12C容器数据库的备份恢复及闪回CDB与PDB。

2020-09-12 00:00:00 文件备份报错日志恢复

12C之五备份恢复、闪回CDB与PDB。

（1）备份整个CDB

rman target /

backup as compressed backupset database plus archivelog delete all input;

（2）备份pluggable库

backup as compressed backupset pluggable database orclpdb;

（3）恢复PDB的system表空间

sqlplus system/oracle@orclpdb

select file_name from dba_data_files;

! rm /u01/app/oracle/oradata/orcl/orclpdb/system01.dbf

rman target /

shutdown abort

start mount

restore tablespace orclpdb:system;

recover tablespace orclpdb:system;

alter database open;

alter pluggable database orclpdb open;

至此恢复完毕

也可以将整个PDB恢复到方式实现

rman target /

shutdown abort;

startup mount;

restore pluggable database orclpdb;

recover pluggable database orclpdb;

alter database open;

alter pluggable database orclpdb open;

（4）恢复PDB的非关键文件（高可用恢复）

创建测试表空间

create tablespace users datafile ‘/u01/app/oracle/oradata/orcl/orclpdb/users01.dbf’ size 10m;

create table t (id number)

insert into t values(100);

commit;

alter table t move tablespace users;

!rm /u01/app/oracle/oradata/orcl/orclpdb/users01.dbf

(这里清除Buffer cache验证表t无法查询数据）

sqlplus system/oracle@orclpdb

alter tablespace users offline immediate

(这里文件已经删除了，必须使用immediate,否则报错文件不存在）

rman target /

restore tablespace orclpdb:users;

recover tablespace orclpdb:users;

sqlplus system/oracle@orclpdb

alter tablespace users online;

（5）备份控制文件夹

alter database backup controlfile to trace;

cd /u01/app/oracle/diag/rdbms/orcl/orcl/trace/

ls -ltr | tail -10

配置RMAN自动备份

(6) 恢复控制文件

方法1：直接拷贝完好的副本

方法2：使用trace的脚本

方法3：使用RMAN自动备份恢复。

select name from v$controlfile;

!rm /u01/app/oracle/fast_recovery_area/orcl/control02.ctl

此时尝试关闭数据库，无法关库（关库需要写控制文件，完全检查点），报错

shutdown abort;

rman target /

startup nomount;

restore controlfile from autobackup;

alter database mount;

recover database;

alter database open resetlogs;

sqlplus / as sysdba

alter pluggable database orclpdb open;

后再做一次全备份

rman target /

backup as compressed backupset database plus archivelog delete all input;

（7）恢复重做日志文件

先向日志组添加成员

alter database add logfile member '/u01/app/oracle/oradata/orcl/redo01_2’ to group 1;

alter database add logfile member '/u01/app/oracle/oradata/orcl/redo02_2' to group 2;

alter database add logfile member '/u01/app/oracle/oradata/orcl/redo03_2’ to group 3;

切换日志，让所有新成员都有数据。

查看当前日志状态

SQL> select group#,sequence#,archived,status from v$log;

GROUP# SEQUENCE# ARC STATUS

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

1 10 NO CURRENT

2 8 YES INACTIVE

3 9 YES INACTIVE

我们删除当前Group1 的某个日志成员

!rm -f /u01/app/oracle/oradata/orcl/redo01_2

group1—/u01/app/oracle/oradata/orcl/redo01_2

group1—/u01/app/oracle/oradata/orcl/redo01

如何恢复呢，通过clear的方式尝试重新格式化重做日志

alter database clear logfile group 1;

但是该日志的当前在用的，系统报错

下面尝试切换日志

alter system switch logfile;

切换成功后，查询日志状态

SQL> select group#,sequence#,archived,status from v$log;

GROUP# SEQUENCE# ARC STATUS

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

1 10 YES ACTIVE

2 11 NO CURRENT

3 9 YES INACTIVE

再clear，可能依然报错，需要通过完全检查点推进，使得日志组1实例恢复不再需要。

alter database clear logfile group 1;

ERROR at line 1:

ORA-01624: log 1 needed for crash recovery of instance orcl (thread 1)

ORA-00312: online log 1 thread 1: '/u01/app/oracle/oradata/orcl/redo01.log'

ORA-00312: online log 1 thread 1: '/u01/app/oracle/oradata/orcl/redo01_2

alter system checkpoint;

继续clear

alter database clear logfile group 1;

此时成功！

验证：

!ls -l /u01/app/oracle/oradata/orcl/redo*

其他

如新增日志组

alter database add logfile group 1 '/u01/app/oracle/oradata/orcl/redo01.log' size 200m;

删除日志成员

alter database drop logfile member '/u01/app/oracle/oradata/orcl/redo01.log'

注意：操作系统文件依然存在，需要手工删除

相关文章