MD3200i - Failed disk, hot spare didn't swapping in + unreadable sectors

Hi,

Can you private message me the service tag so we can get some additional information?

If it is failing the rebuild due to the drive with bad sectors you are going to need to replace the bad drives, create a new array and restore a backup. http://www.dell.com/support/article/us/en/4/SLN111497

Hi Josh,

PM with service tag sent.

Thanks,

Brian

Thanks, Do you have a backup of the data?

The virtual disk is a cluster shared volume in a hyperV cluster - I have a backups at the guest OS level, but not at the LUN level.

It's still operational, so I can assign a driver letter in the cluster host and take a backup of the virtual disk data that way when all the VMS are shutdown.

Recovery guru is only showing one virtual disk as degraded - I was unsure if other virtual disks were affected (i.e. did I need to back up the full storage array).

Can I get a quote for a replacement disk, or disks?

Is there any alternative to deleting the virtual disk?

Is it possible to tell whether the disk will continue to degrade to the point where the virtual disk will go offline?

Thanks.,

Brian

If it is not rebuilding there is not another way around recreating the array, since the bad blocks on the drive are preventing the rebuild. Unfortunately, I am not able to do quotes, it has to come from the sales department.  

Thanks Josh,

>Recovery guru is only showing one virtual disk as degraded - I was unsure if other virtual disks were affected (i.e. did I need to back up the full storage array).

Is it possible that only one virtual disk is affected?

Yes, it depends on where the bad blocks are located, they could all be in one virtual disk.

Hi Josh,

Would it be possible for someone to have a look at the support bundle so that I can be sure that the correct drives are replaced?

Thanks,.

Brian

Yes, I sent you a PM.

Support bundle sent as requested.

Thanks,

Brian

Drive 7 is showing failed and drive 10 is impending failure, so those drives should be replaced. The other virtual disks look fine.

Hi Josh,

OK, so just to confirm the process:

1) Copy my data off the degraded virtual disk

2) Delete the degraded virtual disk

3) Insert a new disk in slot 7 and assign it as the global hot spare

4) Fail the disk in slot 10 and the hot spare in disk 7 should replace it

5) Insert a new disk in slot 10

6) Create a new virtual disk

7) Copy my data back

I'm worried about the data on the other virtual disk - I though that all disks were striped across all drives (excluding the hot spare), so I can't have more that two failed drives  at any one time.

There is currently a reconstruction operation in progress (since I swapped in the original hot spare) Can that operation be cancelled?

Thanks,

Brian

It is one big raid 5 for the disk group, so not just that virtual disk will be affected by the replacement of the drives, they all will be. Sorry if I didn’t make it clear. The reason why only that one virtual disk was affected was the bad blocks just happened to be located on the same virtual disk.  So you will have to recreate all of the virtual disks and restore all of the data.

Since the drive is currently reconstructing you could let it finish then replace the drive that has the pending failure, then once that finishes rebuilding delete just that one virtual disk and then recreate it and restore the backup.