通过asm来存储数据库文件，在linux下能够通过asmlib的方式来管理块设备，也能够直接使用裸设备来建立asm磁盘。在asmlib方式下，磁盘设备启动顺序和名称的改变不会影响到asm的使用。但假设直接使用裸设备会怎么样那？我们知道asm会在磁盘中存储与asm有关的元数据，通过这些元数据asm能够了解磁盘的相关信息，因此理论上裸设备名称的改变不会影响asm的正常使用。以下，通过实验来验证一下。
首先看下面。裸设备的配置文件
node1
[root@node1 ~]# cat /etc/udev/rules.d/60-raw.rules
# Enter raw device bindings here.
#
# An example would be:
# ACTION=="add", KERNEL=="sda", RUN+="/bin/raw /dev/raw/raw1 %N"
# to bind /dev/raw/raw1 to /dev/sda, or
# ACTION=="add", ENV{MAJOR}=="8", ENV{MINOR}=="1", RUN+="/bin/raw /dev/raw/raw2 %M %m"
# to bind /dev/raw/raw2 to the device with major 8, minor 1.
ACTION=="add", KERNEL=="sdb1", RUN+="/bin/raw /dev/raw/raw1 %N"
ACTION=="add", KERNEL=="sdb2", RUN+="/bin/raw /dev/raw/raw2 %N"
ACTION=="add", KERNEL=="sdb5", RUN+="/bin/raw /dev/raw/raw3 %N"
ACTION=="add", KERNEL=="sdb6", RUN+="/bin/raw /dev/raw/raw4 %N"
ACTION=="add", KERNEL=="sdb7", RUN+="/bin/raw /dev/raw/raw5 %N"
ACTION=="add", KERNEL=="sdb8", RUN+="/bin/raw /dev/raw/raw6 %N"
ACTION=="add", KERNEL=="sdb9", RUN+="/bin/raw /dev/raw/raw7 %N"
node2:
[root@node2 rules.d]# cat 60-raw.rules
# Enter raw device bindings here.
#
# An example would be:
# ACTION=="add", KERNEL=="sda", RUN+="/bin/raw /dev/raw/raw1 %N"
# to bind /dev/raw/raw1 to /dev/sda, or
# ACTION=="add", ENV{MAJOR}=="8", ENV{MINOR}=="1", RUN+="/bin/raw /dev/raw/raw2 %M %m"
# to bind /dev/raw/raw2 to the device with major 8, minor 1.
ACTION=="add", KERNEL=="sdb1", RUN+="/bin/raw /dev/raw/raw1 %N"
ACTION=="add", KERNEL=="sdb2", RUN+="/bin/raw /dev/raw/raw2 %N"
ACTION=="add", KERNEL=="sdb5", RUN+="/bin/raw /dev/raw/raw3 %N"
ACTION=="add", KERNEL=="sdb6", RUN+="/bin/raw /dev/raw/raw4 %N"
ACTION=="add", KERNEL=="sdb7", RUN+="/bin/raw /dev/raw/raw5 %N"
ACTION=="add", KERNEL=="sdb8", RUN+="/bin/raw /dev/raw/raw6 %N"
ACTION=="add", KERNEL=="sdb9", RUN+="/bin/raw /dev/raw/raw7 %N"
asm下的磁盘信息例如以下：
[oracle@node1 ~]$ export ORACLE_SID=+ASM1 [oracle@node1 ~]$ sqlplus / as sysdba SQL*Plus: Release 10.2.0.5.0 - Production on Sun Aug 31 12:55:25 2014 Copyright (c) 1982, 2010, Oracle. All Rights Reserved. Connected to: Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production With the Partitioning, Real Application Clusters, OLAP, Data Mining and Real Application Testing options SQL> col name for a30 SQL> col label for a30 SQL> col path for a30 SQL> set linesize 200 SQL> / NAME LABEL PATH ------------------------------ ------------------------------ ------------------------------ DG1_0000 /dev/raw/raw3 DG2_0000 /dev/raw/raw4 DG3_0000 /dev/raw/raw5 DG3_0001 /dev/raw/raw6 DG4_0000 /dev/raw/raw7 /dev/raw/raw2 /dev/raw/raw1 7 rows selected. SQL> ho ssh node2 Last login: Sat Aug 30 17:56:54 2014 from node1 [oracle@node2 ~]$ export ORACLE_SID=+ASM2 [oracle@node2 ~]$ sqlplus / as sysdba SQL*Plus: Release 10.2.0.5.0 - Production on Sun Aug 31 12:56:38 2014 Copyright (c) 1982, 2010, Oracle. All Rights Reserved. Connected to: Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production With the Partitioning, Real Application Clusters, OLAP, Data Mining and Real Application Testing options SQL> col name for a30 SQL> col label for a30 SQL> col path for a30 SQL> set linesize 200 SQL> select name,label,path from v$asm_disk order by 1; NAME LABEL PATH ------------------------------ ------------------------------ ------------------------------ DG1_0000 /dev/raw/raw3 DG2_0000 /dev/raw/raw4 DG3_0000 /dev/raw/raw5 DG3_0001 /dev/raw/raw6 DG4_0000 /dev/raw/raw7 /dev/raw/raw2 /dev/raw/raw1 7 rows selected.
