This section describes the tasks you can perform on disks.
To display information about the disks that comprise a storage aggregate, use the storage disk show command. By default, the command displays the following information about all disks:
Disk name. See “Disk Naming Conventions” for information about disk names.
Size of the disk, in MB.
Name of the disk shelf on which the disk is located.
Name of the bay in which the disk is located.
Disk state. Possible values include partner, broken, zeroing, spare, copy, pending, reconstructing, and present.
RAID type. Possible values include pending, parity, dparity, or data.
Aggregate name.
Name of the node that owns the disk.
Other information is displayed in a detailed view and if other parameters are specified. See the reference page for the command for details.
The following example displays information about all disks that are used as data disks in the RAID group:
node::>storage disk show -raidtype dataDisk UsedSize(MB) Shelf Bay State RAID Type Aggr Owner node1a:0c.18 135979 1 2 present data node_ node1a 1_a_aggr0 node1a:0c.21 135979 1 5 present data test1 node1a node1a:0c.24 135979 1 8 present data foobar node1a node1a:0c.27 135979 1 11 present data test3 node1a node1b:0c.18 135979 1 2 present data node_ node1b 1_b_aggr0 node1b:0c.21 135979 1 5 present data node_ node1b 1_b_test1 node1b:0c.24 135979 1 8 present data test4 node1b node1b:0c.27 135979 1 11 present data test5 node1b
To modify the owning node of a disk, use the storage disk modify. For most operations, the command prompts you for confirmation before proceeding.
The following example changes the owning node of a disk named node1b:0c.27 to a node named node1a.
node::>storage disk modify -disk node1b:0c.27 -owner node1a
The storage disk setstate command sets the state of a specified disk by performing a specified action. Possible actions include the following:
fail—Fails a file system disk by removing it from the RAID configuration and spinning it down.
failnow—Immediately and ungracefully fails a file system disk, and marks the disk's RAID group as degraded.
unfail—Makes a failed disk a spare disk, or clears the prefail marker from a disk. If you specify a value of unfail, the interface prompts you for confirmation to force the disk to become a spare. If you respond yes to the prompt, the disk becomes a spare; if you respond no to the prompt, the prefail marker is cleared from the disk.
remove—Removes a spare disk from the RAID configuration and spins it down
The following example turns a failed disk named node1b:0c.29 into a spare disk.
node::>storage disk setstate -disk node1b:0c.29 -action unfail
The storage disk option show command displays the settings of the following disk options:
Background firmware update
Automatic copying of data to a spare in the event of a predictive failure
Automatic assignment of disks
The following example displays the disk-option settings for all nodes.
node::>storage disk option showNode Bkg. FW Upg. Auto Copy Auto Assign node0a on on on node0b on on on node1a on on on node1b on on on
You can use the storage disk option modify command to modify the following disk options:
Whether firmware updates run in the background
Whether data is automatically copied to a spare in the event of a predictive failure
Whether unowned disks are automatically assigned
By default, all of these options are set to on.
The following example disables background firmware upgrades for disks on all nodes.
node::>storage disk option modify -node * -bkg-firmware-update off
The storage disk replace command starts or stops the replacement of a file system disk with spare disk. When you start a replacement, Rapid RAID Recovery begins copying data from the specified file system disk to a spare disk. When the process is complete, the spare disk becomes the active file system disk and the file system disk becomes a spare disk. If you stop a replacement, the data copy is halted, and the file system disk and spare disk retain their initial roles. When you run the command, you can specify a spare disk or allow Data ONTAP 10.0 to select a spare disk to use.
The following example begins replacing a file system disk named node1a:0c.27. The Data ONTAP 10.0 system selects the spare disk that is used as the replacement disk.
node::>storage disk replace -disk node1a:0c.27 -action start
If a node completely fails and is replaced by another node (for instance, due to a motherboard failure), use the storage disk reassign command to reassign the failed node's disks to the new node. You can specify the nodes either by their names or by their serial numbers. If the failed node's disks have been taken over by the failed node's failover partner, specify the -force true parameter to force the reassignment operation. This command requires the advanced privilege level or higher.
In the following example, a node named node0a has failed. Its disks have not been taken over by its storage-failover partner. A replacement node named spare-node has been installed and connected to node0a's disk shelves. To assign node0a's disks to the new node, ensure that the new node is running and run the following commands:
node::>set -privilege advancednode::*>storage disk reassign -home node0a -newhome spare-node -force falsenode0a:1a.11, node0a:1a.12, node0a:1a.13, node0a:1a.14, node0a:1a.15, node0a:1a.16, node0a:1a.23 and node0a:1a.24 were reassigned to new owner spare-node.
You can remove ownership from a disk by using the storage disk removeowner command. After a disk's ownership is removed, another node can take ownership of it.
The following example removes ownership from a disk named node0a:0a.10.
node::>storage disk removeowner -disk node0a:0a.10
The storage disk zerospares command enables you to zero all spare disks owned by a node. Zeroing a disk means to overwrite all data on it with zeroes. This is useful if you are redeploying a set of disks from a development environment into a production environment.
The following example zeroes all spare disks owned by a node named node1b.
node::>storage disk zerospares -owner node1b
You can download disk and shelf firmware from a URL or URI to a specified node by using the storage firmware download. After the firmware is downloaded, you can update the disk firmware by using the storage disk updatefirmware command.
The following example downloads a firmware package with the path ftp://now.netapp.com/diskfw/fw-1.3.tgz to a node named node0a.
node::>storage firmware download -node node0a -package ftp://now.netapp.com/diskfw/fw-1.3.tgz
You can update the firmware on one or more disks by using the storage disk updatefirmware command. You can update the firmware on all local disks by running the command without any options, or you can update the firmware on one or more specific disks by running the command with the -disk parameter.list_of_disk_names
To download the latest firmware, use the storage firmware download command.
The following example updates the firmware on all disks.
node::>storage disk updatefirmware
If you need to locate one or more disks on a disk shelf, you can use the storage disk setled command to control disk LEDs. You can turn LEDs on or off or cause them to blink or stop blinking. When you run the command, you specify a disk name or name pattern, the state to which the LED or LEDs are to be set, and the time, in minutes, that the LED or LEDs are to remain in the specified state.
The following example causes the LEDs on all disks whose names match the pattern node1b* to blink for five minutes:
node::>storage disk setled -disk node1b* -action blink -time 5