Rubin Shrestha
Rubin Shrestha
Blog

How can I remove two failed newly-added SSDs from an HPE Smart Array RAID10 and return to the original 4-disk layout?

by |Published

I have an HPE Smart Array P440ar controller.

Originally I had a 4x SSD RAID 10 array and everything was working correctly.

I then tried to expand the array by adding 2 more SSDs so that I could later increase the capacity. Unfortunately, those two SSDs turned out to be incompatible with the HPE server/controller setup, and I had to remove them physically.

As a result:

  • the controller marked the two newly added drives as failed / hot removed

  • the array pulled in the hot spare

  • the logical drive is now in degraded / interim recovery mode

  • the logical drive itself was not expanded, so the usable logical drive size never changed

My goal is to return to the original 4-SSD RAID10 configuration until I get proper compatible SSDs for expansion.

Current relevant situation:

  • original RAID10 was 4 SSDs

  • I added 2 SSDs to the same array

  • those 2 new SSDs failed / were removed

  • the logical drive size was not increased

  • the array is now degraded

  • the failed disks are no longer physically present

From ssacli, the failed drives are still shown as members of the RAID10 layout, and the logical drive is shown as Interim Recovery Mode.

What is the least destructive / least painful way to remove those two failed added drives from the array configuration and return to the original 4-disk RAID10 layout?

More specifically:

  1. Is there a supported way to shrink an HPE Smart Array RAID10 array by removing the failed added pair?

  2. Can this be done online with ssacli, or does it require offline SSA / reboot?

  3. If the failed drives are already physically gone, is there any safe way to tell the controller to forget them?

  4. If this is not possible, is the only option backup + recreate the array?

I would prefer to avoid destroying and recreating the array if there is any supported rollback path, especially because the logical drive was never extended.

Example current status summary:

  • controller: HPE Smart Array P440ar

  • RAID level: 1+0

  • 2 newly added SSDs now show as failed / hot removed (which is correct)

  • hot spare activated

  • logical drive status: Interim Recovery Mode

  • logical drive size unchanged

Any advice from someone familiar with HPE Smart Array behavior in this situation would be appreciated.

You may also like