IBM totalstorage fastt100, Инструкция по установке, эксплуатации и обслуживанию, Страница: 106 / 160

Replacing the drives one at a time
Use this procedure to replace all the drives one at a time. You cannot use this
procedure on RAID 0 logical drives (use the procedure in “Replacing all drives at
the same time” on page 74).
Note: If your storage subsystem has hot spares assigned, you might want to
unassign the hot spares while you perform this procedure. If you do not,
reconstruction might start on the hot spare before you insert the new drive.
The data on the new drive is still rebuilt, but the process takes longer for
each drive. Remember to reassign the hot spares when you are finished with
this procedure.
To replace the drives one at a time, perform the following steps:
1. Read the following information:
v “Adding larger-capacity drives” on page 73, particularly the paragraphs that
discuss the differences between the two possible upgrade procedures
v
Your software documentation regarding drive upgrades and installation
v
The documentation that comes with the new drives
Read all precautionary notes, kit instructions, and other information. Kit
instructions often contain the most current information regarding the drives
and their installation, plus upgrade or servicing procedures. Compare the kit
instructions with this procedure to determine if you need to modify this
procedure.
2. Use the Storage Manager client to check the status of the unit. Correct any
problems that are reported.
3. Back up the data on the drives that you are replacing.
Attention: When you handle static-sensitive devices, take precautions to
avoid damage from static electricity. For details about handling static-sensitive
devices, see “Handling static-sensitive devices” on page 69.
4. Unpack the new drives.
Set the drives on a dry, level surface away from magnetic fields. Save the
packing material and documentation in case you need to return the drives.
5. Use the storage management software to ensure that the array that was
defined using these drives is in optimal (not degraded) state before manually
failing the first drive you want to replace. If the array is in degraded state, use
the recovery procedures to bring the array into the optimal state.
Make sure that:
v
You fail only one drive
v
The software status display shows a failed status for the appropriate drive
v
The amber Drive fault LED (on the front bezel below the drive) is on
Attention: Removing the wrong drive can cause data loss. Make sure that
you remove only the failed drive CRU. The Drive fault LED below the failed
drive CRU should be on.
If you remove an active drive accidentally, wait at least 30 seconds and then
reinstall it. Because you failed two drives in a RAID array, the array might be
marked failed by the controller. This array will not be available to the host for
I/O. See the Storage Manager client for further recovery instructions. Do not
attempt to replace any drives until the array is brought back to optimal state.
76
IBM TotalStorage FAStT100 Storage Server: Installation, User’s, and Maintenance Guide