Storage Administration - Oracle Exadata Recipes: A Problem-Solution Approach - page 261

Now, proceed to dropping your Exadata cell disks. First, query the cell disk name and interleaving attribute:
[oracle@cm01dbm01 ~]$ dcli -g ./cell_group cellcli -e list celldisk where diskType=HardDisk
attributes name,interleaving
cm01cel01: CD_00_cm01cel01 normal_redundancy
cm01cel01: CD_01_cm01cel01 normal_redundancy
cm01cel01: CD_02_cm01cel01 normal_redundancy
... output omitted
Next, issue the following cellcli command through dcli to drop your cell disks:
[oracle@cm01dbm01 ~]$ dcli -g ./cell_group cellcli -e drop celldisk all harddisk force
cm01cel01: CellDisk CD_00_cm01cel01 successfully dropped
cm01cel01: CellDisk CD_01_cm01cel01 successfully dropped
cm01cel01: CellDisk CD_02_cm01cel01 successfully dropped
cm01cel01: CellDisk CD_03_cm01cel01 successfully dropped
... output omitted for brevity
Create New Cell Disks and Grid Disks
After the cell disks are dropped, proceed to recreate them using the following cellcli command. In this example, we
will not specify cell disk interleaving but cover this topic in Recipe 9-8.
[oracle@cm01dbm01 ~]$ dcli -g ./cell_group cellcli -e create celldisk all harddisk
cm01cel01: CellDisk CD_00_cm01cel01 successfully created
cm01cel01: CellDisk CD_01_cm01cel01 successfully created
cm01cel01: CellDisk CD_02_cm01cel01 successfully created
... output omitted for brevity
Validate your newly created cell disks using the following cellcli command:
[oracle@cm01dbm01 ~]$ dcli -g ./cell_group cellcli -e list celldisk where \
> disktype=HardDisk attributes name,interleaving
cm01cel01: CD_00_cm01cel01 none
cm01cel01: CD_01_cm01cel01 none
cm01cel01: CD_02_cm01cel01 none
cm01cel01: CD_03_cm01cel01 none
... output omitted
Next, you must rebuild your Exadata grid disks. Using information documented and captured from previous steps in
this recipe, we will start by building our DATA , SDATA , RECO , and SRECO grid disks. Your grid disk prefixes may be different
from these, but the DATA and RECO grid disk prefixes are common in many organizations. It is considered a best practice
to create your grid disks uniformly across each storage cell in your Exadata storage grid, so this is what we will show:
[oracle@cm01dbm01 ~]$ dcli -g ./cell_group cellcli -e create griddisk all \
> harddisk prefix='DATA', size=284G
cm01cel01: GridDisk DATA_CD_00_cm01cel01 successfully created
cm01cel01: GridDisk DATA_CD_01_cm01cel01 successfully created
cm01cel01: GridDisk DATA_CD_02_cm01cel01 successfully created
... output omitted
[oracle@cm01dbm01 ~]$ dcli -g ./cell_group cellcli -e create griddisk all \
> harddisk prefix='SDATA', size=141G