关于alter system dump datafile n block m的疑问

2^#

发表于 2012-5-11 20:43:17

ODM TEST:

SQL> select file#,name, RFILE# from v$datafile;

FILE# NAME                                                          RFILE#
-------- ------------------------------------------------------------ ----------
   1 /s01/oradata/G10R25/datafile/o1_mf_system_7ch7d4mn_.dbf             1
   2 /s01/oradata/G10R25/datafile/o1_mf_undotbs1_7ch7d4nt_.dbf          2
   3 /s01/oradata/G10R25/datafile/o1_mf_sysaux_7ch7d4n2_.dbf             3
   4 /s01/oradata/G10R25/datafile/o1_mf_users_7ch7d4nx_.dbf             4
   5 /s01/oradata/G10R25/datafile/o1_mf_example_7ch7f9vx_.dbf             5
   6 /s01/oradata/G10R25/datafile/o1_mf_bct_test_7tgwoyyd_.dbf          6
   7 /s01/oradata/G10R25/datafile/o1_mf_bigs_7tnktkkv_.dbf             1024

7 rows selected.

SQL> oradebug setmypid;
Statement processed.

SQL> alter system dump datafile 1024 block 10;

System altered.

SQL> oradebug tracefile_name;
/s01/admin/G10R25/udump/g10r25_ora_23969.trc

Error: alter system dump datafile/tempfile: invalid input file # 1024

==> dump datafile block 应当使用 absolute FILE No 绝对文件号

SQL> update tv set t1=rpad('MACLEAN',3000,'@');

1 row updated.

SQL> select DIRTY,status from v$bh where file#=1  and block#=66930;

D STATUS
- -------
N cr
N cr
Y xcur

SQL> alter system dump datafile 1 block 66930;

System altered.

SQL> oradebug setmypid;
Statement processed.
SQL> oradebug tracefile_name;
/s01/admin/G10R25/udump/g10r25_ora_24004.trc

Start dump data blocks tsn: 0 file#: 1 minblk 66930 maxblk 66930
buffer tsn: 0 rdba: 0x00410572 (1/66930)
scn: 0x0000.001e1b06 seq: 0x01 flg: 0x00 tail: 0x1b060601
frmt: 0x02 chkval: 0x0000 type: 0x06=trans data
Hex dump of block: st=0, typ_found=1

tl: 3006 fb: --H-FL-- lb: 0x2  cc: 1
col  0: [3000]
4d 41 43 4c 45 41 4e 40 40 40 40

==》　dump datafile  获得的是Current Block

3^#

发表于 2012-5-12 10:41:23

谢谢刘大，不过我的实验结果和你的不一样：

drop table t;

create table t(processed_flag varchar2(1)) tablespace users;

insert into t values('N');

insert into t values('Y');

commit;

select * from t;

insert into t values('A');

--在另外一个会话查询t以生成一致块

select dbms_rowid.rowid_relative_fno(rowid) REL_FNO,dbms_rowid.rowid_block_number(rowid) BLOCKNO from t;

select file#,block#,status from v$bh where file# = 4 and block#=539;

FILE# BLOCK# STATUS
4 539 xcur
4 539 cr
4 539 cr

alter system dump datafile 4 block 531;

