11G新特性 -- flashback data archive(2)

创建Flashback Data Archive
用户需要授予dba或flashback archive administer系统特权。flashback archive administer系统特权包含：create flashback archive，alter flashback archive，drop flashback archive权利。

SQL> select * from dba_sys_privs where privilege like ‘%FLASH%‘;

GRANTEE                        PRIVILEGE                                ADM COM
------------------------------ ---------------------------------------- --- ---
SYS                            FLASHBACK ANY TABLE                      NO  NO
DBA                            FLASHBACK ANY TABLE                      NO  NO
SYS                            FLASHBACK ARCHIVE ADMINISTER             NO  NO
DBA                            FLASHBACK ARCHIVE ADMINISTER             NO  NO
MDSYS                          FLASHBACK ANY TABLE                      NO  NO

SQL> grant flashback archive administer to hr;
SQL> create tablespace flash_tbs datafile ‘/u01/app/oracle/oradata/yb/flash01.dbf‘ size 300m autoextend on next 30m maxsize 5g;
SQL> create flashback  archive flash1 tablespace flash_tbs retention 4 year;

创建Flashback Data Archive：

SQL> create tablespace tbs1 datafile ‘/u01/app/oracle/oradata/dbt/tbs101.dbf‘ size 2G;
SQL> create flashback archive default fla1 tablespace tbs1 quota 1g retention 1 month;

修改Flashback Data Archive：

#设置默认的flashback data archive
SQL> alter flashback archive fla1 set default;
#flashback data archive 添加表空间
SQL> alter flashback archive fla1 add tablespace tbs2 quota1g;
SQL> alter flashback archive fla1 add tablespace tbs3;
#修改配额
SQL> alter flashback archive fla1 modify tablespace tbs3 quota 20G;
#修改期限
SQL> alter flashback archive fla1 modify retention 2 year;
#移除一个表空间
SQL> alter flashback archive fla1 remove tablespace tbs2;
#purge历史数据
SQL> alter flashback archive fla1 purge all;
SQL> alter flashback archive fla1 purge before timestamp(systimestamp-inteval ‘1‘ day);
SQL> alter flashback archive fla1 purge before scn 123456;

删除Flashback Data Archive：

SQL> drop flashback archive fla1;

开启、取消flashback data archive：
缺省情况下没有对表开启Flashback Data Archive

SQL> create table test1( name varchar2(30),address varchar2(30)) flashback archive flash1;
SQL> select * from dba_flashback_archive_tables;

TABLE_NAME                OWNER_NAME                FLASHBACK_ARCHIVE_NAME         ARCHIVE_TABLE_NAME                                    STATUS
------------------------- ------------------------- ------------------------------ ----------------------------------------------------- -------------
TEST1                     HR                        FLASH1                         SYS_FBA_HIST_106929                                   ENABLED

SQL> create table test2( name varchar2(30),address varchar2(30)) ;
SQL> alter table test2 flashback archive;
SQL> select * from dba_flashback_archive_tables;

TABLE_NAME                OWNER_NAME                FLASHBACK_ARCHIVE_NAME         ARCHIVE_TABLE_NAME                                    STATUS
------------------------- ------------------------- ------------------------------ ----------------------------------------------------- -------------
TEST1                     HR                        FLASH1                         SYS_FBA_HIST_106929                                   ENABLED
TEST2                     HR                        FLASH1                         SYS_FBA_HIST_106930                                   ENABLED

SQL> alter table test2 no flashback archive;
SQL> drop flashback archive flash1;

Flashback Data Archive开启对DDL的支持：
只支持以下DDL语句：
-alter table
-truncate table
-rename table
如果使用了不受支持的DDL语句，可以先用dbms_flashback_archive.disassociate_fba过程使得表和flashback data archive分离，DDL操作完成后使用dbms_flashback_archive.reassociate_fba重新关联

示例：

SQL> create table test01(id int,name varchar2(20),job varchar2(20)) flashback archive fla1;

Table created.

SQL> insert into test01 values(1,‘hello‘,‘tec‘);

1 row created.

SQL> commit;

Commit complete.

SQL> select to_char(systimestamp, ‘YYYY-MM-DD HH24:MI:SS‘) FROM dual;

TO_CHAR(SYSTIMESTAM
-------------------
2015-11-03 10:03:04

SQL> update test01 set job=‘dev‘ where id=1;

1 row updated.

SQL> commit;

Commit complete.

SQL> select to_char(systimestamp, ‘YYYY-MM-DD HH24:MI:SS‘) FROM dual;

TO_CHAR(SYSTIMESTAM
-------------------
2015-11-03 10:06:04

SQL>  update test01 set job=‘dba‘ where id=1;

1 row updated.

SQL> commit;

Commit complete.

SQL> select to_char(systimestamp, ‘YYYY-MM-DD HH24:MI:SS‘) FROM dual;

TO_CHAR(SYSTIMESTAM
-------------------
2015-11-03 10:09:02

SQL> select * from test01;

        ID NAME                 JOB
---------- -------------------- --------------------
         1 hello                dba

SQL> SELECT * FROM test01 AS OF TIMESTAMP TO_TIMESTAMP (‘2015-11-03 10:06:04‘, ‘YYYY-MM-DD HH24:MI:SS‘);

        ID NAME                 JOB
---------- -------------------- --------------------
         1 hello                dev

SQL>