Skip to main content
A newer release of this product is available.

volume reallocation start

Contributors
Suggest changes
PDFs

Start reallocate job

Availability: This command is available to cluster administrators at the admin privilege level.

Description

Begins a reallocation scan on a LUN, NVMe namespace, file, or volume when you specify the path. If a volume has several small files that would benefit from periodic optimization, specify the /vol/volname.

Before performing a reallocation scan, the reallocation job normally performs a check of the current layout optimization. If the current layout optimization is less than the threshold, then the system does not perform a reallocation on the LUN, NVMe namespace, file, or volume.

You can define the reallocation scan job so that it runs at a specific interval, or you can use the volume reallocation schedule command to schedule reallocation jobs.

Note This command is not supported for FlexGroups or FlexGroup constituents.

Parameters

-vserver <vserver name> - Vserver

Specifies the Vserver.

-path <text> - Path

Specifies the path of the reallocation for a LUN, NVMe namespace, file, or volume.

{ [-i, -interval <text>] - Interval Schedule

Specifies the reallocation scan interval in

  • m for minutes

  • h for hours

  • d for days

For example, 30m is a 30 minute interval. The countdown to the next scan begins after the first scan is complete.

The default interval is 24 hours.

| [-o, -once <true>] - Once

Specifies that the job runs once and then is automatically removed from the system when set to true. If you use this command without specifying this parameter, its effective value is false and the reallocation scan runs as scheduled. If you enter this parameter without a value, it is set to true and a reallocation scan runs once.

| [-f, -force <true>] - Force }

Performs a one-time full reallocation on a LUN, file, or volume when set to true. A forced reallocation rewrites blocks on a LUN, file, or volume unless the reallocation would result in worse performance. If you use this command without specifying this parameter, its effective value is false and a forced reallocation is not performed. If you specify this parameter without a value, it is set to true, and a forced reallocation is performed.

{ [-p, -space-optimized <true>] - Space Optimized

Specifies that snapshot blocks are not copied to save space when set to true. If you use this command without specifying this parameter, its effective value is false and snapshot blocks are copied. However, reads from snapshots might have a slightly higher latency. If you specify this parameter without a value, it is set to true and snapshot blocks are not copied. You cannot use the space-optimized option with the unshare option.

| [-u, -unshare <true>] - Unshare Deduplicated Blocks }

Specifies that blocks that are shared by deduplication will be unshared. This option can help remove fragmentation caused on dense volumes. This may result in increased disk usage, especially for full reallocation. You cannot use the unshare option with the space-optimized option.

{ [-t, -threshold <integer>] - Threshold

Specifies the threshold when a LUN, NVMe namespace, file, or volume is considered unoptimized and a reallocation should be performed. Once the threshold is reached, the system creates a diagnostic message that indicates that a reallocation might improve performance.

The threshold range is from 3 (the layout is moderately optimized) to 10 (the layout is not optimal). The threshold default is 4.

| [-n, -no-check <true>] - No Threshold Check }

Does not check the current layout to determine if a reallocation is needed when set to true. If you use this command without specifying this parameter, its effective value is false and the system does check the current layout to determine if a reallocation is needed. If you specify this parameter without a value, it is set to true and the system does not check the current layout to determine if a reallocation is needed.

Examples

cluster1::> volume reallocation start  -path /vol/vol2 -interval 30m
[Job 165] Job is queued: Reallocate Job.

Starts a reallocation job on volume vol2 every 30 minutes.