MSA2050 Replacing a disk in a RAID5 disk-group

[Alpaca]

Hi there, I have different disk-group with different drive capacities and RAIDs. one disk-group is made up of 900 GD drives on RAID5 (say, gpa-R5-9), another is made up of 6000 GB drives and RAID5 (say, gpb-R5-6) , and another is made up of 2.4 TB drives and RAID5 (say, gpc-R5-2.4).

gpa-R5-9 now has one 2.4 TB drive while the rest of its drives are 9000 GB.

We have 600 GB and 900 GB Global spare (newly configured) with none for 2.4 TB. There is also one available 9000 GB (which I intend to make a Global spare as well).

My suspicion is that the 2.4 TB drive in gpa-R5-9 was a Global spare and kicked in there since it was the only global spare equal/greater than 9000 GB.

How do I remove the 2.4 TB drive from gpa-R5-9 and let it failover to the new 9000 GB Global space and then configure the 2.4 TB drive as a Global space? Going through the Wed UI, there is no feature to from a drive from a group. Your expert reply will be appreciated. Thank you,

[ArunKKR · Accepted solution]

Hi,

I believe that all the hard drive in the arry are standard tier hard drives (10k/15k rpm SAS hard drives).
You might need to remove global spare configuration and disable dynamic sparing (if enabled) to ensure that larger capacity hard drive does not kick in when a lower capacity hard drive fails in the future.
You could configure a drive as global spare manually when ever there is a drive failure.
Unfortunately this is the only option when you have hard drives of different capacity in an array.

1. Click on System > Action > Advanced settings. Ensure that dynamic sparing is unchecked.

2. Click on Pools > Action > Global spare . Remove all disks configured as global spares in the list. 

3. Physically remove the 2.4TB disk which is part of gpa-R5-9 disk group out by 1-2 inches from the chassis and reinsert it after 1-2 minutes.

4. The 2.4TB  disk would appear in left over state. Clear the metadata on the drive. Click on System > Action > Clear metadata.

5. Configure 900GB drive as global spare and allow it join disk group and reconstruction to get completed.

Please note that there is a very small possibility of ghost / duplicate disk group appearing when you remove and reinsert 2.4TB drive.

If that happens only way to clear the configuration on the 2.4TB disk is by logging an HPE support ticket which requires a valid support contract/warranty.

[ArunKKR]

Hi , We wanted to follow up and check if the information provided addressed the issue. Request you to please respond with a Yes/No. You may also click on "Accept as a solution" button if the issue is resolved.

[Alpaca]

@ArunKKR 

Thanks so much for your support. Your support gave me the confidence to take the following steps:

- Created a Global spare of the 900 GB

- Pulled out the 2.4 TB drive. The 900 GB kicked in immediately and the RAID started rebuilding.

-After the rebuild completed, put back the 2.4 TB, which appeared as a fromer member of a RAID and wouldn't allow being configured as a spare.

- Select System on the left, click Action and select Clear Metadata. Select the 2.4 TB drive and clear its metadata. 

- You can now configure the drive as a global spare. Either, System -> Action -> Configure Spare.

[ArunKKR]

HI @Alpaca ,

Glad to know that the issue has been resolved.