查看trace文件：
Block dump from cache:
Dump of buffer cache at level 4 for tsn=4, rdba=16777755
BH (0xe4fae4b8) file#: 4 rdba: 0x0100021b (4/539) class: 1 ba: 0xe479e000
  set: 20 pool 3 bsz: 8192 bsi: 0 sflg: 2 pwc: 29,19
  dbwrid: 0 obj: 74941 objn: 74941 tsn: 4 afn: 4 hint: f
  hash: [0xe4f94958,0x124980b10] lru: [0xe4fae6d0,0xe4fae470]
  ckptq: [NULL] fileq: [NULL] objq: [NULL]
  st: CR md: NULL tch: 1
  cr: [scn: 0x0.1a6e6f],[xid: 0x7.1a.418],[uba: 0xc0059e.ec.8],[cls: 0x0.1a6e6f],[sfl: 0x0],[lc: 0x0.1a6e6f]
  flags: only_sequential_access
  cr pin refcnt: 0 sh pin refcnt: 0
BH (0xe4f948a8) file#: 4 rdba: 0x0100021b (4/539) class: 1 ba: 0xe44e8000
  set: 19 pool 3 bsz: 8192 bsi: 0 sflg: 2 pwc: 32,19
  dbwrid: 0 obj: 74941 objn: 74941 tsn: 4 afn: 4 hint: f
  hash: [0xe3f880a8,0xe4fae568] lru: [0xe4f94ac0,0xe4f94860]
  ckptq: [NULL] fileq: [NULL] objq: [NULL]
  st: CR md: NULL tch: 1
  cr: [scn: 0x0.1a6e6e],[xid: 0x7.1a.418],[uba: 0xc0059e.ec.8],[cls: 0x0.1a6e6e],[sfl: 0x0],[lc: 0x0.1a6e6e]
  flags: only_sequential_access
  cr pin refcnt: 0 sh pin refcnt: 0
BH (0xe3f87ff8) file#: 4 rdba: 0x0100021b (4/539) class: 1 ba: 0xe3396000
  set: 17 pool 3 bsz: 8192 bsi: 0 sflg: 2 pwc: 20,19
  dbwrid: 0 obj: 74941 objn: 74941 tsn: 4 afn: 4 hint: f
  hash: [0xecfd9298,0xe4f94958] lru: [0xe3f88210,0xe3f87fb0]
  obj-flags: object_ckpt_list
  ckptq: [0xe3f877a8,0xe3f88258] fileq: [0x12320f880,0x12320f880] objq: [0x120f316e8,0x120f316e8]
  st: XCURRENT md: NULL tch: 3
  flags: buffer_dirty redo_since_read
  LRBA: [0x1a.a6c4.0] LSCN: [0x0.1a6e67] HSCN: [0x0.1a6e6f] HSUB: [1]
  cr pin refcnt: 0 sh pin refcnt: 0
Block dump from disk:
buffer tsn: 4 rdba: 0x0100021b (4/539)
scn: 0x0000.00000000 seq: 0x01 flg: 0x05 tail: 0x00000001
frmt: 0x02 chkval: 0xa41b type: 0x00=unknown
Hex dump of block: st=0, typ_found=0
Dump of memory from 0x00002B3B78047000 to 0x00002B3B78049000
2B3B78047000 0000A200 0100021B 00000000 05010000  [................]
2B3B78047010 0000A41B 00000000 00000000 00000000  [................]
2B3B78047020 00000000 00000000 00000000 00000000  [................]
      Repeat 508 times
2B3B78048FF0 00000000 00000000 00000000 00000001  [................]
Dump of memory from 0x00002B3B78047014 to 0x00002B3B78048FFC
2B3B78047010       00000000 00000000 00000000    [............]
2B3B78047020 00000000 00000000 00000000 00000000  [................]
      Repeat 508 times
2B3B78048FF0 00000000 00000000 00000000          [............]
End dump data blocks tsn: 4 file#: 4 minblk 539 maxblk 539
--可以见到这里没有找到插入的数据

alter system flush buffer_cache;
alter system dump datafile 4 block 531;

查看trace文件：
Block header dump:  0x0100021b
Object id on Block? Y
seg/obj: 0x124bd  csc: 0x00.1a6e6f  itc: 2  flg: E  typ: 1 - DATA
   brn: 0  bdba: 0x1000218 ver: 0x01 opc: 0
   inc: 0  exflg: 0

Itl          Xid                Uba       Flag  Lck       Scn/Fsc
0x01 0x0005.010.00000571  0x00c0066a.01bd.20  C--- 0  scn 0x0000.001a6e68
0x02 0x0007.01a.00000418  0x00c0059e.00ec.08  ---- 1  fsc 0x0000.00000000
bdba: 0x0100021b
data_block_dump,data header at 0x2b3b78047064
===============
tsiz: 0x1f98
hsiz: 0x18
pbl: 0x2b3b78047064
   76543210
flag=--------
ntab=1
nrow=3
frre=-1
fsbo=0x18
fseo=0x1f89
avsp=0x1f65
tosp=0x1f65
0xe:pti[0]    nrow=3  offs=0
0x12:pri[0]    offs=0x1f93
0x14:pri[1]    offs=0x1f8e
0x16:pri[2]    offs=0x1f89
block_row_dump:
tab 0, row 0, @0x1f93
tl: 5 fb: --H-FL-- lb: 0x0  cc: 1
col  0: [ 1]  4e
tab 0, row 1, @0x1f8e
tl: 5 fb: --H-FL-- lb: 0x0  cc: 1
col  0: [ 1]  59
tab 0, row 2, @0x1f89
tl: 5 fb: --H-FL-- lb: 0x2  cc: 1
col  0: [ 1]  41
end_of_block_dump
End dump data blocks tsn: 4 file#: 4 minblk 539 maxblk 539
--这里找到了插入的数据

