1、ASM 3种关闭启动方式
-----使用SQLPLUS 连接到SYSASM直接SHUTDOWN IMMEDIATE，启动直接STARTUP即可
-----使用SRVCTL 命令
[grid@tasm ~]srvctlstopdiskgroup−gARCH[grid@tasm ] srvctl stop diskgroup -g DATA
[grid@tasm ~]$ srvctl stop asm -o immediate
启动使用srvctl start asm -o open 即可
-----使用ASMCMD命令
ASMCMD> shutdown -h
Unknown option: h
usage: shutdown [ --normal |--abort|--immediate ]
help: help shutdown
ASMCMD> startup -h
Unknown option: h
usage: startup [--nomount] [--restrict] [ --pfile pfile ]
help: help startup
2、
create diskgroup data normal redundancy
failgroup T1 disk
'/dev/raw/raw1' name d1 --force 可以删除以前的数据
failgroup T2 disk
'/dev/raw/raw2' name d2;
CREATE DISKGROUP data2 External Redundancy DISK '/dev/raw/raw8'
ATTRIBUTE 'compatible.asm' = '11.2', 'compatible.rdbms' = '11.1'
High Redundancy
-----------------
drop diskgroup data including contents;
-----删除磁盘组，如果磁盘组包含文件必须使用INCLUDING CONTENTS。
SQL> alter diskgroup data add disk
2 '/dev/raw/raw8' name T3D ;
Diskgroup altered.
--------完成后进行平衡需要一定的时间
SQL> alter diskgroup data drop disk t3d;
Diskgroup altered.
--------完成后需要删除磁盘需要一定的时间把AU进行平衡，如果是普通冗余当你删除倒数第二个磁盘时会报错，因为最小2个失效组
SQL> select state,NAME,MODE_STATUS,FREE_MB,path,HEADER_STATUS from v$asm_disk;
STATE NAME MODE_STATUS FREE_MB PATH HEADER_STATUS
-------- ------------------------------ ----------- ---------- -------------------------------------------------------------------------------- -------------
NORMAL DATA_0000 ONLINE 816 /dev/raw/raw5 MEMBER
NORMAL DATA_0001 ONLINE 815 /dev/raw/raw6 MEMBER
NORMAL ARCH_0000 ONLINE 2642 /dev/raw/raw7 MEMBER
DROPPING T3D ONLINE 2759 /dev/raw/raw8 MEMBER
这里状态为DROPPING在DROPPING期间即可进行UNDROP，但是一个删除完成就不可以了
SQL> alter diskgroup data undrop disks;
Diskgroup altered.
--------在还没有删除完成前可以进行回退操作
SQL> select state,NAME,MODE_STATUS,FREE_MB,path,HEADER_STATUS from v$asm_disk;
STATE NAME MODE_STATUS FREE_MB PATH HEADER_STATUS
-------- ------------------------------ ----------- ---------- -------------------------------------------------------------------------------- -------------
NORMAL DATA_0000 ONLINE 761 /dev/raw/raw5 MEMBER
NORMAL DATA_0001 ONLINE 761 /dev/raw/raw6 MEMBER
NORMAL ARCH_0000 ONLINE 2642 /dev/raw/raw7 MEMBER
NORMAL ONLINE 0 /dev/raw/raw8 FORMER
删除完成后，状态变为NORMAL，HEADER状态变为FROMER，表示曾经使用过，
3、兼容性
compatible.asm 管理者版本，管理ASM diskgroup的实例必须大于等于这个设置
compatible.rdbms 使用者版本，使用ASM diskgroup 的实例必须大于等于这个设置
而compatible.asm必须大于等于compatible.rdbms ，否则报错如下：
SQL> CREATE DISKGROUP data2 DISK '/dev/raw/raw8'
2 ATTRIBUTE 'compatible.asm' = '11.1', 'compatible.rdbms' = '11.2';
CREATE DISKGROUP data2 DISK '/dev/raw/raw8'
*
ERROR at line 1:
ORA-15018: diskgroup cannot be created
ORA-15234: target RDBMS compatibility (11.2.0.0.0) exceeds ASM compatibility
(11.1.0.0.0)
（SQL> CREATE DISKGROUP data2 DISK '/dev/raw/raw8'
2 ATTRIBUTE 'compatible.asm' = '11.2', 'compatible.rdbms' = '11.1';
CREATE DISKGROUP data2 DISK '/dev/raw/raw8'
*
ERROR at line 1:
ORA-15018: diskgroup cannot be created
ORA-15072: command requires at least 2 regular failure groups, discovered only
1
）
SQL> CREATE DISKGROUP data2 External Redundancy DISK '/dev/raw/raw8'
2 ATTRIBUTE 'compatible.asm' = '11.2', 'compatible.rdbms' = '11.1';
Diskgroup created.
