movevdisk
Use the movevdisk command to move the preferred node of a volume either within the same caching I/O group or to another caching I/O group.
Syntax
Parameters
- -iogrp iogrp_id | iogrp_name
- (Optional) Specifies the I/O group to move the volume to.
- -nocachingiogrp
- (Optional) If you specify this parameter, the volume has no caching I/O group. This option should be used under the direction of IBM Support only. The -nocachingiogrp parameter is mutually exclusive with the -iogrp and -node parameters.
- -force
- (Optional) Use the force parameter to force the volume to be removed from an I/O group.
This option overrides the cache flush mechanism. Remember:
- If you specify the -force parameter, the contents of the cache are discarded and the volume might be corrupted by the loss of the cached data. Use the -force parameter with caution.
- If the force parameter is used to move a volume that has out-of-sync copies, a full resynchronization is required.
- -node node_id | node_name
- (Optional) Specifies the node ID or name that is assigned as the preferred node.
- vdisk_id | vdisk_name
- (Required) Specifies the volume to move.
Description
Use the movevdisk to migrate a single volume to a new I/O group - repeat this action for other volumes as required. This command can also move the preferred node of a volume without changing the caching I/O group, but it does not change which I/O groups can access the volume (only the caching I/O group is changed).
If you are moving volumes in preparation to remove the original I/O group completely, the FlashCopy® mappings must be deleted first. Otherwise, the source volumes go offline.
This command is not supported to change the I/O group if either copy is thin-provisioned or compressed and within a data reduction pool. The preferred node can be changed for volumes that are in data reduction pools.
A compressed volume can also be moved, and you can specify the preferred node in the new I/O
group. You can move a volume that is in a FlashCopy
mapping, but the FlashCopy bitmaps remain in the
original I/O group. You cannot move volumes when the FlashCopy mapping is in preparing
or prepared
state.
Additionally, a volume can be moved if it is the target of a FlashCopy mapping that is in stopping
state.
You cannot move a volume to change the caching I/O group for a volume that is in a Global Mirror, Metro Mirror, or HyperSwap® relationship, regardless of whether it is a primary, secondary, or change volume. To move a volume in a Global Mirror, Metro Mirror, or HyperSwap relationship, the relationship must first be deleted. You can change a preferred node without changing caching I/O group for this type of a volume.
If the volume is offline, use one of the recovervdisk commands to recover the volume and bring it back online. To specify a preferred node for the volume, use the -node node_id | node_name parameter with the movevdisk command. Use the movevdisk command to change the I/O group with which this volume is associated.
- A volume to an offline I/O group under any circumstance. Remember: To avoid data loss, make sure that the I/O group is online before you move the volume.
- An offline volume to the recovery I/O group.
You can migrate a volume to a new I/O group to manually balance the workload across the nodes in the clustered system. You might end up with a pair of nodes that are overworked and another pair of nodes that are underworked.
For systems that support multiple I/O groups, if the volume is a target of a FlashCopy mapping with a source volume in an
active-active
relationship, then the new I/O group must be in the same site as
the source volume. The system allows moving a volume in a remote copy relationship if the move
does not change the I/O group (it changes the preferred node). If the volume is in an
active-active
relationship, the new I/O group must be in the same site as the
source I/O group.
An invocation example to move DB_Volume
to I/O group
2
movevdisk -iogrp 2 DB_Volume
The resulting output:
No feedback
An invocation example to move DB_Volume
to I/O group
IOGRP3
with a new preferred node ID
7
movevdisk -iogrp IOGRP3 -node 7 DB_Volume
The resulting output:
No feedback
An invocation example to change preferred node of volume DB_Volume with a new preferred node ID as 8 in the same IOGRP
movevdisk -node 8 DB_Volume
The resulting output:
No feedback