Oraclecorruptblock（坏块）详解(5)

event= “10231 trace name context forever, level 10”

之前发过一个帖子,当我是用stm32的spi1(主)和spi2(从)互相通信并将结果通过串口发出的时候,出现的现象是spi1发送spi2接收正常,但是spi2发送spi1接收却接收不到数据,在群里找大神请教了一下问题所在,解释的原因是因为当spi1先发送,那么当spi1发送完毕后,sclk是处于空闲状态的,这时候spi2是无法发送出数据的.我把程序改为spi2先发,然后spi1再发,然后spi1接收,然后spi2接收,这样一来程序就接受正常了.通过这个实验得出的结果是:主spi的sclk只有在发送的时候是有波形的,但是当接收的时候,主设备sclk是没有波形的,我已经在实验中用示波器看过,while循环中只有spi接受函数的时候sclk没有变化一直是低电平,while循环中只有spi发送函数的时候sclk才有波形,所以如果让spi2(从)可以正常的发送给spi1(主),必须spi2(从)先发送然后spi1(主)在发送,这样主就可以提供给从一个时钟信号用来发送.但是疑问又来了,当stm32和的，如果坏块的位置已知，可以按照坏块的范围设置这两个值，节省时间，还不伤盘哦~。

方法二：ROWID method

Extractthe data that does not belong to the corrupted block using ROWIDs. Every row inevery table in an Oracle database has a ROWID column which is usually not displayedunless SELECTed explicitly.

无论是客户基本信息、客户产品信息、客户交易流水等结构化数据、社交网站网页、工商局网页等半结构化数据，还是客户财报、人行征信报告等非结构化的数据，通过数据采集的方式被整合进了中，通过属性抽取、关系抽取、实体抽取等信息获取流程形成知识。在大数据平台上，集团通过建设全媒体内容资源库，整合传统媒体读者资源和互联网平台活跃用户数据，采用大数据的智能方式对资源进行抽取，运用到新闻业务中。优点：通过创建唯一性索引，可以保证中每一行数据的唯一性，2，可以大大加快数据的检索速度，这也是创建索引的最主要的原因，3.可以加速表与表之间的连接，特别是在实现数据的参考完整性。

Oracle Rowid 介绍

（1）通过函数构建ROWID

函数参数如下：

function ROWID_CREATE(rowid_type IN number,

object_number INnumber,

relative_fno IN number,

block_number IN number,

row_number IN number)

return ROWID;

-- rowid_type - type(restricted=0/extended=1)

-- object_number - data objectnumber

-- relative_fno - relative filenumber

-- block_number - block numberin this file

-- row_number - row number inthis block

这些参数可以通过如下方式获取：

ROWID_TYPE:

Thisis 1 because we are using the extended rowid format.

RELATIVE_FNO:

Thisshould have been available when you came to this article. It can also be foundfrom the DBA_EXTENTS view given the absolute file number and block number ofthe corrupt block:

SQL>SELECTtablespace_name, relative_fno, segment_type, owner, segment_name, partition_nameFROM dba_extents WHERE file_id = <AFN> AND <BL> between block_id and block_id + blocks-1;

OBJECT_NUMBER:

For a non-partitioned table, select the DATA_OBJECT_ID from DBA_OBJECTS for theproblem table:

sql>select ename, sal, b.c from emp a,(select job,avg(sal) c from emp group by job) b where a.sal > b.c and a.job =b.job。select sql server object explorer from the visual studio view menu and right-click the sql server object in the window that appears. select add sql server from the pop-up menu, as shown in figure 13-2.。select sql from (select * from sqlite_master union all select * from sqlite_temp_master)where tbl_name like '%s' and type。

Note that a partitionedtable has an object number for each partition.

--注意，对于分区表，每个分区有一个对象id：

Select the DATA_OBJECT_ID from DBA_OBJECTSthus:

SQL>SELECTdata_object_id FROM dba_objects WHERE object_name = '<TABLE-NAME>' ANDowner = '<TABLE-OWNER>' AND subobject_name = '<PARTITION-NAME>' ;

相对与普通表，普通表多了一个subojbect_name 条件。

BLOCK_NUMBER andROW_NUMBER:

Theblock number of the corrupt block should be available before you came to thisarticle. (Eg: It is reported in an ORA-1578 error, or as a Page Number byDBVerify).

// the greater the number of tokens, the greater chance to forging a new block。-227 ora00227: corrupt block detected in controlfile: (block , # blocks )。ora-10563: test recovery had to corrupt data block (file# , block# ) in order to proceed。

（2）创建ROWID

You can now create the rowid strings to use in a predicate thus:

现在可以创建rowid：

The "LOW_RID"is the lowest rowid INSIDE the corrupt block:

SQL>SELECT dbms_rowid.rowid_create(1,<DATA_OBJECT_ID>,<RFN>,<BL>,0)LOW_RID from DUAL;

The "HI_RID" isthe first rowid AFTER the corrupt block:

SQL>SELECT dbms_rowid.rowid_create(1,<DATA_OBJECT_ID>,<RFN>,<BL>+1,0)HI_RID from DUAL;

（3）重建数据

Itis now possible to use CREATE TABLE AS SELECT or INSERT ... SELECT to get datawithout accessing the corrupt block using a query of the form:

根据刚才查询的rowid，跳过corrupt block来进行createtable 或者insert 操作：

1),delete from tablename where id not in (select max(id) from tablename group by col1,col2,...)。select ename,sal,rowid into v_ename,v_sal,v_rowid from emp where empno=7788。='sqlite_sequence' and rootpage>0select 'create index vacuum_db.' || substr(sql,14) from sqlite_master where sql like 'create index %'select 'create unique index vacuum_db.' || substr(sql,21) from sqlite_master where sql like 'create unique index %'select 'insert into vacuum_db.' || quote(name) || ' select * from ' || quote(name) || '。

SQL>INSERT INTO salvage_table SELECT /*+ROWID(A) */ * FROM <owner.tablename> A WHERE rowid >='<hi_rid>';

注意hint里的A是表的别名。

Fora table partition then only the problem partition need be selected from byusing the PARTITION(xxx) option in the FROM clause:

对于分区表，仅需要对问题分区进行处理：

1),delete from tablename where id not in (select max(id) from tablename group by col1,col2,...)。sql>select'invalidobt',owner,object_name,object_type from dba_objects wherestatus='invalid' and owner in (select username from dba_users where account_status='open') 。='sqlite_sequence' and rootpage>0select 'create index vacuum_db.' || substr(sql,14) from sqlite_master where sql like 'create index %'select 'create unique index vacuum_db.' || substr(sql,21) from sqlite_master where sql like 'create unique index %'select 'insert into vacuum_db.' || quote(name) || ' select * from ' || quote(name) || '。

本文来自电脑杂谈，转载请注明本文网址：
https://www.pc-fly.com/a/jisuanjixue/article-117663-5.html