SQL> select * from vasmdiskgroup;GROUPNUMBERNAMESECTORSIZEBLOCKSIZEALLOCATIONUNITSIZESTATETYPETOTALMBFREEMBHOTUSEDMBCOLDUSEDMBREQUIREDMIRRORFREEMBUSABLEFILEMBOFFLINEDISKSCOMPATIBILITYDATABASECOMPATIBILITYVOTINGFILES−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−1ARCH51240961048576CONNECTEDEXTERN28702642022802642011.2.0.0.010.1.0.0.0N2DATA51240961048576CONNECTEDNORMAL57401502042380751011.2.0.0.011.2.0.0.0N3DATA251240961048576MOUNTEDEXTERN2870281805202818011.2.0.0.011.1.0.0.0N4、查看属性SQL>SELECTdg.nameASdiskgroup,t.name,t.stripe,t.redundancy,t.primaryregion,t.mirrorregion2FROMVASM_DISKGROUP dg, V$ASM_TEMPLATE t
3 WHERE dg.group_number = t.group_number and dg.name = 'DATA' ORDER BY t.name;
DISKGROUP NAME STRIPE REDUNDANCY PRIMARY_REGION MIRROR_REGION
------------------------------ ------------------------------ ------ ---------- -------------- -------------
DATA ARCHIVELOG COARSE MIRROR COLD COLD
DATA ASMPARAMETERFILE COARSE MIRROR COLD COLD
DATA AUTOBACKUP COARSE MIRROR COLD COLD
DATA BACKUPSET COARSE MIRROR COLD COLD
DATA CHANGETRACKING COARSE MIRROR COLD COLD
DATA CONTROLFILE FINE HIGH COLD COLD
DATA DATAFILE COARSE MIRROR COLD COLD
DATA DATAGUARDCONFIG COARSE MIRROR COLD COLD
DATA DUMPSET COARSE MIRROR COLD COLD
DATA FLASHBACK COARSE MIRROR COLD COLD
DATA FLASHFILE COARSE MIRROR COLD COLD
DATA OCRFILE COARSE MIRROR COLD COLD
DATA ONLINELOG COARSE MIRROR COLD COLD
DATA PARAMETERFILE COARSE MIRROR COLD COLD
DATA TEMPFILE COARSE MIRROR COLD COLD
DATA XTRANSPORT COARSE MIRROR COLD COLD
SQL> SELECT dg.name AS diskgroup, SUBSTR(a.name,1,18) AS name,
2 SUBSTR(a.value,1,24) AS value, read_only FROM VASMDISKGROUPdg,3VASM_ATTRIBUTE a WHERE dg.name = 'DATA'
4 AND dg.group_number = a.group_number and a.NAME NOT LIKE 'template%';
DISKGROUP NAME VALUE READ_ONLY
------------------------------ -------------------------------------------------------------------------------- -------------------------------------------------------------------------------- ---------
DATA disk_repair_time 3.6h N
DATA au_size 1048576 Y
DATA sector_size 512 Y
DATA compatible.asm 11.2.0.0.0 N
DATA compatible.rdbms 11.2.0.0.0 N
DATA cell.smart_scan_ca FALSE N
DATA access_control.ena FALSE N
DATA access_control.uma 066 N
alter diskgroup data offline disk DATA_0000 drop after 0.5m; 可以指定disk_repair_time的时间，本列30秒后删除DISK。
SQL> select * from v$asm_disk;
GROUP_NUMBER DISK_NUMBER COMPOUND_INDEX INCARNATION MOUNT_STATUS HEADER_STATUS MODE_STATUS STATE REDUNDANCY LIBRARY OS_MB TOTAL_MB FREE_MB HOT_USED_MB COLD_USED_MB NAME FAILGROUP LABEL PATH UDID PRODUCT CREATE_DATE MOUNT_DATE REPAIR_TIMER READS WRITES READ_ERRS WRITE_ERRS READ_TIME WRITE_TIME BYTES_READ BYTES_WRITTEN PREFERRED_READ HASH_VALUE HOT_READS HOT_WRITES HOT_BYTES_READ HOT_BYTES_WRITTEN COLD_READS COLD_WRITES COLD_BYTES_READ COLD_BYTES_WRITTEN VOTING_FILE SECTOR_SIZE FAILGROUP_TYPE
