migrateexts

The migrateexts command migrates extents from one managed disk to another.

Syntax


>>- svctask -- -- migrateexts -- ------------------------------->

>-- -source --+- source_mdisk_id ---+-- ------------------------>
              '- source_mdisk_name -'      

>-- -target --+- target_mdisk_id ---+-- ------------------------>
              '- target_mdisk_name -'      

>-- -exts -- number_of_extents -- ------------------------------>

>--+---------------------------------+--+--------------+-- ----->
   '- -threads -- number_of_threads -'  '- -copy-- id -'      

>-- -vdisk --+- vdisk_id ---+----------------------------------><
             '- vdisk_name -'

Parameters

-source source_mdisk_id | source_mdisk_name: (Required) Specifies the MDisk on which the extents currently reside.
-target target_mdisk_id | target_mdisk_name: (Required) Specifies the MDisk to migrate the extents to.
-exts number_of_extents: (Required) Specifies the number of extents to migrate.
-threads number_of_threads: (Optional) Specifies the number of threads to use while migrating these extents. You can specify 1 - 4 threads. The default number of threads is 4.
-copy id: (Required if the specified VDisk has more than one copy) Specifies the VDisk copy that the extents belong to.
-vdisk vdisk_id | vdisk_name: (Required) Specifies the VDisk that the extents belong to.

Description

This command migrates a given number of extents from the source virtual disk and the managed disk that contains extents that are used to make up the virtual disk. The target is a managed disk within the same managed disk group.

If a large number of extents are being migrated, you can specify 1 - 4 threads. You can issue the svcinfo lsmigrate command to check the progress of the migration.

The migrateexts command fails if there are insufficient free extents on the target managed disk. To avoid this problem, do not issue new commands that use extents until the extents migration is completed.

The migrateexts command fails if the target or source VDisk is offline. Correct the offline condition before attempting to migrate the VDisk.

Note: Migration activity on a single managed disk is limited to a maximum of 4 concurrent operations. This limit does not take into account whether the managed disk is the source or the destination target. If more than four migrations are scheduled for a particular managed disk, further migration operations are queued pending the completion of one of the currently running migrations. If a migration operation is stopped for any reason, a queued migration task can be started. However, if a migration is suspended, the current migration continues to use resources and a pending migration is not started. For example, the following setup is a possible initial configuration:

MDiskGrp 1 has VDisk 1 created in it
MDiskGrp 2 has VDisk 2 created in it
MDiskGrp 3 has only one MDisk

With the previous configuration, the following migration operations are started:

Migration 1 migrates VDisk 1 from MDiskGrp 1 to MDiskGrp 3, running with 4 threads.
Migration 2 migrates VDisk 2 from MDiskGrp 2 to MDiskGrp 3, running with 4 threads.

Due to the previous limitations, the two migration operations do not always run at the same speed. MDiskGrp 3 has only one MDisk and the two migration operations have a total of 8 threads that are trying to access the one MDisk. Four threads are active. The remaining threads are in standby mode waiting to access the MDisk.

An invocation example

svctask migrateexts -vdisk vdisk4 -source mdisk4 -exts
	64 -target mdisk6 -threads 4

The resulting output

No feedback