IBM Storwize V7000 更换硬盘快速维护手册

Storwize V7000 drive replacement procedure

Before removing a drive, you need to make sure it is safe to do so.

Go to the Physical Storage>Internal panel in the GUI

Select the drive to be removed, and select Actions -> Show Dependent Volumes

If any volumes are reported, do not continue.

WARNING: If you are intending to remove multiple drives, then you need to select all of the drives to be removed (using shift click or control click) and check for redundancy as a group, rather than individually.

If the drive has a use of 'member', then use the following steps to mark the drive as failed before continuing:

Go to the Physical Storage>Internal panel in the GUI Select the drive to be removed, and select Actions -> Take Offline

Follow the actions in the Troubleshooting, Recovery, and Maintenance Guide to perform the physical reseat/replaceme.

If this is a reseat procedure

Mark the error as fixed

This will cause the system to re-enable the SAS Ports on the drive, and the drive will come back online.

The drive will be in the failed state., and the following actions are needed to bring it back into the system:

Go to the Physical Storage>Internal panel in the GUI

Select the drive which was reseated, and select Actions -> Mark as -> Candidate

Select the drive which was reseated, and select Actions -> Mark as -> Spare

Once the drive is marked as a spare, the RAID system will

automatically queue up a member exchcange to swap the current array member with the reseated drive. This member exchange will begin once the rebuild operation has completed and will swap the two drives without reducing

redundancy.

If this is a replace procedure

then the old drive needs to be 'cleaned up', and the new drive needs to be configured for use:

Remove ('Clean up') the Old Drive with the following procedure

Go to the Physical Storage>Internal panel in the GUI

Select the old drive (will be offline) and select Actions -> Mark as -> Unused ( Note the drive must already be marked as failed ! )

This will cause the offline drive object to be deleted from the system

Make New Drive Useable

Select the new drive id (currently use=unused) and select Actions -> Mark as -> Candidate followed by Actions -> Mark as -> Spare

If the drive in this slot had previously been a member of an array, then once the drive is marked as a spare, the RAID system will automatically queue up a member exchcange to swap the current array member with the new drive.

This member exchange will begin once the rebuild operation has completed and will swap the two drives without reducing redundancy.

If the new drive is not a perfect replacement (size, RPM & tech_type ) for the old drive, additional steps may be required. See the procedure for Error Code 1692 for the additional steps.

Show dependent volumes => svcinfo lsdependentvdisks -drive <list of colon delimited drive IDs>
Take Offline => svctask chdrive -use failed <drive ID>
Mark as Unused => svctask chdrive -use unused <drive ID>
Mark as Candidate => svctask chdrive -use candidate <drive ID>
Mark as Spare => svctask chdrive -use spare <drive ID>

关于Show Dependent Volumes的结果， 请忽略CMMVC6912E的warning。 需要关注Storage Pool这一栏中是否有内容。