NetApp E5700 - Page 65

• [drawerID] is the identifier of the drive drawer that you want to replace. Drawer ID values
are 1 (top drawer) to 5 (bottom drawer). You must enclose the value for drawerID in square
brackets.
Example
This command ensures you can remove the top-most drawer in drive shelf 10:
SMcli <ctlr_IP1> -p "safety-1" -c "set tray [10] drawer [1]
serviceAllowedIndicator=forceOnWarning;"
3.
Determine if you need to stop host I/O activity, as follows:
• If the command succeeds, you do not need to stop host I/O activity. All drives in the drawer
are in pools or volume groups with Drawer Loss Protection. Go to
Remove cable chains on
page 66.
Attention: Possible damage to drives – Wait 30 seconds after the command completes
before you open the drive drawer. Waiting 30 seconds allows the drives to spin down, which
prevents possible damage to the hardware.
• If a warning is displayed indicating that this command could not be completed, you must stop
host I/O activity before removing the drawer. The warning is displayed because one or more
drives in the affected drawer are in pools or volume groups without Drawer Loss Protection.
To avoid losing data, you must complete the next steps to stop host I/O activity and to power
off the drive shelf and the controller shelf.
4. Ensure that no I/O operations are occurring between the storage array and all connected hosts. For
example, you can perform these steps:
• Stop all processes that involve the LUNs mapped from the storage to the hosts.
• Ensure that no applications are writing data to any LUNs mapped from the storage to the
hosts.
• Unmount all file systems associated with volumes on the array.
Note: The exact steps to stop host I/O operations depend on the host operating system and the
configuration, which are beyond the scope of these instructions. If you are not sure how to stop
host I/O operations in your environment, consider shutting down the host.
5. If the storage array participates in a mirroring relationship, stop all host I/O operations on the
secondary storage array.
Attention: Possible data loss – If you continue this procedure while I/O operations are
occurring, the host application might lose data because the storage array will not be accessible.
6. Wait for any data in cache memory to be written to the drives.
The green Cache Active LED on the back of each controller is on when cached data needs to be
written to the drives. You must wait for this LED to turn off.
Drives |
65