可见alter system dump应该是从硬盘上取的数据，刘大怎么看？

4^#

发表于 2012-5-12 10:42:03

忘记说了，我的版本是11.2.0.1

5^#

发表于 2012-5-12 10:45:37

"为什么这里找到了插入的数据
可见alter system dump应该是从硬盘上取的数据？"

什么逻辑得到了这个结论？

6^#

发表于 2012-5-12 10:49:53

回复 5# 的帖子

alter system flush buffer_cache;
alter system dump datafile 4 block 531;

因为flush buffer到硬盘后就可以dump出数据，而flush之前dump出来是没有数据的，如果是从当前块读取的数据不应该会出现这种情况吧。

7^#

发表于 2012-5-12 11:01:01

SQL> select * from V$version;

BANNER
----------------------------------------------------------------
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bi
PL/SQL Release 10.2.0.5.0 - Production
CORE 10.2.0.5.0    Production
TNS for Linux: Version 10.2.0.5.0 - Production
NLSRTL Version 10.2.0.5.0 - Production

SQL> create table t(processed_flag varchar2(1)) tablespace users;

Table created.

SQL> insert into t values('N');

1 row created.

SQL> commit;

Commit complete.

SQL> alter system checkpoint;

System altered.

SQL> oradebug setospid 16684;
Oracle pid: 5, Unix process pid: 16684, image: oracle@vrh8.oracle.com (DBW0)
SQL> oradebug suspend;
Statement processed.

把 dbwr hang住这会导致  脏块无法写出

SQL> insert into t values('A');

1 row created.

SQL> select dbms_rowid.rowid_block_number(rowid),dbms_rowid.rowid_relative_fno(rowid) from t;

DBMS_ROWID.ROWID_BLOCK_NUMBER(ROWID) DBMS_ROWID.ROWID_RELATIVE_FNO(ROWID)
------------------------------------ ------------------------------------
                           197780                                  4
                           197780                                  4



SQL> alter system dump datafile 4 block 197780;

System altered.

SQL> oradebug setmypid;
Statement processed.
SQL> oradebug tracefile_name;
/s01/admin/G10R25/udump/g10r25_ora_26285.trc

tab 0, row 0, @0x1f93
tl: 5 fb: --H-FL-- lb: 0x1  cc: 1
col  0: [ 1]  4e
tab 0, row 1, @0x1f8e
tl: 5 fb: --H-FL-- lb: 0x2  cc: 1
col  0: [ 1]  41
end_of_block_dump

SQL> select dump(processed_flag,16) from t;

DUMP(PROCESSED_FLAG,16)
--------------------------------------------------------------------------------
Typ=1 Len=1: 4e
Typ=1 Len=1: 41

8^#

发表于 2012-5-12 11:12:03

谢谢刘大，那我那个实验为什么第一次dump出来没有数据的，而flush之后就可以dump出数据，会不会是版本问题？

liuxbgiant

9^#

发表于 2012-5-12 16:26:46

问刘大………？

16:15:48 SYS@orcl>create table t(processed_flag varchar2(1)) tablespace users;

表已创建。

16:15:58 SYS@orcl>
16:15:58 SYS@orcl> insert into t values('N');

已创建 1 行。

16:16:08 SYS@orcl>commit;

提交完成。

16:16:14 SYS@orcl>alter system checkpoint;

系统已更改。

16:16:21 SYS@orcl>
16:16:21 SYS@orcl> insert into t values('A');

已创建 1 行。

16:16:47 SYS@orcl>select dbms_rowid.rowid_block_number(rowid),dbms_rowid.rowid_r
elative_fno(rowid) from t;

DBMS_ROWID.ROWID_BLOCK_NUMBER(ROWID) DBMS_ROWID.ROWID_RELATIVE_FNO(ROWID)
------------------------------------ ------------------------------------
                              7348                                  4
                              7348                                  4

16:16:58 SYS@orcl>alter system dump datafile 4 block 7348;

