诗檀软件专业数据库修复团队
适用于：
Oracle Database - Enterprise Edition – 版本 11.2.0.1及以上
本文信息适用于任何平台。
*** 12-Mar-2015检查相关性***
症状
ASM DISK WITH MOUNT_STATUS=CLOSED HEADER_STATUS=MEMBER GROUP_NUMBER=0
GROUP_NUMBER DISK_NUMBER MOUNT_S HEADER_STATU MODE_ST STATE OS_MB TOTAL_MB FREE_MB NAME FAILGROUP LABEL PATH
0 0 CLOSED MEMBER ONLINE NORMAL 69711 0 0 EMCORA43 ORCL:EMCORA43
0 1 CLOSED MEMBER ONLINE NORMAL 69711 0 0 EMCORA47 ORCL:EMCORA47
KFED IS SHOWING THESE DISKS ARE MEMBER OF ALREADY MOUNTED DISKGROUP (PRD_GOLD_FRA1)
GROUP_NUMBER NAME STATE
4 PRD_GOLD_FRA1 MOUNTED
/dev/oracleasm/disks/EMCORA47
disk /dev/oracleasm/disks/EMCORA47 is disk_number 12
kfdhdb.dsknum: 12 ; 0x024: 0x000c <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
kfdhdb.grptyp: 1 ; 0x026: KFDGTP_EXTERNAL
kfdhdb.hdrsts: 3 ; 0x027: KFDHDR_MEMBER
kfdhdb.dskname: EMCORA68 ; 0x028: length=8
kfdhdb.grpname: PRD_GOLD_FRA1 ; 0x048: length=13
kfdhdb.fgname: EMCORA68 ; 0x068: length=8
/dev/oracleasm/disks/EMCORA43
disk /dev/oracleasm/disks/EMCORA43 is disk_number 11
kfdhdb.dsknum: 11 ; 0x024: 0x000b <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
kfdhdb.grptyp: 1 ; 0x026: KFDGTP_EXTERNAL
kfdhdb.hdrsts: 3 ; 0x027: KFDHDR_MEMBER
kfdhdb.dskname: EMCORA66 ; 0x028: length=8
kfdhdb.grpname: PRD_GOLD_FRA1 ; 0x048: length=13
kfdhdb.fgname: EMCORA66 ; 0x068: length=8
原因
虽然 PRD_GOLD_DATA1 dg 的mount显示它在取消分配De-assign磁盘EMCORA43 和EMCORA47 ，这表示它们不被正常包括在磁盘组内。
这可能是由于在添加这些磁盘时，未在节点间共享的设备：
alert_+ASM1.log.15
===================
Sun Mar 25 12:43:15 2012
.......................
NOTE: Instance updated compatible.asm to 10.1.0.0.0 for grp 3
SUCCESS: diskgroup PRD_GOLD_DATA1 was mounted
NOTE: De-assigning number (4,11) from disk (ORCL:EMCORA43)
NOTE: De-assigning number (4,12) from disk (ORCL:EMCORA47)
kfdp_query(PRD_GOLD_FRA1): 38
kfdp_queryBg(): 38
NOTE: Instance updated compatible.asm to 10.1.0.0.0 for grp 4
SUCCESS: diskgroup PRD_GOLD_FRA1 was mounted
+ASM1_ora_10972.trc
======================
Validated domain 11, flags = 0x0
ASM name of disk 0x70882a00 (4:11:EMCORA66:ORCL:EMCORA43) is being cleared
NOTE: De-assigning number (4,11) from disk (ORCL:EMCORA43)
ASM name of disk 0x708819a8 (4:12:EMCORA68:ORCL:EMCORA47) is being cleared
NOTE: De-assigning number (4,12) from disk (ORCL:EMCORA47)
或由于以下错误：
WARNING: ignoring disk o/192.168.54.52/RECO_CD_11_ldnpso0322admin in deep discovery
ERROR: ORA-15033 signalled during reconfiguration of diskgroup RECO
....
