AFF

Align spares

Baiju625
901 Views

Hi Experts,

 

Clus1::> storage aggregate show-spare-disks -original-owner n1

Original Owner: n1
Pool0
Root-Data1-Data2 Partitioned Spares
Local Local
Data Root Physical
Disk Type Class RPM Checksum Usable Usable Size Status
---------------- ------ ----------- ------ -------------- -------- -------- -------- --------
1.0.23 SSD-NVM
solid-state - block 862.9GB 62.35GB 1.75TB zeroed

clus1::> storage aggregate show-spare-disks -original-owner -n2

Original Owner: 2
Pool0
Root-Data1-Data2 Partitioned Spares
Local Local
Data Root Physical
Disk Type Class RPM Checksum Usable Usable Size Status
---------------- ------ ----------- ------ -------------- -------- -------- -------- --------
1.0.5 SSD-NVM
solid-state - block 0B 62.35GB 1.75TB zeroed
1.0.23 SSD-NVM
solid-state - block 862.9GB 0B 1.75TB zeroed
2 entries were displayed.

 

This o/p is from a AFF400 system with 12 SSD and using root-data-data partition. Do I need to align the spares on n2?

if yes what is the correct CLI

 

storage disk replace  -disk 1.0.5  -replacement 1.0.23 -action start 

or

storage disk replace  -disk 1.0.23  -replacement 1.0.5 -action start 

 

Also noticed that spare-core status is false, what needs to be done so that it is enabled? please advise.

 

clus1::> storage aggregate show-spare-disks -fields is-sparecore
original-owner disk is-sparecore
----------------- ------ ------------
n2 1.0.5 false
n1 1.0.23 false
n2 1.0.23 false
3 entries were displayed.

 

 

 

 

 

 

5 REPLIES 5

TMACMD
889 Views

How about starting with this output instaead:

storage aggregate show-spare-disks

 

aladd
844 Views

Spares should automatically be distributed appropriately.

 

The spare-core status is to specify if there is a spare core present on any of these disks. A spare core is generated when a panic occurs. This is normal output.

Baiju625
834 Views

Here is the requested o/p.I am referring to ONTAP 9.8 "Disk and aggregate management" , page 16.

It says:

"When you add partitioned disks to an aggregate, you must leave a disk with both the root and data
partition available as spare for every node. If you do not and your node experiences a disruption,
ONTAP cannot dump the core to the spare data partition."

1) Does this mean on the n2 node it needs both root and spare partition on 1.0.5 disk? 

 

 

Clus1::> storage aggregate show-spare-disks

Original Owner: n1
Pool0
Root-Data1-Data2 Partitioned Spares
Local Local
Data Root Physical
Disk Type Class RPM Checksum Usable Usable Size Status
---------------- ------ ----------- ------ -------------- -------- -------- -------- --------
1.0.23 SSD-NVM
solid-state - block 862.9GB 62.35GB 1.75TB zeroed

Original Owner: n2
Pool0
Root-Data1-Data2 Partitioned Spares
Local Local
Data Root Physical
Disk Type Class RPM Checksum Usable Usable Size Status
---------------- ------ ----------- ------ -------------- -------- -------- -------- --------
1.0.5 SSD-NVM
solid-state - block 0B 62.35GB 1.75TB zeroed
1.0.23 SSD-NVM
solid-state - block 862.9GB 0B 1.75TB zeroed
3 entries were displayed.

aladd
831 Views

spare core is when we dump a core file to a spare disk following a panic.

 

in the event we don't have a spare disk available, a spray core is performed. It seems you have a spare disk on each node, which should be sufficient for a coredump as needed.

 

 

andris
764 Views

Except... the AFF A400 platform (as do most of the newer models) dumps a coredump into a special partition on the boot device, now.  So, the content about needing spare partitions for coredump no longer applies. That said, you still want spare partitions for RAID recovery operations.

Public