[20181212]truncate的另类恢复5.txt

最新推荐文章于 2024-09-01 11:55:12 发布

cuixun7780

最新推荐文章于 2024-09-01 11:55:12 发布

阅读量138

点赞数

文章标签：数据库操作系统

[20181212]truncate的另类恢复5.txt

--//前一阵子,尝试了truncate的恢复,链接如下:

http://blog.itpub.net/267265/viewspace-2156936/

http://blog.itpub.net/267265/viewspace-2157144/

--//上午尝试了truncate reuse storage的恢复,链接如下:

http://blog.itpub.net/267265/viewspace-2284921/

--//通过修改数据字典,段头的高低水位信息以及段号,相对容易完成修复,通过redo转储获得truncate前的信息.

--//在实际上操作中很少有人使用truncate reuse storage,如果直接truncate,这样Extent Map,Auxillary Map被删除.

--//这样恢复难度相对有点大,下午尝试这样的恢复.

1.环境:

SCOTT@book> @ ver1

PORT_STRING VERSION BANNER

------------------------------ -------------- --------------------------------------------------------------------------------

x86_64/Linux 2.4.xx 11.2.0.4.0 Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production

2.建立测试环境:

SCOTT@book> create table t as select * from all_objects ;

Table created.

SCOTT@book> select object_id,data_object_id from dba_objects where owner=user and object_name='T';

OBJECT_ID DATA_OBJECT_ID

---------- --------------

90557 90557

SCOTT@book> select count(*) from t;

COUNT(*)

----------

84767

SCOTT@book> column PARTITION_NAME noprint

SCOTT@book> select * from dba_extents where owner=user and segment_name='T';

OWNER SEGMENT_NAME SEGMENT_TYPE TABLESPACE_NAME EXTENT_ID FILE_ID BLOCK_ID BYTES BLOCKS RELATIVE_FNO

----- ------------ ------------ --------------- --------- ------- -------- ------- ------ ------------

SCOTT T TABLE USERS 0 4 520 65536 8 4

SCOTT T TABLE USERS 1 4 528 65536 8 4

SCOTT T TABLE USERS 2 4 536 65536 8 4

SCOTT T TABLE USERS 3 4 544 65536 8 4

SCOTT T TABLE USERS 4 4 552 65536 8 4

SCOTT T TABLE USERS 5 4 560 65536 8 4

SCOTT T TABLE USERS 6 4 672 65536 8 4

SCOTT T TABLE USERS 7 4 680 65536 8 4

SCOTT T TABLE USERS 8 4 688 65536 8 4

SCOTT T TABLE USERS 9 4 696 65536 8 4

SCOTT T TABLE USERS 10 4 704 65536 8 4

SCOTT T TABLE USERS 11 4 712 65536 8 4

SCOTT T TABLE USERS 12 4 720 65536 8 4

SCOTT T TABLE USERS 13 4 728 65536 8 4

SCOTT T TABLE USERS 14 4 736 65536 8 4

SCOTT T TABLE USERS 15 4 744 65536 8 4

SCOTT T TABLE USERS 16 4 768 1048576 128 4

SCOTT T TABLE USERS 17 4 896 1048576 128 4

SCOTT T TABLE USERS 18 4 1024 1048576 128 4

SCOTT T TABLE USERS 19 4 1152 1048576 128 4

SCOTT T TABLE USERS 20 4 1280 1048576 128 4

SCOTT T TABLE USERS 21 4 1408 1048576 128 4

SCOTT T TABLE USERS 22 4 1536 1048576 128 4

SCOTT T TABLE USERS 23 4 1664 1048576 128 4

SCOTT T TABLE USERS 24 4 1792 1048576 128 4

25 rows selected.

SCOTT@book> select SEGMENT_NAME,HEADER_FILE,HEADER_BLOCK from dba_segments where owner=user and segment_name='T';

SEGMENT_NAME HEADER_FILE HEADER_BLOCK

-------------------- ----------- ------------

T 4 522

3.建立执行脚本:

$ cat tr.txt

column member new_value v_member

column member noprint

set numw 12

pause run alter system archive log current or alter system switch logfile;

--//12c不允许在pluggable database执行这条命令

alter system archive log current;

SELECT member FROM v$log a, v$logfile b WHERE a.group#(+) = b.group# and a.STATUS='CURRENT' and rownum=1;

column curr1 new_value v_curr1

select current_scn curr1 from v$database;