------------ ----------- -------------- ----------- ------------ ------------- ----------- -------- ---------- ---------------------------------------------------------------- ---------- ---------- ---------- ----------- ------------ ------------------------------ ------------------------------ ------------------------------- -------------------------------------------------------------------------------- ---------------------------------------------------------------- -------------------------------- ----------- ----------- ------------ ---------- ---------- ---------- ---------- ---------- ---------- ---------- ------------- -------------- ---------- ---------- ---------- -------------- ----------------- ---------- ----------- --------------- ------------------ ----------- ----------- --------------
2 1 33554433 3940853700 CACHED MEMBER ONLINE NORMAL UNKNOWN System 2870 2870 751 0 2119 DATA_0001 DATA_0001 /dev/raw/raw6 2013/8/13 2 2013/8/25 1 0 12399 15727 0 0 305.731401 384.269159 2402186240 2244893696 U 559735068 0 0 0 0 9428 5562 141222912 155028480 N 512 REGULAR
1 0 16777216 3940853698 CACHED MEMBER ONLINE NORMAL UNKNOWN System 2870 2870 2642 0 228 ARCH_0000 ARCH_0000 /dev/raw/raw7 2013/8/14 0 2013/8/25 1 0 8593 14440 0 0 37.040936 79.919776 136900608 163531776 797869928 0 0 0 0 8094 9978 132796416 145255424 N 512 REGULAR
2 0 33554432 3940853699 MISSING UNKNOWN OFFLINE FORCING UNKNOWN System 0 2870 751 0 2119 _DROPPED_0000_DATA DATA_0000 2013/8/13 2 2013/8/25 1 0 U 1316795825 N 0 REGULAR
删除后NAME更名为_DROPPED_0000_DATA，状态均有变化，这里实际上是如果出现磁盘读写错误，ASM会自动的设置DISK为OFFLINE
SQL> select state,NAME,MODE_STATUS,FREE_MB,path,HEADER_STATUS from v$asm_disk;
STATE NAME MODE_STATUS FREE_MB PATH HEADER_STATUS
-------- ------------------------------ ----------- ---------- -------------------------------------------------------------------------------- -------------
NORMAL DATA_0001 ONLINE 751 /dev/raw/raw6 MEMBER
NORMAL ARCH_0000 ONLINE 2642 /dev/raw/raw7 MEMBER
FORCING _DROPPED_0000_DATA OFFLINE 751 UNKNOWN
NORMAL ONLINE 0 /dev/raw/raw8 FORMER
NORMAL ONLINE 0 /dev/raw/raw5 MEMBER ---这里是原始的磁盘
这种状态下没有办法恢复了，只有重新键入磁盘
SQL> alter diskgroup data add disk '/dev/raw/raw5' force;
Diskgroup altered.
SQL> r
1* select state,NAME,MODE_STATUS,FREE_MB,path,HEADER_STATUS,GROUP_NUMBER from v$asm_disk
STATE NAME MODE_ST FREE_MB PATH HEADER_STATU GROUP_NUMBER
-------- ------------------------------ ------- ---------- ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ------------ ------------
NORMAL ONLINE 0 /dev/raw/raw8 FORMER 0
FORCING _DROPPED_0000_DATA OFFLINE 1108 UNKNOWN 2
NORMAL DATA_0001 ONLINE 751 /dev/raw/raw6 MEMBER 2
NORMAL DATA_0002 ONLINE 2511 /dev/raw/raw5 MEMBER 2
NORMAL ARCH_0000 ONLINE 2642 /dev/raw/raw7 MEMBER 1
DATA_0002 free_mb不断减少，_DROPPED_0000_DATA free_mb不断增多,实际上_DROPPED_0000_DATA并不是一块真实的盘而是一个占位而已表示曾经的DATA_0000磁盘出现了故障OFFLINE删除了，并且
如果需要新增盘的话，需要多少空间而已，当然这个特性只能在高冗余或者普通冗余有用，当你新加入盘后数据会从健康的盘同步到新的盘中，同步完成后会自动消失，因为磁盘组又恢复到正常状态，
及普通冗余有2个失效组，高冗余有3个失效组。
SQL> r
1* select state,NAME,MODE_STATUS,FREE_MB,path,HEADER_STATUS,GROUP_NUMBER from v$asm_disk
STATE NAME MODE_ST FREE_MB PATH HEADER_STATU GROUP_NUMBER
-------- ------------------------------ ------- ---------- ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ------------ ------------
NORMAL ONLINE 0 /dev/raw/raw8 FORMER 0
NORMAL DATA_0001 ONLINE 751 /dev/raw/raw6 MEMBER 2
NORMAL DATA_0002 ONLINE 751 /dev/raw/raw5 MEMBER 2
NORMAL ARCH_0000 ONLINE 2642 /dev/raw/raw7 MEMBER 1
最后恢复完成
