数据库环境Oracle 19.3.0.0.0
与checkpoint相关的SCN号有四个,其中三个存在控制文件中,一个存放在数据文件头中。
这四个分别是:
1.System Checkpoint SCN
当checkpoint完成后,ORACLE将System Checkpoint SCN号存放在控制文件中。
我们可以通过下面SQL语句查询:
select checkpoint_change# from v$database;
2.Datafile Checkpoint SCN
当checkpoint完成后,ORACLE将Datafile Checkpoint SCN号存放在控制文件中。
我们可以通过下面SQL语句查询所有数据文件的Datafile Checkpoinnt SCN号。
select name,checkpoint_change# fromv$datafile;
3.Start SCN号
ORACLE将Start SCN号存放在数据文件头中。
这个SCN用于检查数据库启动过程是否需要做media recovery.
我们可以通过以下SQL语句查询:
select name,checkpoint_change# fromv$datafile_header;
4.End SCN号
ORACLE将End SCN号存放在控制文件中。
这个SCN号用于检查数据库启动过程是否需要做instance recovery.
我们可以通过以下SQL语句查询:
select name,last_change# from v$datafile;
在数据库正常运行的情况下,对可读写的,online的数据文件,该SCN号为NULL.
在数据库启动过程中,当System Checkpoint SCN、Datafile Checkpoint SCN和Start SCN号都相同时,数据库可以正常启动,不需要做media recovery.三者当中有一个不同时,则需要做media recovery.
ORACLE在启动过程中首先检查是否需要media recovery,然后再检查是否需要instance recovery.
如果数据库的正常关闭的话,将会触发一个checkpoint,同时将数据文件的END SCN号设置为相应数据文件的Start SCN号。当数据库启动时,发现它们是一致的,则不需要做instance recovery。在数据库正常启动后,ORACLE会将END SCN号设置为NULL.
如果数据库异常关闭的话,则END SCN号将为NULL.则需要做instance recovery.
bbed 工具分32位和64位,并且从11g开始不提供bbed,故需要从10g的库找下面4个文件,上传到19c环境重新编译。
ls /home/oracle/bbed/bbed
bbedus.msb bbedus.msg sbbdpt.o ssbbded.o
cp s* /u01/app/oracle/product/19.9.0/dbhome_1/rdbms/lib/
cp bbedus.ms* /u01/app/oracle/product/19.9.0/dbhome_1/rdbms/mesg/
make -f /u01/app/oracle/product/19.9.0/dbhome_1/rdbms/lib/ins_rdbms.mk BBED=$ORACLE_HOME/bin/bbed $ORACLE_HOME/bin/bbed
bbed 默认密码为:blockedit,可以编辑参数文件把密码、BLOCKSIZE 、FILENAME 等进去,省的 后面设置,也可以手动一项一项设置。
[oracle@test1 ~]$ bbed
Password:
BBED: Release 2.0.0.0.0 - Limited Production on Sun Feb 14 17:23:30 2021
Copyright (c) 1982, 2019, Oracle and/or its affiliates. All rights reserved.
************* !!! For Oracle Internal Use only !!! ***************
BBED> info all;
File# Name Size(blks)
----- ---- ----------
BBED> help all;
SET DBA [ dba | file#, block# ]
SET FILENAME 'filename'
SET FILE file#
SET BLOCK [+/-]block#
SET OFFSET [ [+/-]byte offset | symbol | *symbol ]
SET BLOCKSIZE bytes
SET LIST[FILE] 'filename'
SET WIDTH character_count
SET COUNT bytes_to_display
SET IBASE [ HEX | OCT | DEC ]
SET OBASE [ HEX | OCT | DEC ]
SET MODE [ BROWSE | EDIT ]
SET SPOOL [ Y | N ]
SHOW [ <SET parameter> | ALL ]
INFO
MAP[/v] [ DBA | FILENAME | FILE | BLOCK ]
DUMP[/v] [ DBA | FILENAME | FILE | BLOCK | OFFSET | COUNT ]
PRINT[/x|d|u|o|c] [ DBA | FILE | FILENAME | BLOCK | OFFSET | symbol | *symbol ]
EXAMINE[/Nuf] [ DBA | FILE | FILENAME | BLOCK | OFFSET | symbol | *symbol ]
</Nuf>:
N - a number which specifies a repeat count.
