【備份恢復】之scn續

接着上篇繼續說scn...//

 

數據庫非正常關閉

SQL> startup mount
ORACLE instance started.

Total System Global Area    238530560 bytes
Fixed Size                                    1335724 bytes
Variable Size                         159387220 bytes
Database Buffers                     75497472 bytes
Redo Buffers                                2310144 bytes
Database mounted.
SQL> select checkpoint_change# from v$database;

CHECKPOINT_CHANGE#
------------------
                        880846

SQL> select checkpoint_change# from v$datafile;

CHECKPOINT_CHANGE#
------------------
                        880846
                        880846
                        880846
                        880846

SQL> select last_change# from v$datafile;

LAST_CHANGE#
------------                                        ------終止scn爲null

SQL> select checkpoint_change# from v$datafile_header;

CHECKPOINT_CHANGE#
------------------
                        880846
                        880846
                        880846
                        880846

數據庫開啓前比較v$datafile_header.checkpoint_change#v$database.checkpoint_change#
v$datafile_header.checkpoint_change#和v$datafile.last_change#此時啓動scn與終止scn不同，那麼數據庫需要實例恢復。

 

SQL> alter database open;

Database altered.

SQL> select checkpoint_change# from v$database;

CHECKPOINT_CHANGE#
------------------
                        905596

SQL> select checkpoint_change# from v$datafile;

CHECKPOINT_CHANGE#
------------------
                        905596
                        905596
                        905596
                        905596

SQL> select last_change# from v$datafile;

LAST_CHANGE#
------------

SQL> select checkpoint_change# from v$datafile_header;

CHECKPOINT_CHANGE#
------------------
                        905596
                        905596
                        905596
                        905596

此時數據庫smon進程做了實例恢復，所謂實例恢復就是指內存和進程，數據庫把redo中需要恢復的數據讀入內存.其過程分爲前滾、數據庫開啓、回滾三個階段。