Diskgroup resource shows status offline after restart even it is mounted in ASM instance

　　今天在自己的测试库上，添加ASM磁盘组后，重启CRS服务后，新添加的ASM磁盘组无法自动ONLINE，环境如下

　　　　oracle 11.2.0.1.0 + ASM +REDHAT     RAC

　　　　OS VERSION：redhat 5.8 64-bit

　　　　DB VERSION：oracle 11.2.0.1.0 64-bit  

　　手动online该磁盘组可以成功，后来在metalink上查了下，原来这是一个bug。文档id 1065158.1

 

 Diskgroup resource shows status offline after restart even it is mounted in ASM instance (文档 ID 1065158.1)     
________________________________________
修改时间:2011-12-28 类型:PROBLEM 
 
In this Document
  Symptoms
  Changes
  Cause
  Solution
  References
________________________________________
APPLIES TO: 
　　Oracle Server - Enterprise Edition - Version: 11.2.0.1.0 and later   [Release: 11.2 and later ]
　　Information in this document applies to any platform.
SYMPTOMS
　　After restart the 11.2 HAS stack, the diskgroup (.dg) resource is showing "OFFLINE" status, but diskgroup is mounted in the ASM instance.

　　For example:
　　　　$ crsctl stat res ora.DATA.dg

　　　　NAME=ora.DATA.dg
　　　　TYPE=ora.diskgroup.type
　　　　TARGET=OFFLINE, ONLINE
　　　　STATE=OFFLINE, ONLINE on host1

　　From ASM when you query , the diskgroup shows as mounted.
　　　　SQL> conn / as sysdba
　　　　Connected.
　　　　SQL> select name,state from v$asm_diskgroup;

　　　　NAME                            STATE
　　　　------------------------------ -----------
　　　　DATA                             MOUNTED

CHANGES
　　From the crsd.log:
　2010-02-05 16:05:09.988: [    AGFW][56] ora.DATA.dg host1 1 received state from probe request. Old state = UNKNOWN, New state = OFFLINE

　　From oraagent_oracle.log:
　2010-02-05 16:05:09.799: [    AGFW][9] ora.DATA.dg host1 1 state changed from: UNKNOWN to: OFFLINE

CAUSE
　　This is caused by an unpublished bug 8663101 11.2 DG RES OFFLINE, BUT DISKGROUP MOUNTED AFTER RESTART HAS STACK 

　　The diskgroup is mounted in asm before crsd is up so crsd is never notified of the diskgroup resource start. 
SOLUTION
　　The bug has been fixed in 11.2.0.2 patchset.

Two workarounds available:
1. Manually start the diskgroup:
    srvctl start diskgroup -g <dg_name> [-n "<node_list>"]

or
2. Modify the attribute AUTO_START to "always"
     crsctl modify resource ora.<diskgroup>.dg -attr AUTO_START=always

REFERENCES
NOTE:1060920.1 - ALERT.LOG for ASM Shows "WARNING: failed to online diskgroup resource ora.GI.dg (unable to communicate with CRSD/OHASD)"