1.描述
生产环境dg 10.2.0.4,是由RMAN异机恢复，主库存在的redo路径在备库并未提取创建，也并未rename 日志文件，因此recover databaes完毕之后，其余MRP进程，同步应用日志，发现redo备库不存在，如下是操作日志
2.操作
尝试取消mrp
alter database recover managed standby database cancel;
确认dg角色

SQL> select dbid,name, database_role,open_mode,SWITCHOVER_STATUS,FLASHBACK_ON,PROTECTION_MODE,FORCE_LOGGING  from v$database;

       DBID NAME       DATABASE_ROLE

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

       3xx9 WxxxB      PHYSICAL STANDBY

 参数调整

SQL> alter system set standby_file_management=manual;

 查询日志组状态

SQL> select group# ,thread#,bytes//,status from v$log;

    GROUP#    THREAD# BYTES// STATUS

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

                                CLEARING

                                CLEARING

                                CLEARING_CURRENT

                                CLEARING

                                CLEARING

                                CLEARING

                                CLEARING

                                CLEARING

 rows selected.

手工对日志组进行clear后续进行drop 发现报错！

SQL> ALTER DATABASE CLEAR LOGFILE GROUP ;

alter database drop logfile group ;

ORA-01624: log 6 needed for crash recovery of instance webdb (thread 1)
ORA-00312: online log 6 thread 1: '/oracle/oradata/redolog/redo6.log'

根据MOS文档，Drop Redo Logs on Standby操作失败！！！
Sync Redo Logs on Primary and Standby 这部分区域存在一句话描述

7. Start Redo Apply:

When using Oracle Database >=10.2.0.1 and LOG_FILE_NAME_CONVERT is configured the online redo logs will

be automatically cleared and physically created when starting recovery.

因此采用上述建议
恢复参数


SQL> alter system set standby_file_management=auto;

设置参数，与主库相同，本次并不做转换

alter system set LOG_FILE_NAME_CONVERT ='/oracle/oradata/redolog/','/oracle/oradata/redolog/' scope=spfile;

SQL> startup force mount;

重启dg生效参数
开启mrp进程

SQL>recover managed standby database disconnect from session;

观察db_alert

Managed Standby Recovery not using Real Time Apply

 parallel recovery started with  processes

Tue Nov  ::

Waiting for all non-current ORLs to be archived...

Tue Nov  ::

Errors in file /oracle/admin/webdb/bdump/webdb_mrp0_13806.trc:

ORA-: open failed for members of log group  of thread

ORA-: online log  thread : '/oracle/oradata/redolog/redo1.log'

ORA-: unable to obtain file status

Linux-x86_64 Error: : No such file or directory

Additional information:

Tue Nov  :: 

操作系统验证-日志存在

[oracle@ytwebdb redolog]$ cd /oracle/oradata/redolog/

[oracle@ytwebdb redolog]$ ls -lrt

total

-rw-r-----  oracle dba  Nov  : redo1.log

-rw-r-----  oracle dba  Nov  : redo2.log

-rw-r-----  oracle dba  Nov  : redo3.log

-rw-r-----  oracle dba  Nov  : redo4.log