系统已更改。

16:18:08 SYS@orcl>oradebug setmypid;
已处理的语句
16:18:21 SYS@orcl>oradebug tracefile_name;
e:\app\administrator\diag\rdbms\orcl\orcl\trace\orcl_ora_6004.trc

Block header dump:  0x01001cb4
Object id on Block? Y
seg/obj: 0x1274e  csc: 0x00.1452fb  itc: 2  flg: E  typ: 1 - DATA
   brn: 0  bdba: 0x1001cb0 ver: 0x01 opc: 0
   inc: 0  exflg: 0

Itl          Xid                Uba       Flag  Lck       Scn/Fsc
0x01 0x0002.021.000003d0  0x00c0023d.00c4.2f  --U- 1  fsc 0x0000.001452fd
0x02 0x0000.000.00000000  0x00000000.0000.00  ---- 0  fsc 0x0000.00000000
bdba: 0x01001cb4
data_block_dump,data header at 0x8941264
===============
tsiz: 0x1f98
hsiz: 0x14
pbl: 0x08941264
   76543210
flag=--------
ntab=1
nrow=1
frre=-1
fsbo=0x14
fseo=0x1f93
avsp=0x1f7b
tosp=0x1f7b
0xe:pti[0]       nrow=1       offs=0
0x12:pri[0]       offs=0x1f93
block_row_dump:
tab 0, row 0, @0x1f93
tl: 5 fb: --H-FL-- lb: 0x1  cc: 1
col  0: [ 1]  4e
end_of_block_dump
End dump data blocks tsn: 4 file#: 4 minblk 7348 maxblk 7348

16:18:33 SYS@orcl>select dump(processed_flag,16) from t;

DUMP(PROCESSED_FLAG,16)
--------------------------------------------------------------------------------

Typ=1 Len=1: 4e
Typ=1 Len=1: 41

col  0: [ 1]  4e
有此数据，但是没有41对应的dump结果？请解释…………………………

而commit 之后，dump结果：
flag=--------
ntab=1
nrow=2
frre=-1
fsbo=0x16
fseo=0x1f8e
avsp=0x1f70
tosp=0x1f70
0xe:pti[0] nrow=2 offs=0
0x12:pri[0] offs=0x1f93
0x14:pri[1] offs=0x1f8e
block_row_dump:
tab 0, row 0, @0x1f93
tl: 5 fb: --H-FL-- lb: 0x1  cc: 1
col  0: [ 1]  4e
tab 0, row 1, @0x1f8e
tl: 5 fb: --H-FL-- lb: 0x2  cc: 1
col  0: [ 1]  41
end_of_block_dump
End dump data blocks tsn: 4 file#: 4 minblk 7348 maxblk 7348

[ 本帖最后由 liuxbgiant 于 2012-5-12 16:35 编辑 ]

10^#

发表于 2012-5-12 19:46:01

FOR 11G

SQL>  select * from v$version;

BANNER
--------------------------------------------------------------------------------
Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production
PL/SQL Release 11.2.0.3.0 - Production
CORE 11.2.0.3.0    Production
TNS for Linux: Version 11.2.0.3.0 - Production
NLSRTL Version 11.2.0.3.0 - Production

SQL> alter system checkpoint;

System altered.

SQL>  select ora_rowscn from t;

ORA_ROWSCN
----------
2228203                =>scn: 0x0000.0021ffeb
2228203
2228203

SQL>  insert into t values('X');

1 row created.

SQL> commit;

Commit complete.

SQL> select ora_rowscn from t;

ORA_ROWSCN
----------
2228327
2228327
2228327
2228327

SQL>  alter system dump datafile 4 block  663;

System altered.

SQL>  oradebug setmypid;
Statement processed.

SQL>  oradebug tracefile_name
/s01/orabase/diag/rdbms/vprod/VPROD1/trace/VPROD1_ora_13521.trc

pin ktspbwh1: ktspfsrch dba 0x1000297:1 time 883713082
pin release    396 ktspbwh1: ktspfsrch dba 0x1000297:1