u - a letter which specifies a unit size:
b - b1, ub1 (byte)
h - b2, ub2 (half-word)
w - b4, ub4(word)
r - Oracle table/index row
f - a letter which specifies a display format:
x - hexadecimal
d - decimal
u - unsigned decimal
o - octal
c - character (native)
n - Oracle number
t - Oracle date
i - Oracle rowid
FIND[/x|d|u|o|c] numeric/character string [ TOP | CURR ]
COPY [ DBA | FILE | FILENAME | BLOCK ] TO [ DBA | FILE | FILENAME | BLOCK ]
MODIFY[/x|d|u|o|c] numeric/character string
[ DBA | FILE | FILENAME | BLOCK | OFFSET | symbol | *symbol ]
ASSIGN[/x|d|u|o] <target spec>=<source spec>
<target spec> : [ DBA | FILE | FILENAME | BLOCK | OFFSET | symbol | *symbol ]
<source spec> : [ value | <target spec options> ]
SUM [ DBA | FILE | FILENAME | BLOCK ] [ APPLY ]
PUSH [ DBA | FILE | FILENAME | BLOCK | OFFSET ]
POP [ALL]
REVERT [ DBA | FILE | FILENAME | BLOCK ]
UNDO
HELP [ <bbed command> | ALL ]
VERIFY [ DBA | FILE | FILENAME | BLOCK ]
CORRUPT [ DBA | FILE | FILENAME | BLOCK ]
BBED>
目前还不知道怎么使用bbed读取asm磁盘组中的数据文件,所以需要文件在文件系统中,asm磁盘组中的数据文件需要copy出来,修改好后,再copy回去,用rman还是cp都可以。
SQL> create tablespace test datafile '/home/oracle/test001.dbf' size 10m autoextend on next 1m;
Tablespace created.
SQL> create table t1 tablespace test as select rownum n ,'wen' a from dual connect by rownum<20;
Table created.
SQL> select * from t1;
N A
---------- ---
1 wen
2 wen
3 wen
4 wen
5 wen
6 wen
7 wen
8 wen
9 wen
10 wen
11 wen
12 wen
13 wen
14 wen
15 wen
16 wen
17 wen
18 wen
19 wen
19 rows selected.
SQL> SELECT a.FILE#, a.NAME,a.CHECKPOINT_CHANGE#,a.LAST_CHANGE#,status FROM v$datafile a;
FILE# NAME CHECKPOINT_CHANGE# LAST_CHANGE# STATUS
---------- ----------------------------- ------------------ ------------ -------
1 +DG1/TEST/system01.dbf 2307067 SYSTEM
2 /home/oracle/test001.dbf 2307067 ONLINE
3 +DG1/TEST/sysaux01.dbf 2307067 ONLINE
4 +DG1/TEST/undotbs01.dbf 2307067 ONLINE
5 +DG1/TEST/undotbs02.dbf 2307067 ONLINE
7 +DG1/TEST/users01.dbf 2307067 ONLINE
SQL> delete from t1 where n<17;
16 rows deleted.
SQL> commit;
Commit complete.
SQL> select * from t1;
N A
---------- ---
17 wen
18 wen
19 wen
SQL> select CHECKPOINT_CHANGE# from v$datafile_header;
CHECKPOINT_CHANGE#
------------------
2810204
2405924 ---file /home/oracle/test001.dbf
2810204
2810204
2810204
2810204
6 rows selected.
