_minimum_giga_scn的含义(摘抄自dbsnake)

_minimum_giga_scn=n的含义是把datafile header里的checkpoint SCN往前推进到nG，但请注意，只有在datafile header的checkpoint SCN小于nG的时候才会用到这个隐含参数，反之则oracle会置这个隐含参数于不顾。

 

我们来看一个实际的例子：

BBED> set file 1 block 1

        FILE#           1

        BLOCK#          1

 

BBED> p kcvfhckp.kcvcpscn

struct kcvcpscn, 8 bytes                    @140    

   ub4 kscnbas                              @140      0x000ed811

   ub2 kscnwrp                              @144      0x0000

 

BBED> set file 1 block 122

        FILE#           1

        BLOCK#          122

 

BBED> p kcbh 

struct kcbh, 20 bytes                       @0      

   ub1 type_kcbh                            @0        0x06

   ub1 frmt_kcbh                            @1        0x02

   ub1 spare1_kcbh                          @2        0x00

   ub1 spare2_kcbh                          @3        0x00

   ub4 rdba_kcbh                            @4        0x0040007a

   ub4 bas_kcbh                             @8        0x000f2631

   ub2 wrp_kcbh                             @12       0x0000

   ub1 seq_kcbh                             @14       0x01

   ub1 flg_kcbh                             @15       0x06 (KCBHFDLC, KCBHFCKV)

   ub2 chkval_kcbh                          @16       0x686f

   ub2 spare3_kcbh                          @18       0x0000

 

BBED> p ktbbh

struct ktbbh, 48 bytes                      @20     

   ub1 ktbbhtyp                             @20       0x01 (KDDBTDATA)

   union ktbbhsid, 4 bytes                  @24     

      ub4 ktbbhsg1                          @24       0x00000012

      ub4 ktbbhod1                          @24       0x00000012

   struct ktbbhcsc, 8 bytes                 @28     

      ub4 kscnbas                           @28       0x000f2630

      ub2 kscnwrp                           @32       0x0000

   b2 ktbbhict                              @36       1

   ub1 ktbbhflg                             @38       0x02 (NONE)

   ub1 ktbbhfsl                             @39       0x00

   ub4 ktbbhfnx                             @40       0x00000000

   struct ktbbhitl[0], 24 bytes             @44     

      struct ktbitxid, 8 bytes              @44     

         ub2 kxidusn                        @44       0x0010

         ub2 kxidslt                        @46       0x002a

         ub4 kxidsqn                        @48       0x0000010a

      struct ktbituba, 8 bytes              @52     

         ub4 kubadba                        @52       0x02c00444

         ub2 kubaseq                        @56       0x002a

         ub1 kubarec                        @58       0x0e

      ub2 ktbitflg                          @60       0x2001 (KTBFUPB)

      union _ktbitun, 2 bytes               @62     

         b2 _ktbitfsc                       @62       0

         ub2 _ktbitwrp                      @62       0x0000

      ub4 ktbitbas                          @64       0x000f2631

 

现在这个库是启不起来的：

SQL_testdb>startup pfile=/dras20/testdb/inittestdb.ora

ORACLE instance started.

 

Total System Global Area  504858456 bytes

Fixed Size                   743256 bytes

Variable Size             285212672 bytes

Database Buffers          218103808 bytes

Redo Buffers                 798720 bytes

Database mounted.

ORA-01092: ORACLE instance terminated. Disconnection forced

 

[P550_04_LA:oracle@:/cadrasu01/app/oracle/admin/testdb/bdump]#tail -n 200 alert_testdb.log

CKPT started with pid=5

......省略显示部分内容

SMON: enabling cache recovery

Mon Dec  6 14:24:59 2010

Errors in file /cadrasu01/app/oracle/admin/testdb/udump/testdb_ora_1851540.trc:

ORA-00600: internal error code, arguments: [4000], [16], [], [], [], [], [], []

Mon Dec  6 14:25:00 2010

Errors in file /cadrasu01/app/oracle/admin/testdb/udump/testdb_ora_1851540.trc:

ORA-00600: internal error code, arguments: [16513], [1403], [4], [], [], [], [], []

Mon Dec  6 14:25:00 2010

Errors in file /cadrasu01/app/oracle/admin/testdb/udump/testdb_ora_1851540.trc:

ORA-00704: bootstrap process failure

ORA-00704: bootstrap process failure

ORA-00600: internal error code, arguments: [16513], [1403], [4], [], [], [], [], []

Mon Dec  6 14:25:00 2010

Error 704 happened during db open, shutting down database

USER: terminating instance due to error 704

Instance terminated by USER, pid = 1851540

ORA-1092 signalled during: ALTER DATABASE OPEN...