--//以下操作内容:

truncate table t ;

column curr2 new_value v_curr2

select current_scn curr2 from v$database;

prompt exec DBMS_LOGMNR.START_LOGMNR(STARTSCN => &&v_curr1 ,ENDSCN => &&v_curr2 ,OPTIONS => DBMS_LOGMNR.DICT_FROM_ONLINE_CATALOG + DBMS_LOGMNR.CONTINUOUS_MINE);

prompt alter system dump logfile '&&v_member' scn min &&v_curr1 scn max &&v_curr2;

alter system dump logfile '&&v_member' scn min &&v_curr1 scn max &&v_curr2;

SCOTT@book> @ tr.txt

run alter system archive log current or alter system switch logfile

System altered.

CURR1

------------

13815795309

Table truncated.

CURR2

------------

13815795378

exec DBMS_LOGMNR.START_LOGMNR(STARTSCN => 13815795309 ,ENDSCN => 13815795378 ,OPTIONS => DBMS_LOGMNR.DICT_FROM_ONLINE_CATALOG + DBMS_LOGMNR.CONTINUOUS_MINE)

alter system dump logfile '/mnt/ramdisk/book/redo03.log' scn min 13815795309 scn max 13815795378

System altered.

--//前面我已经做了修复高低水位的恢复.

$ egrep -A1 "^Low HWM|High HWM" /u01/app/oracle/diag/rdbms/book/book/trace/book_ora_9704.trc

Low HWM

Highwater:: 0x01000757 ext#: 24 blk#: 87 ext size: 128

High HWM

Highwater:: 0x01000757 ext#: 24 blk#: 87 ext size: 128

Low HWM

Highwater:: 0x0100020b ext#: 0 blk#: 3 ext size: 8

High HWM

Highwater:: 0x0100020b ext#: 0 blk#: 3 ext size: 8

--//通过转储日志很容易获得高低水位信息.

SCOTT@book> select object_id,data_object_id from dba_objects where owner=user and object_name='T';

OBJECT_ID DATA_OBJECT_ID

---------- --------------

90557 90558

--//段号也很容易获得.

4.恢复.

--//首先恢复数据字段:

UPDATE tab$ set dataobj#=90557 where obj#=90557;

UPDATE seg$ set hwmincr=90557 where hwmincr=90558;

update obj$ set dataobj#=90557 where obj#=90557;

commit ;

SYS@book> alter system flush buffer_cache;

System altered.

SYS@book> alter system flush shared_pool ;

System altered.

SCOTT@book> select * from t where rowid=dbms_rowid.ROWID_CREATE(1,90557,4, 523 ,0);

OWNER OBJECT_NAME SUBOBJECT_ OBJECT_ID DATA_OBJECT_ID OBJECT_TYPE CREATED LAST_DDL_TIME TIMESTAMP STATUS T G S NAMESPACE EDITION_NAME

------ ----------- ---------- ---------- -------------- ----------- ------------------- ------------------- ------------------- ------- - - - ---------- ------------

SYS ICOL$ 20 2 TABLE 2013-08-24 11:37:35 2013-08-24 11:47:37 2013-08-24:11:37:35 VALID N N N 1

--//能查询到，说明以上修改没有问题。

5.修复段头高低水位相关信息以及段号.

segment header 每个 offset 对应的含义，如下：

----------------------------------------------

offset desc

----------------------------------------------

36 total extents

40 total blocks

48 高HWM的ext#

52 高HWM的blk#(从0开始)

56 高HWM的ext size#

60 高HWM的dba地址

76 高HWM下有多少个block

92 低HWM的ext#

96 低HWM的blk#(从0开始)

100 低HWM的ext size

104 低HWM的dba地址

120 高LHWM下有多少个block

124 Level 1 BMB for High HWM block

128 Level 1 BMB for Low HWM block

213 block size

220 L2 Array start offset

224 First Level 3 BMB

228 L2 Hint for inserts

236 Last Level 1 BMB

240 Last Level II BMB

244 Last Level III BMB

264 extents

272 obj#

280 ext#为0的block_id

284 ext#为0的extent blocks

288 ext#为1的block_id

292 ext#为1的extent blocks

...... 以此类推循环

2736 aux map信息，ext#为0的L1 dba

2740 aux map信息，ext#为0的data dba

2744 aux map信息，ext#为1的L1 dba

2748 aux map信息，ext#为1的data dba

...... 以此类推循环

5192 Second Level Bitmap block DBAs

$ egrep -A1 "^Low HWM|High HWM" /u01/app/oracle/diag/rdbms/book/book/trace/book_ora_9704.trc