alert_+ASM8.log:
Tue Apr 20 07:46:05 2010
ORA-15032: not all alterations performed
ORA-15033: disk 'o/192.168.54.52/RECO_CD_00_ldnpso0322admin' belongs to diskgroup "RECO"
ERROR: alter diskgroup RECO add disk 'o/192.168.54.52/RECO_CD_00_ldnpso0322admin' rebalance power 10 wait
v$asm_disk 查询输出中不显示磁盘号11 和12 ：
SELECT GROUP_NUMBER,DISK_NUMBER,MOUNT_STATUS,HEADER_STATUS,MODE_STATUS,STATE,OS_MB,TOTAL_MB,FREE_MB,NAME,FAILGROUP,LABEL,PATH FROM V$ASM_DISK WHERE GROUP_NUMBER=4;
GROUP_NUMBER DISK_NUMBER MOUNT_S HEADER_STATU MODE_ST STATE OS_MB TOTAL_MB FREE_MB NAME FAILGROUP LABEL PATH
4 0 CACHED MEMBER ONLINE NORMAL 69711 69711 21499 EMCORA24 EMCORA24 EMCORA24 ORCL:EMCORA24
4 1 CACHED MEMBER ONLINE NORMAL 69711 69711 21486 EMCORA25 EMCORA25 EMCORA25 ORCL:EMCORA25
4 2 CACHED MEMBER ONLINE NORMAL 69711 69711 21647 EMCORA26 EMCORA26 EMCORA26 ORCL:EMCORA26
4 3 CACHED MEMBER ONLINE NORMAL 69711 69711 21625 EMCORA27 EMCORA27 EMCORA27 ORCL:EMCORA27
4 4 CACHED MEMBER ONLINE NORMAL 69711 69711 21578 EMCORA30 EMCORA30 EMCORA30 ORCL:EMCORA30
4 5 CACHED MEMBER ONLINE NORMAL 69711 69711 21640 EMCORA31 EMCORA31 EMCORA31 ORCL:EMCORA31
4 6 CACHED MEMBER ONLINE NORMAL 69711 69711 21853 EMCDB004 EMCDB004 EMCDB004 ORCL:EMCDB004
4 7 CACHED MEMBER ONLINE NORMAL 69711 69711 21767 EMCORA34 EMCORA34 EMCORA34 ORCL:EMCORA34
4 8 CACHED MEMBER ONLINE NORMAL 69711 69711 21692 EMCORA35 EMCORA35 EMCORA35 ORCL:EMCORA35
4 9 CACHED MEMBER ONLINE NORMAL 69711 69711 21754 EMCORA36 EMCORA36 EMCORA36 ORCL:EMCORA36
4 10 CACHED MEMBER ONLINE NORMAL 69711 69711 22076 EMCORA65 EMCORA65 EMCORA65 ORCL:EMCORA65
4 13 CACHED MEMBER ONLINE NORMAL 69711 69711 22342 EMCORA64 EMCORA64 EMCORA64 ORCL:EMCORA64
解决方案
- 取磁盘的10 MB dd 备份：
dd if=/dev/oracleasm/disks/EMCORA43 of=/tmp/EMCORA43.txt bs=1M count=10
dd if=/dev/oracleasm/disks/EMCORA47 of=/tmp/EMCORA43.txt bs=1M count=10
- Change the permission of these 2 disks to 000
#chmod 000 /dev/oracleasm/disks/EMCORA47
#chmod 000 /dev/oracleasm/disks/EMCORA43
- 然后dismount 并mount 磁盘组PRD_GOLD_DATA1
- 如果磁盘组被成功mount，则我们能使用dd命令通过清除头删除这两个磁盘。
dd if=/dev/zero of=/dev/oracleasm/disks/EMCORA43 bs=1M count=1
dd if=/dev/zero of=/dev/oracleasm/disks/EMCORA47 bs=1M count=1
- 或你能将这两个磁盘从asm中删除，通过将标签label从asmlib oracleasm中删除。（你可能需要将磁盘的权限更改回660 ，通过以下操作）
/etc/init.d/oracleasm deletedisk EMCORA43
/etc/init.d/oracleasm deletedisk EMCORA47
- 或使用force选项以这些磁盘创建一个新的磁盘组，然后如果一切正常，则你能继续并drop新创建的磁盘组并将这些磁盘添加到需要的磁盘组。