SQL> select file_name,tablespace_name from dba_data_files;
FILE_NAME TABLESPACE_NAME
-------------------------------------------- ------------------------------
+DG4/easy/datafile/system.272.856543875 SYSTEM
+DG4/easy/datafile/undotbs1.273.856543885 UNDOTBS1
+DG4/easy/datafile/sysaux.274.856543891 SYSAUX
+DG4/easy/datafile/undotbs2.276.856543901 UNDOTBS2
+DG4/easy/datafile/users.277.856543905 USERS
SQL> create table t1 (id number,name varchar2(20)) tablespace users;
Table created.
SQL> insert into t1 values(1,111);
1 row created.
SQL> insert into t1 select * from t1;
1 row created.
SQL> /
2 rows created.
SQL> /
4 rows created.
SQL> /
8 rows created.
SQL> /
16 rows created.
SQL> /
32 rows created.
SQL> /
64 rows created.
SQL> commit;
Commit complete.
SQL> select count(*) from t1;
COUNT(*)
----------
128
SQL> update t1 set id=rownum,name=rownum;
128 rows updated.
SQL> commit;
Commit complete.
改动裸设备的路径名称并重起集群
[root@node1 ~]# cat /etc/udev/rules.d/60-raw.rules
# Enter raw device bindings here.
#
# An example would be:
# ACTION=="add", KERNEL=="sda", RUN+="/bin/raw /dev/raw/raw1 %N"
# to bind /dev/raw/raw1 to /dev/sda, or
# ACTION=="add", ENV{MAJOR}=="8", ENV{MINOR}=="1", RUN+="/bin/raw /dev/raw/raw2 %M %m"
# to bind /dev/raw/raw2 to the device with major 8, minor 1.
ACTION=="add", KERNEL=="sdb1", RUN+="/bin/raw /dev/raw/raw1 %N"
ACTION=="add", KERNEL=="sdb2", RUN+="/bin/raw /dev/raw/raw2 %N"
ACTION=="add", KERNEL=="sdb5", RUN+="/bin/raw /dev/raw/raw7 %N"
ACTION=="add", KERNEL=="sdb6", RUN+="/bin/raw /dev/raw/raw6 %N"
ACTION=="add", KERNEL=="sdb7", RUN+="/bin/raw /dev/raw/raw5 %N"
ACTION=="add", KERNEL=="sdb8", RUN+="/bin/raw /dev/raw/raw4 %N"
ACTION=="add", KERNEL=="sdb9", RUN+="/bin/raw /dev/raw/raw3 %N"
[oracle@node1 ~]$ export ORACLE_SID=+ASM1
[oracle@node1 ~]$ sqlplus / as sysdba
SQL*Plus: Release 10.2.0.5.0 - Production on Sun Aug 31 13:13:01 2014
Copyright (c) 1982, 2010, Oracle. All Rights Reserved.
Connected to:
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP, Data Mining
and Real Application Testing options
SQL> col name for a30
SQL> col lable for a33
SQL> col path for a33
SQL> set linesize 222
SQL> col label for a31
SQL> /
NAME LABEL PATH
------------------------------ ------------------------------- ---------------------------------
DG1_0000 /dev/raw/raw7
DG2_0000 /dev/raw/raw6
DG3_0000 /dev/raw/raw5
DG3_0001 /dev/raw/raw4
DG4_0000 /dev/raw/raw3
/dev/raw/raw1
/dev/raw/raw2
7 rows selected.
SQL> ho ssh oracle@node2
Last login: Sun Aug 31 12:56:27 2014 from node1
[oracle@node2 ~]$ export ORACLE_SID=+ASM2
[oracle@node2 ~]$ sqlplus / as sysdba
SQL*Plus: Release 10.2.0.5.0 - Production on Sun Aug 31 13:14:31 2014
Copyright (c) 1982, 2010, Oracle. All Rights Reserved.
Connected to:
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP, Data Mining
and Real Application Testing options
SQL> col name for a31
SQL> col label for a31
SQL> col path for a31
SQL> set linesize 211
SQL> select name,label,path from v$asm_disk order by 1;
NAME LABEL PATH
------------------------------- ------------------------------- -------------------------------
DG1_0000 /dev/raw/raw3
DG2_0000 /dev/raw/raw4
DG3_0000 /dev/raw/raw5
DG3_0001 /dev/raw/raw6
DG4_0000 /dev/raw/raw7
/dev/raw/raw2
/dev/raw/raw1
7 rows selected.
SQL> exit
Disconnected from Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP, Data Mining
and Real Application Testing options
[oracle@node2 ~]$ export ORACLE_SID=easy2
[oracle@node2 ~]$ sqlplus / as sysdba
SQL*Plus: Release 10.2.0.5.0 - Production on Sun Aug 31 13:15:31 2014
Copyright (c) 1982, 2010, Oracle. All Rights Reserved.
Connected to:
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit Production
With the Partitioning, Real Application Clusters, OLAP, Data Mining
and Real Application Testing options
SQL> select count(*) from t1;
COUNT(*)
----------
128
SQL> select min(id),max(id) from t1;
MIN(ID) MAX(ID)
---------- ----------
1 128
由此可见，裸设备路径和名称的改变不会影响asm的使用。可是，我们依旧建议保持路径名称的稳定性，方便管理。