验证索引结构

[oracle@DB-Server ~]$ sqlplus / as sysdba
 
SQL*Plus: Release 11.2.0.1.0 Production on Thu Sep 13 17:42:03 2018
 
Copyright (c) 1982, 2009, Oracle.  All rights reserved.
 
 
Connected to:
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
 
SQL> analyze index kerry.ix_test validate structure;
analyze index kerry.ix_test validate structure
*
ERROR at line 1:
ORA-01578: ORACLE data block corrupted (file # 8, block # 130)
ORA-01110: data file 8: '/u01/app/oracle/oradata/gsp/test_idx_01.dbf'

如上截图所示,如果索引损坏(Corrupt Index),那么使用analyze index validate structure就会报错。要检查整个数据库所有的损坏索引(Corrupt Indexes)的话,就可以借助下面脚本: 

spool analy_index.sql
SET PAGESIZE 50000;
SELECT
   'ANALYZE INDEX  ' || OWNER || '.' || INDEX_NAME|| ' VALIDATE STRUCTURE;' FROM DBA_INDEXES;
 
spool off;
 
@analy_index.sql

2:使用系统视图v$database_block_corruption查看损坏索引

如下所示,我们使用这个脚本来查看出现坏块的索引,发现这个脚本无法找出坏块索引。

set pagesize 50 linesize 170
col segment_name format a30
col partition_name format a30
SELECT DISTINCT file#, 
       segment_name, 
       segment_type, 
       tablespace_name, 
       partition_name 
FROM   dba_extents a, 
       v$database_block_corruption b 
WHERE  a.file_id = b.file# 
       AND a.block_id <= b.block# 
       AND a.block_id + a.blocks >= b.block#;

原因分析如下,视图v$database_block_corruption中有坏块记录,但是我们将索引段的第一个块标记为坏块后,在dba_extents中没有该索引段的记录了。所以这种情况下的索引损坏,这个SQL语句根本无法找出坏块索引。

SQL> SELECT file_id, 
  2         segment_name, 
  3         segment_type 
  4  FROM   dba_extents 
  5  WHERE  file_id = 8 ;
 
no rows selected
 
SQL> SELECT file_id, 
  2         segment_name, 
  3         segment_type 
  4 FROM   dba_extents 
  5  WHERE  owner = 'KERRY';
 
   FILE_ID SEGMENT_NAME                     SEGMENT_TYPE
---------- -------------------------------- ------------------
         7 TEST                             TABLE
         7 TEST                             TABLE
 
SQL>

因为一个段的第一个区的第一个块是FIRST LEVEL BITMAP BLOCK,第二个块是SECOND LEVEL BITMAP BLOCK,这两个块是用来管理free block的,第三个块是PAGETABLE SEGMENT HEADER,这个块才是segment里的HEADER_BLOCK,再后面的块就是用来记录数据的。关于这些知识,可以参考我博客ORACLE关于段的HEADER_BLOCK的一点浅析。我们最上面的例子,是将第一个块构造为坏块,所以导致上面SQL无法查出。 我们重新构造案例,如我们将索引段的数据块构造为坏块,例如下面,将块号148人为构造坏块。那么此时这个脚本就能找出坏块索引了。所以综上述实验可以看出,这个脚本查找坏块索引是有条件的,要看索引段损坏的块是什么类型

SQL> SELECT FILE_ID, 
  2         BLOCK_ID, 
  3         BLOCKS 
FROM   DBA_EXTENTS 
  4    5  WHERE  OWNER ='&OWNER'
  6       AND SEGMENT_NAME = '&TABLE_NAME'; 
Enter value for owner: KERRY
old   5: WHERE  OWNER ='&OWNER'
new   5: WHERE  OWNER ='KERRY'
Enter value for table_name: IX_TEST
old   6:      AND SEGMENT_NAME = '&TABLE_NAME'
new   6:      AND SEGMENT_NAME = 'IX_TEST'
 
   FILE_ID   BLOCK_ID     BLOCKS
---------- ---------- ----------
         8        144          8
         8        152          8
 
SQL> SELECT HEADER_FILE
  2       , HEADER_BLOCK
  3       , BYTES
  4       , BLOCKS
  5       , EXTENTS 
FROM DBA_SEGMENTS 
  6    7  WHERE OWNER='&OWNER' AND SEGMENT_NAME='&SEGMENT_NAME';
Enter value for owner: KERRY
Enter value for segment_name: IX_TEST
old   7: WHERE OWNER='&OWNER' AND SEGMENT_NAME='&SEGMENT_NAME'
new   7: WHERE OWNER='KERRY' AND SEGMENT_NAME='IX_TEST'
 
HEADER_FILE HEADER_BLOCK      BYTES     BLOCKS    EXTENTS
----------- ------------ ---------- ---------- ----------
          8          146     131072         16          2
 
SQL>

RMAN> recover datafile 8 block 148 clear;



相关文章

发表评论

Comment form

(*) 表示必填项

还没有评论。

跳到底部
返回顶部