Asset ID: |
1-72-2299466.1 |
Update Date: | 2017-09-29 |
Keywords: | |
Solution Type
Problem Resolution Sure
Solution
2299466.1
:
ORA-15018 ORA-15031 Creating New ASM Diskgroup In Exadata
Related Categories |
- PLA-Support>Eng Systems>Exadata/ODA/SSC>Oracle Exadata>DB: Exadata_EST
|
In this Document
Created from <SR 3-15421304501>
Applies to:
Exadata X3-2 Hardware - Version All Versions to All Versions [Release All Releases]
Linux x86-64
Symptoms
On : NA version, MICC Transfer to TSC
ACTUAL BEHAVIOR
---------------
2 new X4-2 Exadata Storage Servers were added into the X3-2 Exadata Quarter Rack
X4-2 Exadata Storage Servers have hard disks larger than X3-2 Exadata Storage Srerver.
Hence, each hard disk in X4-2 Exadata Storage Servers has 500G free space which remains after customer added new ASM disks from X4-2 Exadata Storage Servers into the existing ASM disk groups
So my customer wants to create a new ASM disk group 'DATA2' using the remaining disk partitions.
My customer failed to create a new ASM disk group hitting the following errors
ORA-15018: diskgroup cannot be created
ORA-15031: disk specification 'o/X.X.X.9/DATA2*' matches no disks
ORA-15031: disk specification 'o/X.X.X.7/DATA2*' matches no disks
Changes
2 new X4-2 Exadata Storage Servers were added into the X3-2 Exadata Quarter Rack
Cause
1) LUNs, Cell disks, and Grid disks have been already generated when 2 new X4-2 Exadata Storage Servers were added into the X3-2 Exadata Quarter Rack
2) However, the status of the targeted GRID Disks is INACTIVE like below
+ griddisk-detail.out
name: DATA2_CD_00_celll5
asmDiskGroupName:
asmDiskName:
asmFailGroupName:
availableTo:
cachedBy: "FD_04_celll5, FD_05_celll5, FD_07_celll5, FD_06_celll5"
cachingPolicy: default
cellDisk: CD_00_celll5
comment:
creationTime: 2014-09-25T10:15:27+09:00
diskType: HardDisk
errorCount: 0
id: 3b41b8dc-e8da-429a-b05c-6ea137bdaba2
offset: 557.859375G
size: 524.984375G
status: inactive
name: DATA2_CD_01_celll5
asmDiskGroupName:
asmDiskName:
asmFailGroupName:
availableTo:
cachedBy: "FD_13_celll5, FD_12_celll5, FD_14_celll5, FD_15_celll5"
cachingPolicy: default
cellDisk: CD_01_celll5
comment:
creationTime: 2014-09-25T10:15:28+09:00
diskType: HardDisk
errorCount: 0
id: 0ede50c4-e6be-47b0-8845-ca40032d2ea7
offset: 557.859375G
size: 524.984375G
status: inactive
(omission)
3) We need to check the DISK Path for the candidate grid disks. To do it, we can use kfod or query v$asm_disk
Below is the example for getting the information using kfod utility:
We can see that the PATH of all candidate disks is 'o/X.X.X.7;X.X.X.8/DATA2_*' or 'o/X.X.X.9;X.X.X.10/DATA2_*'
[root # kfod asm_diskstring='o/x.x.x.7;x.x.x.8/DATA2_*' disks=ALL
--------------------------------------------------------------------------------
Disk Size Path User Group
================================================================================
1: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_00_cell4
2: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_01_cell4
3: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_02_cell4
4: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_03_cell4
5: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_04_cell4
6: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_05_cell4
7: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_06_cell4
8: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_07_cell4
9: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_08_cell4
10: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_09_cell4
11: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_10_cell4
12: 537584 Mb o/x.x.x.7;x.x.x.8/DATA2_CD_11_cell4
--------------------------------------------------------------------------------
ORACLE_SID ORACLE_HOME
================================================================================
+ASM1 /u01/app/11.2.0.4/grid
+ASM2 /u01/app/11.2.0.4/grid
[root # kfod asm_diskstring=,'o/x.x.x.9;x.x.x.10/DATA2_*' disks=ALL
--------------------------------------------------------------------------------
Disk Size Path User Group
================================================================================
1: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_00_cell5
2: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_01_cell5
3: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_02_cell5
4: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_03_cell5
5: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_04_cell5
6: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_05_cell5
7: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_06_cell5
8: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_07_cell5
9: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_08_cell5
10: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_09_cell5
11: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_10_cell5
12: 537584 Mb o/x.x.x.9;x.x.x.10/DATA2_CD_11_cell5
--------------------------------------------------------------------------------
Solution
Below are the STEPS for creating a new ASM Disk group in Exadata environment
Step 1) Change the status of all candidate grid disks from all targeted to ACTIVE using the following command
For example)
CellCLI> ALTER GRIDDISK gdisk_name ACTIVE
CellCLI> LIST GRIDDISK
Step 2) Check if candidate disks are identified using kfod utility
For example)
# kfod asm_diskstring='o/x.x.x.7;x.x.x.8/DATA2_*' disks=ALL
# kfod asm_diskstring=,'o/x.x.x.9;x.x.x.10/DATA2_*' disks=ALL
Step3) If all candidate disks are identified from step 2, now you can create a new ASM disk group with them.
At this time, you need to input the exact DISK PATH which is gotten from step 2
and you can also get ASM disk attributes of the other existing ASM disks from v$asm_attribute view
For example)
CREATE DISKGROUP DATA2 NORMAL REDUNDANCY
DISK 'o/x.x.x.7;x.x.x.8/DATA2_*','o/x.x.x.9;x.x.x.10/DATA2_*' ATTRIBUTE 'content.type' = 'DATA',
'AU_SIZE' = '4M',
'cell.smart_scan_capable'='TRUE',
'compatible.rdbms'='11.2.0.2',
'compatible.asm'='11.2.0.3';
References
<NOTE:1683842.1> - SRDC - EEST Sundiag
Attachments
This solution has no attachment