Block dump from cache:
Dump of buffer cache at level 4 for tsn=4 rdba=16777879
BH (0xa0f588d8) file#: 4 rdba: 0x01000297 (4/663) class: 1 ba: 0xa0cd2000
  set: 6 pool: 3 bsz: 8192 bsi: 0 sflg: 2 pwc: 28,19
  dbwrid: 0 obj: 77407 objn: 77407 tsn: 4 afn: 4 hint: f
  hash: [0xbada1a50,0xbada1a50] lru: [0xa0f58b38,0xa0f58848]
  obj-flags: object_ckpt_list
  ckptq: [0xbaf61b18,0xbaf61b18] fileq: [0xbaf61b98,0xbaf61b98] objq: [0xa0f59430,0xb42ac2c0] objaq: [0xa0f58b70,0xb42ac2a0]
  st: XCURRENT md: NULL fpin: 'ktspbwh2: ktspfmdb' tch: 10 le: 0x8cfe34b0
  flags: buffer_dirty block_written_once redo_since_read
  LRBA: [0x40.275.0] LSCN: [0x0.220065] HSCN: [0x0.220067] HSUB: [1]


Block dump from disk:

buffer tsn: 4 rdba: 0x01000297 (4/663)
scn: 0x0000.0021ffeb seq: 0x01 flg: 0x06 tail: 0xffeb0601
frmt: 0x02 chkval: 0x3966 type: 0x06=trans data
Hex dump of block: st=0, typ_found=1

0x16:pri[2]    offs=0x1f89
block_row_dump:
tab 0, row 0, @0x1f93
tl: 5 fb: --H-FL-- lb: 0x0  cc: 1
col  0: [ 1]  4e
tab 0, row 1, @0x1f8e
tl: 5 fb: --H-FL-- lb: 0x0  cc: 1
col  0: [ 1]  41
tab 0, row 2, @0x1f89
tl: 5 fb: --H-FL-- lb: 0x1  cc: 1
col  0: [ 1]  5a
end_of_block_dump

GLOBAL CACHE ELEMENT DUMP (address: 0x8cfe34b0):
  id1: 0x297 id2: 0x4 pkey: OBJ#77407 block: (4/663)
  lock: X rls: 0x0 acq: 0x0 latch: 6
  flags: 0x20 fair: 0 recovery: 0 fpin: 'ktspbwh2: ktspfmdb'
  bscn: 0x0.0 bctx: (nil) write: 0 scan: 0x0
  lcp: (nil) lnk: [NULL] lch: [0xa0f58a08,0xa0f58a08]
  seq: 10 hist: 113 424 180 113 424 180 143:0 325 352 32
  LIST OF BUFFERS LINKED TO THIS GLOBAL CACHE ELEMENT:
flg: 0x02200001 state: XCURRENT tsn: 4 tsh: 10
   addr: 0xa0f588d8 obj: 77407 cls: DATA bscn: 0x0.220067

scn: 0x0000.0021ffeb=> 2228203

summary :

11g中加强了 alter system dump datafile block 命令，  该命令目前会  DUMP block 在cache中的 buffer header信息，  on disk的完成块信息FULL block，若在RAC中还会dump该块的  GLOBAL CACHE ELEMENT。

因为只有" Block dump from disk " 从磁盘上转储的是完整的块输出，所以看到的行信息和 10g中直接从cache中读取的dirty block的情况存在区别。

wzhihua

11^#

发表于 2012-5-14 22:23:00

看我的
session 1 :
SQL> create table wzh
  2  (
  3 A VARCHAR2(10)
  4  )
  5  tablespace WZH
  6 pctfree 10
  7 initrans 1
  8 maxtrans 255
  9 storage
10 (
11    initial 64K
12    minextents 1
13    maxextents unlimited
14 );

Table created.

SQL> insert into wzh a values ('a');-------------------此处没有提交

1 row created.

SQL> select dbms_rowid.rowid_block_number(rowid) bl,dbms_rowid.rowid_relative_fno(rowid)fi  from wzh;

      bl       fi
---------- ----------
   630       6

session 2 ：
SQL> select *from wzh;

no rows selected

SQL> alter system dump datafile 6 block 630;

System altered.

trace 文件：
Dump file d:\oracle\product\10.2.0\admin\wzh\udump\wzh_ora_5404.trc
Mon May 14 22:18:49 2012
ORACLE V10.2.0.1.0 - Production vsnsta=0
vsnsql=14 vsnxtr=3
Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - Production
With the Partitioning, OLAP and Data Mining options
Windows XP Version V5.1 Service Pack 3
CPU                : 2 - type 586, 2 Physical Cores
Process Affinity : 0x00000000
Memory (Avail/Total): Ph:2343M/3571M, Ph+PgF:3006M/5459M, VA:322M/2047M
Instance name: wzh

