Lab environment: Oracle 11g R2 RAC (11.2.0.3.5) Oracle Enterprise Linux 5.6x86
1. Create an ASM disk group and mount the disk on two nodes. Then, partition the disk on the first node. [root @ Rac1 ~] # Fdisk-ldisk/dev/SDE: 2147 MB, 2147483648 bytes255 heads, 63 sectors/track, 261 cylindersunits = cylinders of 16065*512 = 8225280 bytes
Device boot start end blocks ID system/dev/sde1 1 261 2096451 83 Linux
Create an ASM disk on node 1st [root @ Rac1 ~] # Oracleasm createdisk goldengate/dev/sde1 [root @ Rac1 ~] # Oracleasm scandisks
[Root @ Rac1 ~] # Oracleasm listdisksdatafragoldengateocr_vote scan the ASM disk on node 2nd [root @ rac2 ~] # Oracleasm listdisksdatafragoldengateocr_vote
Switch to the grid user on the first node and execute asmca to create the ASM disk group. Create the Oracle advm volume and diskgroup on the ASM diskgroup.
COMPATIBLE.ASM
And
COMPATIBLE.ADVM
The compatibility must be set to 11.2.0.2. You can select "advanced options" when creating a disk group to set this attribute for diskgroup. You can also use SQL commands to create or modify this attribute for a disk group. The method is as follows:
ALTER DISKGROUP goldengate SET ATTRIBUTE 'compatible.asm' = '11.2.0.2';
ALTER DISKGROUP goldengate SET ATTRIBUTE 'compatible.rdbms' = '11.2.0.2';
ALTER DISKGROUP goldengate SET ATTRIBUTE 'compatible.advm' = '11.2.0.2';
2. Create an oracle advm volume
Switch to the grid user and enter the asmcmd command line interface Su-grid
[grid@rac1 ~]$ export ORACLE_SID=+ASM1[grid@rac1 ~]$ asmcmd
Verify the actual size of the created ASM disk.
ASMCMD> lsdgState Type Rebal Sector Block AU Total_MB Free_MB Req_mir_free_MB Usable_file_MB Offline_disks Voting_files NameMOUNTED EXTERN N 512 4096 1048576 5114 3027 0 3027 0 N DATA/MOUNTED EXTERN N 512 4096 1048576 5114 4613 0 4613 0 N FRA/MOUNTED EXTERN N 512 4096 1048576 2047 29 0 29 0 N GOLDENGATE/MOUNTED EXTERN N 512 4096 1048576 2047 1651 0 1651 0 Y OCR_VOTE/
Create an oracle advm volume
ASMCMD> volcreate -G GOLDENGATE -s 1900m GOLDENGATE
Note that the value specified by the-s option must not exceed the actual size value displayed by the lsdg. Otherwise, the following error is reported:
ASMCMD> volcreate -G GOLDENGATE -s 1950m GOLDENGATEORA-15032: not all alterations performedORA-15041: diskgroup "GOLDENGATE" space exhausted (DBD ERROR: OCIStmtExecute)
Determine whether the Oracle advm volume is created
ASMCMD> volinfo -G GOLDENGATE GOLDENGATEDiskgroup Name: GOLDENGATE Volume Name: GOLDENGATE Volume Device: /dev/asm/goldengate-257 State: ENABLED Size (MB): 1920 Resize Unit (MB): 32 Redundancy: UNPROT Stripe Columns: 4 Stripe Width (K): 128 Usage: Mountpath:
SQL> col volume_device for a30SQL> col volume_name for a30SQL> SELECT volume_name, volume_device FROM V$ASM_VOLUME 2 WHERE volume_name ='GOLDENGATE';VOLUME_NAME VOLUME_DEVICE------------------------------ ------------------------------GOLDENGATE /dev/asm/goldengate-257
3. Create an oracle ACFs File System
[root@rac1 ~]# /sbin/mkfs -t acfs /dev/asm/goldengate-257mkfs.acfs: version = 11.2.0.3.0mkfs.acfs: on-disk version = 39.0mkfs.acfs: volume = /dev/asm/goldengate-257mkfs.acfs: volume size = 2013265920mkfs.acfs: Format complete.
Create an ACFs file system mount point on node 1 and node 2.
[root@rac1 ~]# su - oracle[oracle@rac1 ~]$ mkdir -p /home/oracle/ggs
[root@rac2 ~]# su - oracle[oracle@rac2 ~]$ mkdir -p /home/oracle/ggs
Register the just-created ACFs File System in the cluster Mount registery or manually mount it without registration.
The advantage of registering an ACFs file system is that when the cluster checks Mount registery every 30 seconds, the file system is automatically mounted on each node, when the cluster software or system is restarted, the file system is automatically mounted.
[root@rac1 ~]# /sbin/acfsutil registry -a /dev/asm/goldengate-257 /home/oracle/ggsacfsutil registry: mount point /home/oracle/ggs successfully added to Oracle Registry
The command for manual mounting is as follows:
[root@rac1 ~]# /bin/mount -t acfs /dev/asm/goldengate-257 /home/oracle/ggs
Wait 30 seconds after running the registration command, and then check whether the ACFs file system is mounted on each node.
[root@rac1 ~]# df -hFilesystem Size Used Avail Use% Mounted on/dev/sda1 30G 20G 7.7G 73% //dev/sda3 1003M 39M 914M 5% /tmptmpfs 1.3G 454M 814M 36% /dev/shm/dev/asm/goldengate-257 1.9G 73M 1.9G 4% /home/oracle/ggs
[oracle@rac2 ~]$ df -hFilesystem Size Used Avail Use% Mounted on/dev/sda1 30G 21G 7.3G 74% //dev/sda3 1003M 39M 914M 5% /tmptmpfs 1.3G 450M 818M 36% /dev/shm/dev/asm/goldengate-257 1.9G 137M 1.8G 8% /home/oracle/ggs
4. Test the File System
[root@rac1 ~]# chown -R oracle:oinstall /home/oracle/ggs
First Node
[oracle@rac1 ~]$ echo "Oracle ACFS File System" > /home/oracle/ggs/acfs.log
[oracle@rac1 ggs]$ more /home/oracle/ggs/acfs.logOracle ACFS File System
Section 2
[oracle@rac2 ~]$ more /home/oracle/ggs/acfs.logOracle ACFS File System
For reprinting, please indicate the author's source and original article links; otherwise, you will be held legally responsible:
Author: xiangsir
Link: http://blog.csdn.net/xiangsir/article/details/10007779
QQ: 444367417
MSN: xiangsir@hotmail.com