 

当我启用_minimum_giga_scn后，上述库可以open：

[P550_04_LA:oracle@:/cadrasu01/app/oracle/admin/testdb/bdump]#cat /dras20/testdb/inittestdb.ora

*.aq_tm_processes=1

CKPT started with pid=5

......省略显示部分内容

SMON: enabling cache recovery

event = "10513 trace name context forever, level 2:10231 trace name context forever, level 10"

*._minimum_giga_scn=1

 

SQL_testdb>startup pfile=/dras20/testdb/inittestdb.ora

ORACLE instance started.

 

Total System Global Area  504858456 bytes

Fixed Size                   743256 bytes

Variable Size             285212672 bytes

Database Buffers          218103808 bytes

Redo Buffers                 798720 bytes

Database mounted.

Database opened.

 

现在我们再去看一下datafile header里记录的checkpoint SCN：

BBED> set file 1 block 1

        FILE#           1

        BLOCK#          1

 

BBED> p kcvfhckp.kcvcpscn

struct kcvcpscn, 8 bytes                    @140    

   ub4 kscnbas                              @140      0x40000088

   ub2 kscnwrp                              @144      0x0000

 

我们可以清晰的看到现在datafile header里记录的checkpoint SCN确实向前推进到了大于1个G：

SQL>  select to_number('40000088','XXXXXXXXXXXX')/(1024*1024*1024) from dual;

 

TO_NUMBER('40000088','XXXXXXXX

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

              1.00000012665987

 

现在我们保留上述隐含参数后再次启库：

SQL_testdb>startup pfile=/dras20/testdb/inittestdb.ora

ORACLE instance started.

 

Total System Global Area  504858456 bytes

Fixed Size                   743256 bytes

Variable Size             285212672 bytes

Database Buffers          218103808 bytes

Redo Buffers                 798720 bytes

Database mounted.

Database opened.

 

BBED> set file 1 block 1

        FILE#           1

        BLOCK#          1

 

BBED> p kcvfhckp.kcvcpscn

struct kcvcpscn, 8 bytes                    @140    

   ub4 kscnbas                              @140      0x400000fc

   ub2 kscnwrp                              @144      0x0000

很明显，因为现在的checkpoint SCN已经大于1G了，所以oracle置上述隐含参数于不顾。

 

这里我把_minimum_giga_scn改为2后再启库：

[P550_04_LA:oracle@:/cadrasu01/app/oracle/product/9.2.0/bin]#cat /dras20/testdb/inittestdb.ora

*.aq_tm_processes=1

......省略显示部分内容

*.log_archive_start=TRUE

*._minimum_giga_scn=2

 

SQL_testdb>startup pfile=/dras20/testdb/inittestdb.ora

ORACLE instance started.

 

Total System Global Area  504858456 bytes

Fixed Size                   743256 bytes

Variable Size             285212672 bytes

Database Buffers          218103808 bytes

Redo Buffers                 798720 bytes

Database mounted.

Database opened.

 

SQL_testdb>show parameter scn;

 

NAME                                 TYPE        VALUE

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

_minimum_giga_scn                    integer     2

 

BBED> set file 1 block 1

        FILE#           1

        BLOCK#          1

 

BBED> p kcvfhckp.kcvcpscn

struct kcvcpscn, 8 bytes                    @140    

   ub4 kscnbas                              @140      0x8000008d

   ub2 kscnwrp                              @144      0x0000

 

SQL> select to_number('8000008d','XXXXXXXXXXXX')/(1024*1024*1024) from dual;

 

TO_NUMBER('8000008D','XXXXXXXX

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

              2.00000013131648