ONTAP Discussions

whether can solve Spares_Low issue by using a internal spare disk

Daniel_66
642 Views

Hello everyone,

 

        there is a  FAS2720 with a external disk shelf DS212C,  and the disks inside the controller is partitioned, but the disks in external disk shelf is non-partitioned. all of 12 disks in external disk shelf is used to create a data aggr, no spare disk is reserved(Ontap  generates "Spares_Low" Error, because there are only partitioned spares, no whole spare disk) . There are many data in the data aggr, and it's not possible to delete aggr and then recreate.  I wonder whether auto-partition of internal disks  can be disabled, that is,  after assigned, the internal spare is still non-partitioned. if this is the case, whether can the data aggr that is comprised of the external disks share the spare disk in the internal controller chassis?

 

Thanks in advance

3 REPLIES 3

chamfer
466 Views

Hi @Daniel_66,

 

You are asking some questions that are not easy to answer.  Arrays with internal disks would generally use partitioning as they would have their disks partitioned for the root aggregates, which are required as this is the most space efficient.  See more details here:

If you don't want to use drive partitioning then you would need to sacrifice 6x drives for the root aggregates (assuming you have dual HA controllers/nodes).

 

You can rebuild the array with ADP and just 8x internal drives so you would have a spare or two for your external disk shelf.

longren
426 Views

i am form china user.

You may operation in cli # storage raid-options modify -node * -name raid.min_spare_count 0

longren_0-1723690228418.png

 

 

Daniel_66
226 Views

Thank you both for your reply. There have been data in our system, so it can't re-builded.  the option "raid.min_spare_count 0" can be used to close Low_spare waining, but I think unpartitioned or partitioned aggr can share spare to improve the  disk system  security.

I found a KB about that partitioned aggr can use a unpartitioned aggr spare, link below: 

https://kb.netapp.com/on-prem/ontap/Ontap_OS/OS-KBs/ONTAP_is_reverting_the_partitioning_and_returning_the_disk_to_unpartitioned_state

the main meaning is that if partitioned aggr uses external shelf spare (unpartitoned spare) to re-builded, then data will copy back to internal shelf  automatically when there are spares in the internal shelf.(remark: I only see the document, and I haven't verify it in the real environment )

Public