RMAN> run{allocate channel c1 type disk; backup full database format '/opt/bak0_%s_%p_%t'; release channel c1;}

allocated channel: c1channel c1: sid=143 devtype=DISK

Starting backup at 24-APR-12using channel ORA_DISK_1channel ORA_DISK_1: starting full datafile backupsetchannel ORA_DISK_1: specifying datafile(s) in backupsetinput datafile fno=00006 name=/tmp/EAS_D_QM01_STANDARD.orainput datafile fno=00007 name=/opt/EAS_D_MEN1_STANDARD.orainput datafile fno=00008 name=/opt/data1/EAS60GOLD_DATA.dbfinput datafile fno=00009 name=/opt/data1/EAS_D_STANDARD.dbfinput datafile fno=00002 name=/opt/oracle/oradata/quanmai/undotbs01.dbfinput datafile fno=00001 name=/opt/oracle/oradata/quanmai/system01.dbfinput datafile fno=00003 name=/opt/oracle/oradata/quanmai/sysaux01.dbfinput datafile fno=00005 name=/opt/oracle/oradata/quanmai/example01.dbfinput datafile fno=00004 name=/opt/oracle/oradata/quanmai/users01.dbfchannel ORA_DISK_1: starting piece 1 at 24-APR-12channel ORA_DISK_1: finished piece 1 at 24-APR-12piece handle=/opt/bak0_3_1_781462215 tag=TAG20120424T165015 comment=NONEchannel ORA_DISK_1: backup set complete, elapsed time: 00:05:25channel ORA_DISK_1: starting full datafile backupsetchannel ORA_DISK_1: specifying datafile(s) in backupsetincluding current control file in backupsetincluding current SPFILE in backupsetchannel ORA_DISK_1: starting piece 1 at 24-APR-12channel ORA_DISK_1: finished piece 1 at 24-APR-12piece handle=/opt/bak0_4_1_781462540 tag=TAG20120424T165015 comment=NONEchannel ORA_DISK_1: backup set complete, elapsed time: 00:00:02Finished backup at 24-APR-12

RMAN-00571: ===========================================================RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============RMAN-00571: ===========================================================RMAN-03002: failure of release command at 04/24/2012 16:55:44RMAN-06012: channel: c1 not allocated

解决方式：Select Name,STATUS From  v$datafile;查看发现数据文件的状态在offline和recover状态修复后正常

如果无法创建如下（来自http://www.ochef.net/tag/rmanora-19504ora-27038）：RMAN-03009: failure of backup command on c1 channel at 08/04/2009 08:34:43

ORA-19504: failed to create file “/home/oracle/backup/full2009_08_A”

ORA-27038: created file already exists

错误分析几个步骤：错误中指定要创建的文件已经，而实际上是并不存在，也没有用先前的备份的数据恢复过数据库。考虑是不是在OS层面删除过备份文件，而没有在RMAN中更新，试着更新RMAN：

RMAN>crosscheck archivelog all;RMAN>delete expired archivelog all;RMAN>crosscheck backup;RMAN>delete expired backup;做了更新之后，问题依旧。

2.会不会是目录（/home/oracle/backup）权限的问题，修改目录权限：#chmod –R 775 /home/oracle/backup再执行备份，问题仍旧。3.试着让oracle自动分配通道，但配置RMAN的并行度为2RMAN>configure device type disk parallelism2；

尝试着再次备份，问题仍然是channel c2没有分配通道。

解决方式：增加%U参数后再次尝试备份：RMAN> run{allocate channel c1 type disk; backup full database format '/opt/bak0%Y_%M_%U_A'; release channel c1;}解决文章参考原文如下：

Solution Description:

=====================

Insert a %U into the format portion of the backup script in order to ensure a unique backup file name.

Problem Explanation:

====================

These errors occur because a backup file name already exists by the name specified in you backup script. For instance, if you are using the line “allocate channel c1 type disk format

‘/oracle/database/rman/backup/df_%d_%p_%c’;”, df_%d_%p_%c formats the backupstring like so;

df_ is simply a name. This could be any set of characters. In this case it means

database full.

%d_ is the database sid.

%p_is the backup piece number within the backup set.

%c_ specifies the copy number of the backup piece within a set of duplexed backup pieces.

There needs to be a %U added to the format string.

%U_ specifies a convenient shorthand that guarantees uniqueness in generated backup filenames. So, if the string were “db_%d_%U_%p_%c a unique name would be generated and it would not be necessary to either rename or move the backup file name prior to the next backup.

If the format is changed to include the %U, for instance;

allocate channel c1 type disk format ‘/oracle/database/rman/backup/df_%d_%U_%p_%c’;

The backup file will automatically have a unique name generated like;

df_JTG_0eblfm65_1_1_1_1

The next one would look like;

df_JTG_0fblfm76_1_1_1_1

翻译：解决方案描述：=====================插入一%到格式部分的备份脚本以保证一个独特的备份文件名。问题解答：====================这些错误的发生是因为一个备份文件名已存在于指定的名称在您的备份脚本。例如，如果您使用的是线”分配渠道Cl型磁盘格式“/数据/数据库/中/备份/ df_ % d_ % p_ %”；“，df_ % d_ % p_ %的backupstring格式一样；df_只是一个名字。这可以是任何字符集。在这种情况下它数据库已满。% d_是数据库。% p_is备份数的备份集。% c_指定的拷贝数的备份片内的一套全双工备份片。需要有一%个你添加到格式字符串。% u_指定一个方便的速记保证唯一性在生成的备份文件名。所以，如果字符串“db_ % % % % d_ u_ p_独特的名字会产生，就没有必要重命名或移动备份文件的名称之前，下一个备份。如果格式化包括%，例如；分配渠道Cl型磁盘格式/开发/数据库/中/备份/ df_ % % % % d_ u_ p_ '；备份文件将自动有一个唯一的名称产生像；df_jtg_0eblfm65_1_1_1_1下一个看起来像；df_jtg_0fblfm76_1_1_1_1