Low HWM

Highwater:: 0x01000757 ext#: 24 blk#: 87 ext size: 128

High HWM

Highwater:: 0x01000757 ext#: 24 blk#: 87 ext size: 128

Low HWM

Highwater:: 0x0100020b ext#: 0 blk#: 3 ext size: 8

High HWM

Highwater:: 0x0100020b ext#: 0 blk#: 3 ext size: 8

--// 25=0x19 24=0x18 87=0x57 128=0x80

--// 0x01000757 大小头对调 57 07 00 01

m /x 19 offset 36 --//total extents

m /x 19 offset 264 --//extents

--//32+16=48 开始每4各字节分别高HWM的ext#,高HWM的blk#,高HWM的ext size. 高HWM的dba地址.

--//64+28=92 开始每4各字节分别低HWM的ext#,低HWM的blk#,低HWM的ext size. 低HWM的dba地址.

m /x 18 offset 48 --//高HWM的ext#

m /x 57 offset 52 --//高HWM的blk#(从0开始)

m /x 80 offset 56 --//高HWM的ext size#

m /x 5707 offset 60 --//HWM的dba地址(低位)

m /x 0001 offset 62 --//HWM的dba地址(高位)

m /x 18 offset 92 --//低HWM的ext#

m /x 57 offset 96 --//低HWM的blk#(从0开始)

m /x 80 offset 100 --//低HWM的ext size

m /x 5707 offset 104 --//低HWM的dba地址(低位)

m /x 0001 offset 106 --//低HWM的dba地址(高位)

--//注意大小头问题。

--//90557=0x161bd 大小头对调 bd 61 01 00

m /x bd61 offset 272 --//修改obj#

m /x 0100 offset 274 --//修改obj#

--//整理如下,别写错了.^_^:

$ cat tr_bbed1.txt

set dba 4,522

m /x 19 offset 36

m /x 19 offset 264

m /x 18 offset 48

m /x 57 offset 52

m /x 80 offset 56

m /x 5707 offset 60

m /x 0001 offset 62

m /x 18 offset 92

m /x 57 offset 96

m /x 80 offset 100

m /x 5707 offset 104

m /x 0001 offset 106

m /x bd61 offset 272

m /x 0100 offset 274

sum apply

--//注:中间有1行y不是多余的,这样可以通过管道执行.因为是修改,中间有提示如下

Warning: contents of previous BIFILE will be lost. Proceed? (Y/N)

--//时就报错.如果害怕错误,使用copy &paste 执行吧.

$ cat tr_bbed1.txt | bbed parfile=bbed.par cmdfile=cmd.par

...

--//检查:

BBED> dump /v offset 48 count 16

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 48 to 63 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

18000000 57000000 80000000 57070001 l ....W.......W...

<32 bytes per line>

BBED> dump /v offset 92 count 16

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 92 to 107 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

18000000 57000000 80000000 57070001 l ....W.......W...

<32 bytes per line>

BBED> dump /v offset 272 count 8

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 272 to 279 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

bd610100 00000010 l .a......

<32 bytes per line>

6.修复段头Extent Map,Auxillary Map信息:

--//Extent Map

$ grep -i "ADD:" /u01/app/oracle/diag/rdbms/book/book/trace/book_ora_9704.trc | tee ext_map.txt