SQL> SELECT a.FILE#, a.NAME,a.CHECKPOINT_CHANGE#,a.LAST_CHANGE#,status FROM v$datafile a;
FILE# NAME CHECKPOINT_CHANGE# LAST_CHANGE# STATUS
---------- ----------------------- ------------------ ------------ -------
1 +DG1/TEST/system01.dbf 2810204 2810204 SYSTEM
2 /home/oracle/test001.dbf 2810204 2810204 ONLINE
3 +DG1/TEST/sysaux01.dbf 2810204 2810204 ONLINE
4 +DG1/TEST/undotbs01.dbf 2810204 2810204 ONLINE
5 +DG1/TEST/undotbs02.dbf 2810204 2810204 ONLINE
7 +DG1/TEST/users01.dbf 2810204 2810204 ONLINE
6 rows selected.
SQL> alter database open;
alter database open
*
ERROR at line 1:
ORA-01113: file 2 needs media recovery
ORA-01110: data file 2: '/home/oracle/test001.dbf'
BBED> info
File# Name Size(blks)
----- ---- ----------
1 /home//oracle/test001.dbf 0
2 /tmp/USERS.260.1064444539 0
BBED> p kcvfhckp
struct kcvfhckp, 36 bytes @484 ## @ 后面是偏移量
struct kcvcpscn, 8 bytes @484
ub4 kscnbas @484 0x0024b624
ub2 kscnwrp @488 0x8000
ub2 kscnwrp2 @490 0x0000
ub4 kcvcptim @492 0x3f72dc0e
ub2 kcvcpthr @496 0x0001
union u, 12 bytes @500
struct kcvcprba, 12 bytes @500
ub4 kcrbaseq @500 0x00000007
ub4 kcrbabno @504 0x0000108e
ub2 kcrbabof @508 0x0010
ub1 kcvcpetb[0] @512 0x06
ub1 kcvcpetb[1] @513 0x00
ub1 kcvcpetb[2] @514 0x00
ub1 kcvcpetb[3] @515 0x00
ub1 kcvcpetb[4] @516 0x00
ub1 kcvcpetb[5] @517 0x00
ub1 kcvcpetb[6] @518 0x00
ub1 kcvcpetb[7] @519 0x00
BBED> d /v dba 1,1 offset 484 count 16
File: /home//oracle/test001.dbf (1)
Block: 1 Offsets: 484 to 499 Dba:0x00400001
-------------------------------------------------------
24b62400 00800000 0edc723f 01000000 l $¶$......۲?....
2405924的16进制为:24b624
2810204的16进制为:2ae15c
注意操作系统的大小端。Linux下字节序,存储是反的,这个困惑了我好久
BBED> modify /x 5ce12a dba 1,1 offset 484
File: /home//oracle/test001.dbf (1)
Block: 1 Offsets: 484 to 499 Dba:0x00400001
------------------------------------------------------------------------
5ce12a00 00800000 0edc723f 01000000
<32 bytes per line>
BBED> sum apply
Check value for File 1, Block 1:
current = 0x4737, required = 0x4737
SQL> r
1* select CHECKPOINT_CHANGE# from v$datafile_header
CHECKPOINT_CHANGE#
------------------
2810204
2810204
2810204
2810204
2810204
2810204
6 rows selected
SQL> alter database open;
alter database open
*
ERROR at line 1:
ORA-01113: file 2 needs media recovery
ORA-01110: data file 2: '/home/oracle/test001.dbf'
SQL> alter database recover;
Database altered.
SQL> alter database open;
Database altered.
SQL> select * from t1;
N A
---------- ---
1 wen
2 wen
3 wen
4 wen
5 wen
6 wen
7 wen
8 wen
9 wen
10 wen
11 wen
N A
---------- ---
12 wen
13 wen
14 wen
15 wen
16 wen
17 wen
18 wen
19 wen
19 rows selected.
可以看到数据已经恢复到了关库冷备时候的数据,后面的删除语句没有生效,已经通过修改文件头的scn,跳过了删除据的redo,也说明这种方法可能会使数据不一致,适用于历史表空间数据,其他情况谨慎使用。