Redo thread mounted by this instance: 1

Oracle process number: 23

Windows thread id: 5404, image: ORACLE.EXE (SHAD)

*** 2012-05-14 22:18:49.343
*** ACTION NAME:() 2012-05-14 22:18:49.328
*** MODULE NAME:(SQL*Plus) 2012-05-14 22:18:49.328
*** SERVICE NAME:(WZH) 2012-05-14 22:18:49.328
*** SESSION ID:(154.475) 2012-05-14 22:18:49.328
Start dump data blocks tsn: 7 file#: 6 minblk 630 maxblk 630
buffer tsn: 7 rdba: 0x01800276 (6/630)
scn: 0x0000.0070ccaa seq: 0x01 flg: 0x00 tail: 0xccaa0601
frmt: 0x02 chkval: 0x0000 type: 0x06=trans data
Hex dump of block: st=0, typ_found=1
Dump of memory from 0x5C726200 to 0x5C728200
5C726200 0000A206 01800276 0070CCAA 00010000  [....v.....p.....]
5C726210 00000000 00000001 0000D0D8 0070CCA5  [..............p.]
5C726220 00000000 00320002 01800271 00290003  [......2.q.....).]
5C726230 00000794 00800712 00280421 00000001  [........!.(.....]
5C726240 00000000 00000000 00000000 00000000  [................]
      Repeat 1 times
5C726260 00000000 00010100 0014FFFF 1F7B1F93  [..............{.]
5C726270 00001F7B 1F930001 00000000 00000000  [{...............]
5C726280 00000000 00000000 00000000 00000000  [................]
      Repeat 502 times
5C7281F0 00000000 2C000000 61010101 CCAA0601  [.......,...a....]
Block header dump:  0x01800276
Object id on Block? Y
seg/obj: 0xd0d8  csc: 0x00.70cca5  itc: 2  flg: E  typ: 1 - DATA
   brn: 0  bdba: 0x1800271 ver: 0x01 opc: 0
   inc: 0  exflg: 0

Itl          Xid                Uba       Flag  Lck       Scn/Fsc
0x01 0x0003.029.00000794  0x00800712.0421.28  ---- 1  fsc 0x0000.00000000
0x02 0x0000.000.00000000  0x00000000.0000.00  ---- 0  fsc 0x0000.00000000

data_block_dump,data header at 0x5c726264
===============
tsiz: 0x1f98
hsiz: 0x14
pbl: 0x5c726264
bdba: 0x01800276
   76543210
flag=--------
ntab=1
nrow=1
frre=-1
fsbo=0x14
fseo=0x1f93
avsp=0x1f7b
tosp=0x1f7b
0xe:pti[0]       nrow=1       offs=0
0x12:pri[0]       offs=0x1f93
block_row_dump:
tab 0, row 0, @0x1f93
tl: 5 fb: --H-FL-- lb: 0x1  cc: 1
col  0: [ 1]  61--------------------------------------------虽然该session中查询不到结果，可是dump文件中看到了该值。
end_of_block_dump
End dump data blocks tsn: 7 file#: 6 minblk 630 maxblk 630

"==》　dump datafile  获得的是Current Block" ???

[ 本帖最后由 wzhihua 于 2012-5-14 22:28 编辑 ]

wzhihua

12^#

发表于 2012-5-14 22:26:01

SQL> select * from v$version;

BANNER
----------------------------------------------------------------
Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - Prod
PL/SQL Release 10.2.0.1.0 - Production
CORE 10.2.0.1.0 Production
TNS for 32-bit Windows: Version 10.2.0.1.0 - Production
NLSRTL Version 10.2.0.1.0 - Production

13^#

发表于 2012-5-14 22:44:44

回复 12# 的帖子

这个上面已经说明了 10g 中是 alter system dump datafile ，会直接转储cache中的 current block

在11g 中加强了 alter system dump datafile block 命令，该命令目前会 DUMP block 在cache中的 buffer header信息， on disk的完成块信息FULL block，若在RAC中还会dump该块的 GLOBAL CACHE ELEMENT。

wzhihua

14^#

发表于 2012-5-14 23:00:45

哦，sorry ，晕了，看到current block，可是想到的还是consistent block。一直以为跟你的解释不一致。
10g中就是把cache中的current block dump出来，明白了。