ADD: dba:0x1000700 len:128 at offset:24

ADD: dba:0x1000680 len:128 at offset:23

ADD: dba:0x1000600 len:128 at offset:22

ADD: dba:0x1000580 len:128 at offset:21

ADD: dba:0x1000500 len:128 at offset:20

ADD: dba:0x1000480 len:128 at offset:19

ADD: dba:0x1000400 len:128 at offset:18

ADD: dba:0x1000380 len:128 at offset:17

ADD: dba:0x1000300 len:128 at offset:16

ADD: dba:0x10002e8 len:8 at offset:15

ADD: dba:0x10002e0 len:8 at offset:14

ADD: dba:0x10002d8 len:8 at offset:13

ADD: dba:0x10002d0 len:8 at offset:12

ADD: dba:0x10002c8 len:8 at offset:11

ADD: dba:0x10002c0 len:8 at offset:10

ADD: dba:0x10002b8 len:8 at offset:9

ADD: dba:0x10002b0 len:8 at offset:8

ADD: dba:0x10002a8 len:8 at offset:7

ADD: dba:0x10002a0 len:8 at offset:6

ADD: dba:0x1000230 len:8 at offset:5

ADD: dba:0x1000228 len:8 at offset:4

ADD: dba:0x1000220 len:8 at offset:3

ADD: dba:0x1000218 len:8 at offset:2

ADD: dba:0x1000210 len:8 at offset:1

--//Auxillary Map

$ grep -i "ADDAXT" /u01/app/oracle/diag/rdbms/book/book/trace/book_ora_9704.trc | tee aux_map.txt

ADDAXT: offset:24 fdba:x01000700 bdba:0x01000702

ADDAXT: offset:23 fdba:x01000680 bdba:0x01000682

ADDAXT: offset:22 fdba:x01000600 bdba:0x01000602

ADDAXT: offset:21 fdba:x01000580 bdba:0x01000582

ADDAXT: offset:20 fdba:x01000500 bdba:0x01000502

ADDAXT: offset:19 fdba:x01000480 bdba:0x01000482

ADDAXT: offset:18 fdba:x01000400 bdba:0x01000402

ADDAXT: offset:17 fdba:x01000380 bdba:0x01000382

ADDAXT: offset:16 fdba:x01000300 bdba:0x01000302

ADDAXT: offset:15 fdba:x010002e0 bdba:0x010002e8

ADDAXT: offset:14 fdba:x010002e0 bdba:0x010002e1

ADDAXT: offset:13 fdba:x010002d0 bdba:0x010002d8

ADDAXT: offset:12 fdba:x010002d0 bdba:0x010002d1

ADDAXT: offset:11 fdba:x010002c0 bdba:0x010002c8

ADDAXT: offset:10 fdba:x010002c0 bdba:0x010002c1

ADDAXT: offset:9 fdba:x010002b0 bdba:0x010002b8

ADDAXT: offset:8 fdba:x010002b0 bdba:0x010002b1

ADDAXT: offset:7 fdba:x010002a0 bdba:0x010002a8

ADDAXT: offset:6 fdba:x010002a0 bdba:0x010002a1

ADDAXT: offset:5 fdba:x01000228 bdba:0x01000230

ADDAXT: offset:4 fdba:x01000228 bdba:0x01000229

ADDAXT: offset:3 fdba:x01000218 bdba:0x01000220

ADDAXT: offset:2 fdba:x01000218 bdba:0x01000219

ADDAXT: offset:1 fdba:x01000208 bdba:0x01000210

--//oracle转储还有点小问题，fdba的输出,前面的16进制少1个0.而前面Extent Map不足8位没有补0,格式不统一.

--//日志文件里面就有这些信息,剩下的事情就简单了,如何回填到原来的位置.

280 ext#为0的block_id

284 ext#为0的extent blocks

288 ext#为1的block_id --//从这里开始

292 ext#为1的extent blocks

...... 以此类推循环

2736 aux map信息，ext#为0的L1 dba

2740 aux map信息，ext#为0的data dba

2744 aux map信息，ext#为1的L1 dba --//从这里开始

2748 aux map信息，ext#为1的data dba

...... 以此类推循环

--//我看了offset 280,2736的信息还在.

BBED> dump /v offset 280 count 16

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 280 to 295 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

08020001 08000000 00000000 00000000 l ................

<32 bytes per line>

BBED> dump /v offset 2736 count 16

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 2736 to 2751 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

08020001 0b020001 00000000 00000000 l ................

<32 bytes per line>

--//仅仅offset 288,2744的信息被清空了.

