A P P E N D I X

A P P E N D I X B

Checklists for Replacing an FC-AL Disk Drive

This section provides a summary of the tasks required to replace a disk drive on a Sun StorEdge A5x00 array or a Sun Fire 880 system running UFS, Volume Manager, or Solstice DiskSuite software. To complete the tasks listed below, you need to complete these procedures:

How to Prepare an FC-AL Drive for Removal

How to Replace an FC-AL Disk Drive

How to Reconfigure an FC-AL Disk Drive

Caution - You must be a qualified system administrator to perform these procedures. Performing hot-plug operations on an active disk drive can result in data loss or data corruption.

Replacing a Disk Drive Using UFS

Identify the faulty disk drive.

Use the df , mount , showmount-a, and ps-ef commands to identify activities or applications attached to the drive you plan to replace.

Stop any activity to the drive.

Back up your system.

Determine and save the partition table for the disk.

Use the umount command to unmount any file systems on the disk.

If your system is running Volume Manager, use the Volume Manager command vxdisk offline to take the disk offline. This is required even if the disk is not being managed by Volume Manager.

Use the luxadm probe , enclosure_name , and display subcommands to determine an address for the drive. An address can be a path name, a WWN, or a box name and slot number.

Use the luxadm remove_device subcommand to remove the disk from the Solaris Operating Environment (deleting the /devices and /dev entries). Use remove_device -F if you are running Solstice DiskSuite software.

Remove the disk drive from the enclosure when prompted by the luxadm remove_device subcommand.

Use the luxadm insert_device subcommand to add the disk to the Solaris Operating Environment (adding the disk's WWN to the /devices and /dev directories).

Insert the disk drive into the enclosure when prompted by the luxadm insert_device subcommand.

Use the prtvtoc and format commands to verify that the partition table is correct for the file system you need.

Use the newfs command to create a new file system on the slice and the mount command to mount the new file system.

Replacing a Disk Drive Using Volume Manager

Use vxdisk to identify the disk you need to replace.

Use vxdiskadm to remove the disk from Volume Manager.

Use the luxadm probe , enclosure_name , and display subcommands to determine an address for the drive. An address can be a path name, a WWN, or a box name and slot number.

Use the luxadm remove_device -F subcommand to take the disk offline from Volume Manager and remove the disk from Solaris (deleting the /devices and /dev entries).

Remove the disk drive from the enclosure when prompted by the luxadm remove_device subcommand.

Use the luxadm insert_device subcommand to add the disk to the Solaris Operating Environment (adding the disks WWN to the /devices and /dev directories) and physically install the drive.

Insert the disk drive into the enclosure when prompted by the luxadm insert_device subcommand.

Use vxdctl enable to notify Volume Manager of the new disk.

Use vxdiskadm to replace the failed disk in Volume Manager.

Replacing a Disk Drive Using Solstice DiskSuite

Use /var/adm/messages and metastat to identify the faulty disk drive.

Use metadb to locate and delete any metadevice database replicas on the faulty disk.

Use metastat to locate any submirrors using slices on the faulty disk.

Use metadetach to detach submirrors that use slices on the faulty disk.

Use metahs to delete hot spares marked "Available" that are on the problem disk.

Use prtvtoc to preserve the disk label if multiple partitions are being used on the disk.

Use metareplace to replace the disk slices if they are not hot spared.

Use the luxadm probe , enclosure_name , and display subcommands to determine an address for the drive. An address can be a path name, a WWN, or a box_name and slot_number.

Use the luxadm remove_device -F subcommand to take the disk offline from Solstice DiskSuite software and remove the disk from the Solaris Operating Environment (deleting the /devices and /dev entries).

Remove the disk drive from the enclosure when prompted by the luxadm remove_device subcommand.

Use the luxadm insert_device command to add the disk to the Solaris Operating Environment (adding the disks WWN to the /devices and /dev directories) .

Insert the disk drive into the enclosure when prompted by the luxadm insert_device subcommand.

If necessary, restore the disk label.

Use metadb -a to add the same number of metadevice state database replicas that were deleted.

Use metahs -a to replace hot spares that were deleted to hot spare pools.

Perform one of the tasks listed in TABLE B-1 , depending on how the slice that failed was used.

TABLE B-1 Slice and command Information
Slice	Action
Simple Slice	Use normal recovery procedures.
Stripe or Concatenation	Run `newfs` for the entire metadevice; restore from backup.
Mirror	Reattach detached submirrors.
RAID5 metadevice	Resync (enable) affected slices.
Trans Metadevice	run `fsck(1M)` .