Identify Candidate Disks

The current disk group configuration, (TESTDB_DATA1 and candidate disks not assigned to any disk group) has the following configuration:

$ ORACLE_SID=+ASM; export ORACLE_SID $ sqlplus "/ as sysdba" SELECT NVL(a.name, '[CANDIDATE]') disk_group_name , b.path disk_file_path , b.name disk_file_name , b.failgroup disk_file_fail_group FROM v$asm_diskgroup a RIGHT OUTER JOIN v$asm_disk b USING (group_number) ORDER BY a.name; Disk Group Name Path File Name Fail Group --------------- --------------- -------------------- --------------- TESTDB_DATA1 /dev/raw/raw1 TESTDB_DATA1_0000 CONTROLLER1 /dev/raw/raw2 TESTDB_DATA1_0001 CONTROLLER1 /dev/raw/raw3 TESTDB_DATA1_0002 CONTROLLER2 /dev/raw/raw4 TESTDB_DATA1_0003 CONTROLLER2 [CANDIDATE] /dev/raw/raw5 /dev/raw/raw6 /dev/raw/raw7

In this example, I will be adding two new disks (/dev/raw/raw5 and /dev/raw/raw6) to the current disk group.

Add Disks to a Disk Group

Finally, let's add the two new disks to the disk group. This needs to be done within the ASM instance and connected as a user with SYSDBA privileges:

$ ORACLE_SID=+ASM; export ORACLE_SID $ sqlplus "/ as sysdba" SQL> ALTER DISKGROUP testdb_data1 ADD 2 FAILGROUP controller1 DISK '/dev/raw/raw5' 3 FAILGROUP controller2 DISK '/dev/raw/raw6' REBALANCE POWER 11; Diskgroup altered.

After submitting the SQL to add the new disks to the disk group, query the dynamic performance view V$ASM_OPERATION in the ASM instance to check the status of the rebalance operation. The V$ASM_OPERATION view will return one row for every active Automatic Storage Management long running operation executing in the Automatic Storage Management instance.

SQL> SELECT group_number, operation, state, power, est_minutes FROM v$asm_operation; GROUP_NUMBER OPERATION STATE POWER EST_MINUTES ------------ --------- ------ ------ ----------- 1 REBAL RUN 11 9

Continue to query this view to monitor the rebalance operation. When this row is gone, the ASM rebalance operation will be complete.

After adding the new disks, this is a new view of the disk group configuration:

Disk Group Name Path File Name Fail Group --------------- --------------- -------------------- --------------- TESTDB_DATA1 /dev/raw/raw1 TESTDB_DATA1_0000 CONTROLLER1 /dev/raw/raw2 TESTDB_DATA1_0001 CONTROLLER1 /dev/raw/raw3 TESTDB_DATA1_0002 CONTROLLER2 /dev/raw/raw4 TESTDB_DATA1_0003 CONTROLLER2 /dev/raw/raw5 TESTDB_DATA1_0004 CONTROLLER1 /dev/raw/raw6 TESTDB_DATA1_0005 CONTROLLER2 [CANDIDATE] /dev/raw/raw7

Drop Disks from a Disk Group

Now, let's drop the same two new disks from the disk group. This needs to be done within the ASM instance and connected as a user with SYSDBA privileges:

$ ORACLE_SID=+ASM; export ORACLE_SID $ sqlplus "/ as sysdba" SQL> ALTER DISKGROUP testdb_data1 DROP 2 DISK testdb_data1_0004, testdb_data1_0005 REBALANCE POWER 11; Diskgroup altered.

The current disk group configuration, (TESTDB_DATA1 and candidate disks not assigned to any disk group) now has the following configuration:

Disk Group Name Path File Name Fail Group --------------- --------------- -------------------- --------------- TESTDB_DATA1 /dev/raw/raw1 TESTDB_DATA1_0000 CONTROLLER1 /dev/raw/raw2 TESTDB_DATA1_0001 CONTROLLER1 /dev/raw/raw3 TESTDB_DATA1_0002 CONTROLLER2 /dev/raw/raw4 TESTDB_DATA1_0003 CONTROLLER2 [CANDIDATE] /dev/raw/raw5 /dev/raw/raw6 /dev/raw/raw7