--//建立如下脚本:

$ cat ext_map.sh

#! /bin/bash

cat ext_map.txt | sed -e 's/^ ADD: //' -e 's/ at//' | tr ": " "=;" | while read i

eval $i

dba=$(printf "%08x" $dba )

#echo $dba $len $offset

a1=${dba:0:2}

a2=${dba:2:2}

a3=${dba:4:2}

a4=${dba:6:2}

#echo $a1 $a2 $a3 $a4

offset=$((($offset-1)*8+288))

echo m /x $a4$a3 offset $offset

offset=$(($offset+2))

echo m /x $a2$a1 offset $offset

offset=$(($offset+2))

len=$(echo "obase=16;$len"| bc)

len=$(printf "%02s" $len )

echo m /x $len ' ' offset $offset

echo

done

$ . ext_map.sh | tee ext_map_bbed.txt

m /x 0007 offset 472

m /x 0001 offset 474

m /x 80 offset 476

...

m /x 1802 offset 296

m /x 0001 offset 298

m /x 8 offset 300

m /x 1002 offset 288

m /x 0001 offset 290

m /x 8 offset 292

$ cat aux_map.sh

#! /bin/bash

cat aux_map.txt | sed -e 's/^ ADDAXT: //' -e 's/0x//' -e 's/x//'| tr ": " "=;" | while read i

eval $i

#echo $offset $fdba $bdba

a1=${fdba:0:2}

a2=${fdba:2:2}

a3=${fdba:4:2}

a4=${fdba:6:2}

#echo $a1 $a2 $a3 $a4

b1=${bdba:0:2}

b2=${bdba:2:2}

b3=${bdba:4:2}

b4=${bdba:6:2}

#echo $b1 $b2 $b3 $b4

offset=$((($offset-1)*8+2744))

echo m /x $a4$a3 offset $offset

offset=$(($offset+2))

echo m /x $a2$a1 offset $offset

offset=$(($offset+2))

echo m /x $b4$b3 offset $offset

offset=$(($offset+2))

echo m /x $b2$b1 offset $offset

echo

done

$ . aux_map.sh | tee aux_map_bbed.txt

m /x 0007 offset 2928

m /x 0001 offset 2930

m /x 0207 offset 2932

m /x 0001 offset 2934

...

m /x 0802 offset 2744

m /x 0001 offset 2746

m /x 1002 offset 2748

m /x 0001 offset 2750

--//然后开头加入set dba 4,522,结尾加入sum apply.

--//注意不要忘记在m开始的第2行加入y,例子如下:

m /x 0007 offset 2928

m /x 0001 offset 2930

$ cat /home/oracle/xxx430/testl/ext_map_bbed.txt | bbed parfile=bbed.par cmdfile=cmd.par

$ cat /home/oracle/xxx430/testl/aux_map_bbed.txt | bbed parfile=bbed.par cmdfile=cmd.par

SCOTT@book> alter system flush buffer_cache;

System altered.

SCOTT@book> alter system flush shared_pool;

System altered.

SCOTT@book> select count(*) from t;

COUNT(*)

----------

84767

--//OK,成功了.

7.继续测试:

--//理论讲恢复到这里最好把表拷贝到别的表空间.

--//而且第一段的L1没有修改段号,这样恢复不能插入数据的,而且文件头的位图区并没有标识这些空间可用,必须修复.

SCOTT@book> insert into t select * from all_objects where rownum<=1;

insert into t select * from all_objects where rownum<=1

ERROR at line 1:

ORA-00600: internal error code, arguments: [ktspgfblk3:kcbz_objdchk], [0], [0], [1], [], [], [], [], [], [], [], []

BBED> set dba 4,520

DBA 0x01000208 (16777736 4,520)

BBED> dump /v offset 192 count 8

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 520 Offsets: 192 to 199 Dba:0x01000208

-----------------------------------------------------------------------------------------------------------

be610100 6f4e7c37 l .a..oN|7

<32 bytes per line>

--//0x161be=90558,当前dba=4,520的数据段号还是90558.

BBED> m /x bd61 offset 192

Warning: contents of previous BIFILE will be lost. Proceed? (Y/N) y

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 520 Offsets: 192 to 199 Dba:0x01000208

------------------------------------------------------------------------------------------------------------------------------------------------

bd610100 6f4e7c37

<64 bytes per line>

BBED> sum apply

Check value for File 4, Block 520:

current = 0xe970, required = 0xe970

--//修复数据文件头的位图区,参考链接:http://blog.itpub.net/267265/viewspace-2128411/

SYS@book> execute dbms_space_admin.TABLESPACE_REBUILD_BITMAPS('USERS');

PL/SQL procedure successfully completed.

SYS@book> alter system checkpoint ;

System altered.

--//依旧无法插入,使用10046跟踪发现.

