数据库管理-m6米乐安卓版下载

第一堂搞完了，感觉其实内容是熟悉而陌生，毕竟上一次摸ocm都是17年的时候，继续第二堂。
第二堂标题是“backup/recovery”备份与恢复。本堂操作均需要开启归档，如果前一堂没有开启归档，本堂每题操作前需要开启归档，同时注意冷备的归档状态，以免后续需要使用。

考点1：configure recovery catalog 配置恢复catalog

连接至emrep数据库作为catalog数据库，catalog用户为rc_admin/oracle，将prodcdb和prod4注册到catalog中：

-- 环境准备
sqlplus sys/oracle123@emrep as sysdba
	create tables catatbs datafile '/u01/app/oracle/oradata/emrep/catatbs01.dbf' size 10m autoextend on maxsize 2g;
	create user rc_admin identified by oracle account unlock;
	grant dba,recovery_catalog_owner to rc_admin,
	alter user rc_admin default tablespace catatbs quota unlimited on catatbs;
rman catalog rc_admin/oracle@emrep
	create catalog;
-- 操作
rman target sys/oracle@prodcdb catalog rc_admin/oracle@emrep
	upgrade catalog;
	upgrade catalog;
	register databaase;
rman target sys/oracle@prod4 catalog rc_admin/oracle@emrep
	upgrade catalog;
	upgrade catalog;
	register databaase;

考点2：perform backup and backup configuration 备份及备份配置

（这里把网页考点2,3，4的位置调整一下，即是先备份后恢复）

1. prod4数据库开启块追踪：

rman target sys/oracle@prod4
	configure backup optimization on;
	-- configure controlfile autobackup on;
	-- configure device type disk backup type to compressed backupset;
sqlplus sys/oracle@prod4 as sysdba
	alter database enable block change tracking using file '/u01/app/oracle/oradata/prod4/block.txt';

2. 为prodcdb创建一个包含所有pdb的备份：

rman target sys/oracle@prodcdb
	backup full database plus archivelog;

3. 备份pdbprod1中的sysaux表空间，并需要保留很长时间。同时需要解决串行备份的性能问题：

rman target sys/oracle@prodcdb catalog rc_admin/oracle@emrep
configure device type disk parallelism 4;
backup section size 100m tablespace pdbprod1:sysaux keep forever format '/u01/app/oracle/fast_recovery_area/prodcdb/%d_%t_%u.bak';

4. 对prod4执行增量/差异备份:

rman target sys/oracle@prod4
	backup incremental level 0 database;
	backup incremental level 1 database;
	backup incremental level 2 database;

5. 其他备份

rman target sys/oracle@prod4
	backup as backupset database tag "ocm01";
	backup datafile 5 section size 25m [tag "ocm02size25m"];
rman target sys/oracle@prod4 catalog rc_admin/oracle@emrep
	change backup tag "ocm02size25m" keep fover;

考点3：configuring flashback database 配置闪回数据库

1. prod4开启闪回，fra目录为/u01/app/oracle/flash，大小5gb：

sqlplus sys/oracle@prod4 as sysdba
	alter system set db_recovery_file_dest='/u01/app/oracle/flash';
	alter system set db_recovery_file_dest_size=5g;
	-- alter system set db_flashback_retention_target=2880 scope=spfile;
	shut immediate
	startup mount
	alter database flashback on;
	alter database open;

2. prod4创建不会过期的还原点rp1:

sqlplus sys/oracle@prod4 as sysdba
	create restore point rp1 guarantee flashback database;

考点4：perform recovery and duplication 恢复与复制

1. prod4执行脚本xxx.sql，最终发现脚本内容误操作，请恢复相关表数据：

sqlplus sys/oracle@prod4 as sysdba
	select to_char(sysdate,'yyyy-mm-dd hh24:mi:ss') from dual;	-- 获取当前时间戳
	-- 执行题目要求脚本
	shut immediate
	startup mount
	flashback database to timestamp to_timestamp('xxxx-xx-xx xx:xx:xx','yyyy-mm-dd hh24:mi:ss'); -- 闪回数据库到之前获取的时间
	alter database open read only; -- 因为仅是需要将异常数据恢复，所以不适用open resetlog方式开启数据库
	-- 找到变更数据exp出来
	-- exp \'/ as sysdba\' tables=username.tabname file=xxx.dmp
	shut immediate
	startup mount
	recover database;
	alter database open resetlog;
	-- 将对应表数据清除后，imp相关表数据，如遇主外键影响，删除外键导入数据后重建外键
-- 其实这里如果不涉及ddl操作，仅为dml操作，使用闪回查询也可以，需要根据题目具体情况选择
	select to_char(current_scn) from v$database; -- 获取当前scn
	select to_char(sysdate,'yyyy-mm-dd hh24:mi:ss') from dual;	-- 获取当前时间戳
	-- 执行题目要求脚本
	create table username.tabname_old as select * from username.tabname as of timestamp to_timestamp('xxxx-xx-xx xx:xx:xx','yyyy-mm-dd hh24:mi:ss'); -- 基于时间点的闪回查询
	create table username.tabname_old as select * from username.tabname as of scn xxxxxxxxxxxxxx; -- 基于scn的闪回查询
	-- 后续在替换数据

2. prod4中从回收站恢复表名为t1，列名包含name的表，恢复表名为t1_old：

sqlplus sys/oracle@prod4 as sysdba
	show recyclebin;/select * from dba_recyclebin;
	desc "bin$....";
	flashback table "bin$...." to before drop remane to t1_old;

总结

第二堂内容应该是基本覆盖完了。
老规矩，知道写了些啥。