FYI：

http://docs.oracle.com/cd/E11882_01/backup.112/e10643/rcmsynta020.htm#RCMRF126

rman duplicate database 必须是同样的平台，可是支持32bit\64bit 间复制，在复制完毕后须要执行￥ORACLE_HOME/rdbms/admin/utlirp.sql
脚本来转换pl/sql 代码


The DUPLICATE command requires one or more auxiliary channels. These channels perform the work of the duplication on the auxiliary database instance. 
rman duplicate 命令须要在备库上至少要分配一个以上辅助通道（auxiliary channels）


If the source database files are in the Oracle Managed Files (OMF) format, then you cannot use the DB_FILE_NAME_CONVERT and 
LOG_FILE_NAME_CONVERT initialization parameters or the fileNameConversionSpec clause to generate new OMF file names for the 
duplicate database. If you do not follow this rule, the new OMF files generated from these three methods can cause problems
  使用duplicate 时必需要注意，假设源库使用OMF，那么我们就不能使用 DB_FILE_NAME_CONVERT 和 LOG_FILE_NAME_CONVERT 和其它的文件转
换方式。否者就会出问题；比如。备库的控制文件中的数据文件名称是依据DB_FILE_NAME_CONVERT 和 LOG_FILE_NAME_CONVERT 转换而来，可是实际

数据文件求不在相应的路径下，实际文件会採用OMF来管理；这时须要使用：

1》RMAN catalog 和 switch database to copy

2》alter database rename file to 和 

3》switch datafile file# to copy 来修正控制文件的信息；

                                      


当然也有例外，假设仅仅是换个asm 磁盘组名，其它路径名不变，能够使用DB_FILE_NAME_CONVERT
The only exception to this rule is when changing only an ASM disk group name. Assume that source data files and online redo 
log files are stored in ASM disk group +SOURCEDSK. You want to store the duplicate database files in ASM disk group +DUPDSK. 
In this case, you can set the initialization parameters as follows:
DB_FILE_NAME_CONVERT = ("+SOURCEDSK","+DUPDSK")
LOG_FILE_NAME_CONVERT = ("+SOURCEDSK","+DUPDSK") 


RMAN uses DB_FILE_NAME_CONVERT or LOG_FILE_NAME_CONVERT to convert the disk group name, and then generates a new, valid file 
name based on the converted disk group name.


-假设源库使用OMF，建议使用下面參数配置：
You have the following other supported options for naming data files when the source files are in the Oracle Managed Files format:

1.Use SET NEWNAME to specify names for individual data files. 
2.Set DB_FILE_CREATE_DEST to make all data files of the new database Oracle-managed files, except the files for which SET 


NEWNAME is used. You should not set DB_FILE_NAME_CONVERT if you set DB_FILE_CREATE_DEST.
DB_CREATE_FILE_DEST 
DB_RECOVERY_FILE_DEST 

对于redolog 假设没有设置DB_RECOVERY_FILE_DEST, or DB_CREATE_ONLINE_LOG_DEST_n. 则会使用DB_CREATE_FILE_DEST 參数指定的文件夹来存放

SQL> show parameter DB_CREATE_ONLINE_LOG_DEST
NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
db_create_online_log_dest_1          string
db_create_online_log_dest_2          string
db_create_online_log_dest_3          string
db_create_online_log_dest_4          string
db_create_online_log_dest_5          string

若单独指定 db_create_online_log_dest_* 參数。则redolog member 会冗余放到这些单独指定的文件夹下

Supported options for naming online redo logs duplicated from Oracle-managed files are DB_CREATE_FILE_DEST, 

DB_RECOVERY_FILE_DEST, or DB_CREATE_ONLINE_LOG_DEST_n.

----DB_CREATE_ONLINE_LOG_DEST_n 參数解释，为redolog，control file指定存放文件夹，指定多个则多路复用
DB_CREATE_ONLINE_LOG_DEST_n (where n = 1, 2, 3, ... 5) specifies the default location for Oracle-managed control files and online redo logs. If more than one DB_CREATE_ONLINE_LOG_DEST_n parameter is specified, then the control file or online redo log is multiplexed across the locations of the other DB_CREATE_ONLINE_LOG_DEST_n parameters. One member of each online redo log is created in each location, and one control file is created in each location.


Specifying at least two parameters provides greater fault tolerance for the control files and online redo logs if one of the locations should fail.

If a file system directory is specified as the default location, then the directory must already exist; Oracle does not create it. The directory must have appropriate permissions that allow Oracle to create files in it. Oracle generates unique names for the files, and a file thus created is an Oracle Managed File.

对于文件系统而言，指定的文件夹必须权限准确，且已经存在；假设没有，oracle 不会自己主动创建