PARSING IN CURSOR #139755626819840 len=56 dep=0 uid=83 oct=2 lid=83 tim=1544597100085300 hv=999879737 ad='7d60ee00' sqlid='26cwtxhxtjx1t'

insert into t select * from all_objects where rownum<=1

END OF STMT

PARSE #139755626819840:c=132980,e=133583,p=0,cr=10,cu=0,mis=1,r=0,dep=0,og=1,plh=2967680183,tim=1544597100085297

WAIT #139755626819840: nam='Disk file operations I/O' ela= 50 FileOperation=2 fileno=4 filetype=2 obj#=90557 tim=1544597100089806

WAIT #139755626819840: nam='db file sequential read' ela= 12 file#=4 block#=521 blocks=1 obj#=90557 tim=1544597100089868

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

*** SESSION ID:(274.9) 2018-12-12 14:45:00.089

OBJD MISMATCH typ=33, seg.obj=-2, diskobj=90558, dsflg=4, dsobj=90557, tid=90557, cls=9

--//dba = 4,521也需要修改数据段号.

BBED> set dba 4,521

DBA 0x01000209 (16777737 4,521)

BBED> dump /v offset 104 count 8

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 521 Offsets: 104 to 111 Dba:0x01000209

-----------------------------------------------------------------------------------------------------------

be610100 01000000 l .a......

<32 bytes per line>

BBED> m /x bd61 offset 104

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 521 Offsets: 104 to 111 Dba:0x01000209

------------------------------------------------------------------------------------------------------------------------------------------------

bd610100 01000000

<64 bytes per line>

BBED> sum apply

Check value for File 4, Block 521:

current = 0x829f, required = 0x829f

SCOTT@book> insert into t select * from all_objects where rownum<=1;

insert into t select * from all_objects where rownum<=1

ERROR at line 1:

ORA-00600: internal error code, arguments: [ktspgetmyb-1], [], [], [], [], [], [], [], [], [], [], []

--//不行!!放弃!!估计还有许多信息需要修改.

segment header 每个 offset 对应的含义，如下：

----------------------------------------------

offset desc

----------------------------------------------

36 total extents

40 total blocks

48 高HWM的ext#

52 高HWM的blk#(从0开始)

56 高HWM的ext size#

60 高HWM的dba地址

76 高HWM下有多少个block

92 低HWM的ext#

96 低HWM的blk#(从0开始)

100 低HWM的ext size

104 低HWM的dba地址

120 高LHWM下有多少个block

124 Level 1 BMB for High HWM block

128 Level 1 BMB for Low HWM block

213 block size

220 L2 Array start offset

224 First Level 3 BMB

228 L2 Hint for inserts

236 Last Level 1 BMB

240 Last Level II BMB

244 Last Level III BMB

264 extents

272 obj#

280 ext#为0的block_id

284 ext#为0的extent blocks

288 ext#为1的block_id

292 ext#为1的extent blocks

...... 以此类推循环

2736 aux map信息，ext#为0的L1 dba

2740 aux map信息，ext#为0的data dba

2744 aux map信息，ext#为1的L1 dba

2748 aux map信息，ext#为1的data dba

...... 以此类推循环

5192 Second Level Bitmap block DBAs

BBED> dump /v offset 40 count 16

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 40 to 55 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

08000000 9c0a0000 18000000 57000000 l ............W...

<32 bytes per line>

BBED> dump /v offset 76 count 16

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 76 to 91 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

00000000 00000000 00000000 00000000 l ................

<32 bytes per line>

BBED> dump /v offset 120 count 16

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 120 to 135 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

00000000 08020001 08020001 00000000 l ................

<32 bytes per line>

BBED> dump /v offset 228 count 4

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 228 to 231 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

09020001 l ....

<32 bytes per line>

BBED> dump /v offset 228 count 4

File: /mnt/ramdisk/book/users01.dbf (4)

Block: 522 Offsets: 228 to 231 Dba:0x0100020a

-----------------------------------------------------------------------------------------------------------

09020001 l ....

<32 bytes per line>

--//还有太多信息需要修改了,放弃!!

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/267265/viewspace-2284984/，如需转载，请注明出处，否则将追究法律责任。

转载于:http://blog.itpub.net/267265/viewspace-2284984/

cuixun7780

关注

0
点赞
踩
0

收藏

觉得还不错? 一键收藏
0
评论
[20181212]truncate的另类恢复5.txt

[20181212]truncate的另类恢复5.txt--//前一阵子,尝试了truncate的恢复,链接如下:http://blog.itpub.net/26726...
复制链接

扫一扫