客户操作系统版本为AIX 6105，Oracle crs版本为10.2.0.5,rdbms和asm版本为10.2.0.1。     当用dbca工具建立磁盘组时，检索不出备选磁盘。在确定rdisk权限等没问题之后，在启动ASM实例之后，尝试在1号节点手动建立磁盘组，不料建磁盘组时出现以下错误。

引用

$ sqlplus "/as sysdba"

    SQL*Plus: Release 10.2.0.1.0 – Production on Thu Mar 31 13:44:37 2011

    Copyright (c) 1982, 2005, Oracle.  All rights reserved.

    Connected to:     Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 – 64bit Production     With the Partitioning, Real Application Clusters, OLAP and Data Mining options

    SQL> create diskgroup data1 external redundancy disk '/dev/rhdisk7' size 2047G;     create diskgroup data1 external redundancy disk '/dev/rhdisk7' size 2047G     *     ERROR at line 1:     ORA-15018: diskgroup cannot be created     IBM AIX RISC System/6000 Error: 6:  No such device or address     Additional information: -1     Additional information: 655360     IBM AIX RISC System/6000 Error: 6: No such device or address     Additional information: -1     Additional information: 655360     IBM AIX RISC System/6000 Error: 6: No such device or address     Additional information: -1     Additional information: 655360     IBM AIX RISC System/6000 Error: 6: No such device or address     Additional information: -1     Additional information: 655360     IBM AIX RISC System/6000 Error: 6: No such device or address     Additional information: -1     Additional information: 655360     IBM AIX RISC System/6000 Error: 6: No such device or address     Additional information: -1     Additional information: 655360     IBM AIX RISC System/6000 Error: 6: No such device or address     Additional information: -1     Additional information: 655360     IBM AIX RISC System/6000 Error: 6: No such device or address     Additional information: -1     Additional information: 655360     IBM AIX RISC System/6000 Error:

    红色标注可以看到，Oracle识别不了磁盘rhdisk7，但实际上rhdisk7确实存在。由于时间紧张，上google搜索，在一份otn的帖子中，一老外同行也碰到过类似问题，后来在解决问题描述时指出：

引用

Hi All,     I have resolved the issue now.     Since we are using SAN disks with raid 5. The disks size visible to ASm was not correct one.     So what i did is that I added SIZE clause during asm diskgroup creation.     create diskgroup oradata external redundany disk '/dev/rhdisk4' size 1000G;     by issuing this statement now I can create diskgroups.     Remember max size of a SAN disk for ASM is 2TB. We can't go beyond 2TB

    询问客户之后，发现本案例中也是raid5 而且rdisk也超过了2T，于是按照网友指出的方法进行创磁盘组尝试。幸好创建成功！

引用

SQL> create diskgroup data external redundancy disk '/dev/rhdisk7' size 1990G;

    Diskgroup created.

    创建成功之后，碰到了ASM重启之后，识别不了磁盘组错误：

引用

SQL> startup     ASM instance started

    Total System Global Area  130023424 bytes     Fixed Size                  2019136 bytes     Variable Size             102838464 bytes     ASM Cache                  25165824 bytes     ORA-15110: no diskgroups mounted

    或者

引用

SQL> startup     ASM instance started

    Total System Global Area  130023424 bytes     Fixed Size                  2019136 bytes     Variable Size             102838464 bytes     ASM Cache                  25165824 bytes     ORA-15032: not all alterations performed     ORA-15063: ASM discovered an insufficient number of disks for diskgroup     "DATA,DATA1"

    解决办法为：     在asm参数文件中显示指明该asm需要mount的磁盘组

引用

asm_diskgroups='DATA','ARCH'     ASM_DISKSTRING ='/dev/rhdisk7','/dev/rhdisk8'

    为什么ASM不能创建超过2T的硬盘呢？上metalink一查，Oracle针对这一现象，解释为bug：Bug 6453944 – ORA-15196 with ASM disks larger than 2TB [ID 6453944.8]

引用

ORA-15196 (ASM block corruption) can occur, if LUNs larger than 2TB are     presented to an ASM diskgroup.

    As a result of the fix, ORA-15099 will be raised if larger than 2TB size     disk is specified. This is irrespective of the presence of asmlib.

    In the future, this limitation will be removed.

    Workaround:       Do not add more than 2 TB size disk to a diskgroup.

    此外metalink 736891.1对此bug引起的后果有较为详细的描述，其中最直接的后果就是损坏ASM元数据（metadata）：

引用

There are no clear indications of what can go wrong with ASM LUNs larger than 2TB – but one example is wrapping: when Oracle gets to the end of 2TB, it may wrap back to the start of the LUN and  overwrite the ASM metadata

，并导致如下错误发生：

引用

ORA-15196 : invalid ASM block header [kfc.c:7997] [hard_kfbh] [4] [4] [34 != 130]

    当然这一bug已经在以下版本修复：

引用

    11.2         10.2.0.4 (Server Patch Set)         11.1.0.7 (Server Patch Set)

    那么除此之外，ASM还有什么限制呢?(metalink:ID 370921.1).我想除了2TB限制，其他限制达到的可能性还是比较低的。     

引用

63 disk groups in a storage system     10,000 ASM disks in a storage system      2 terabyte maximum storage for each ASM disk (the Bug 6453944 allowed larger sizes, but that led to problems, see Note 736891.1 "ORA-15196 WITH ASM DISKS LARGER THAN 2TB")     40 exabyte maximum storage for each storage system     1 million files for each disk group     2.4 terabyte maximum storage for each file