Creating volumes and qtree for LUNs

Guidelines for creating volumes

When you create volumes to contain LUNs, use the following guidelines:

Create volumes at least twice the size of the LUNs when snapshots of LUNs are not required.
Create volumes at least three times the size of the LUNs when snapshots of LUNs are required.
Create volumes for the exclusive use of LUNs:
- Ensure that no other files or directories exist in volumes that contain LUNs.
- Ensure that volumes that are accessed by NAS protocols do not contain LUNs.
Enable space reservation setting for the volume (recommended).
Enable the create_ucode options for the volume.
If multiple hosts share the same volume, create a qtree on the volume to store all LUNs of the same type (recommended).

Creating a volume

To create a volume, complete the following step.

Note
If you want to create volumes using a Web-based interface, launch FilerView and go to Volumes > Manage > Add New Volume. See Launching FilerView for details. With FilerView, you can see what choices you have, for example, which disks are available, if you want to manually select them.

Step	Action
1	On the filer's command line, enter the following command to create a volume: vol create volname [-f][-n][-m][-r raid-group-size] [-l language_code] disk-list volname is the name of the new volume (include the /vol/ prefix). `-f` instructs Data ONTAP to ignore fault isolation rules; for example, allow the volume to use disks on each plex from different spare pools or, with sync_mirror_local licensed, allow an unmirrored volume to use disks for that volume to span spare pools. `-n` displays the command that Data ONTAP executes without making any changes; this is useful if you want to see the automatically selected disks before executing the command. `-m` specifies that the new volume is mirrored (has two plexes) upon creation. If you choose this parameter, the indicated disks are split across the two plexes. By default, new volumes are not mirrored. raid-group-size is the RAID group size (in number of disks) you want to use for this volume. Network Appliance recommends using eight disks. language_code specifies a language other than that of the root volume (See the Data ONTAP System Administrator's Guide for language codes). disk-list is one of the following: ndisks [@disk-size] ndisks is the number of disks to use. It must be at least 2. disk-size is the disk size to use, in gigabytes. You must have at least two available disks of the size you specify. -d disk_name1 disk_name2... disk_nameN disk_name1, disk_name2... disk_nameN are disk IDs of one or more available disks; separate multiple disks with a space. Example `vol create vol1 8` Result A volume called /vol/vol1 is created with 8 disks. Data ONTAP selects the disks from the pool of available disks.

Step

Action

On the filer's command line, enter the following command to create a volume:

vol create volname [-f][-n][-m][-r raid-group-size] 

[-l language_code] disk-list

volname is the name of the new volume (include the /vol/ prefix).

-f instructs Data ONTAP to ignore fault isolation rules; for example, allow the volume to use disks on each plex from different spare pools or, with sync_mirror_local licensed, allow an unmirrored volume to use disks for that volume to span spare pools.

-n displays the command that Data ONTAP executes without making any changes; this is useful if you want to see the automatically selected disks before executing the command.

-m specifies that the new volume is mirrored (has two plexes) upon creation. If you choose this parameter, the indicated disks are split across the two plexes. By default, new volumes are not mirrored.

raid-group-size is the RAID group size (in number of disks) you want to use for this volume. Network Appliance recommends using eight disks.

language_code specifies a language other than that of the root volume (See the Data ONTAP System Administrator's Guide for language codes).

disk-list is one of the following:

ndisks [@disk-size]
ndisks is the number of disks to use. It must be at least 2.

disk-size is the disk size to use, in gigabytes. You must have at least two available disks of the size you specify.
-d disk_name1 disk_name2... disk_nameN
disk_name1, disk_name2... disk_nameN are disk IDs of one or more available disks; separate multiple disks with a space.

Example

vol create vol1 8

Result

A volume called /vol/vol1 is created with 8 disks. Data ONTAP selects the disks from the pool of available disks.

Enabling space reservation

To enable space reservation on a volume, complete the following step.

Note
If you want to enable space reservation using FilerView, launch FilerView, go to Volumes > Manage, look for the name of the volume, then click the Modify icon in the Operations section. Find the field for Space Reservations and select On.

Step	Action
1	On the filer's command line, enter the following command: *vol options volname* create_reserved on** Result This guarantees that space is reserved to write data to the holes that previously did not contain data in the LUN and that space is reserved for block overwrites for files that are created on volname after this command is executed.

Step

Action

On the filer's command line, enter the following command:

vol options volname create_reserved on

Result

This guarantees that space is reserved to write data to the holes that previously did not contain data in the LUN and that space is reserved for block overwrites for files that are created on volname after this command is executed.

Enabling create_ucode

Volumes, file, and LUN names must be written in ucode to support multiple protocol access to LUNs. To force Unicode format directories to be created by default, from both NFS and CIFS, complete the following step.

Step	Action
1	On the filer's command line, enter the following command to enable the vol option to create Unicode format directories by default: *vol options volname* create_ucode on** Example `vol options vol1 create_ucode on`

Organizing LUNs by creating qtrees

You can use qtrees to organize related LUNs into groups. To create a qtree on the filer to host multiple LUNs, complete the following step.

Note
If you want to create a qtree in a volume using FilerView, go to Volumes > Manage > Qtrees > Manage and click the Create button.

Step	Action
1	On the filer's command line, enter the following command to create a qtree: qtree create /vol/volname/qtree-name Note Create the qtree in any volume other than the root volume (/vol/vol0). Include /vol in the path because if it does not begin with a slash (/), the qtree is created in the root volume. Example `qtree create /vol/vol1/qtree1` Result qtree1 is a qtree in /vol/vol1.

Step

Action

On the filer's command line, enter the following command to create a qtree:

qtree create /vol/volname/qtree-name

Note
Create the qtree in any volume other than the root volume (/vol/vol0). Include /vol in the path because if it does not begin with a slash (/), the qtree is created in the root volume.

Example

qtree create /vol/vol1/qtree1

Result

qtree1 is a qtree in /vol/vol1.

Verifying that options are set properly

To verify that space reservation is on, complete the following step.

Step	Action
1	Display the status of several options on the volume by entering the following command: `vol status [volname] -v` Example `vol status /vol1 -v` Result Volume State Status Options vol1 online normal nosnap=on, nosnapdir=off, minra=off, no_atime_update=off, raidsize=14, nvfail=off, snapmirrored=off, resyncsnaptime=60, create_ucode=off convert_ucode=off, maxdirsize=10240, fs_size_fixed=off, create_reserved=on raid_type=RAID4 Plex /vol/vol1/plex0: online, normal, active RAID group /vol/vol1/plex0/rg0: normal Note If you do not specify a volume, status on all volumes is displayed.

Step

Action

Display the status of several options on the volume by entering the following command:

vol status [volname] -v

Example

vol status /vol1 -v

Result

Volume  State   Status   Options

 vol1   online  normal   nosnap=on, nosnapdir=off,

                         minra=off, no_atime_update=off,

                         raidsize=14, nvfail=off,

                         snapmirrored=off,

                         resyncsnaptime=60,

                         create_ucode=off

                         convert_ucode=off,

                         maxdirsize=10240,

                         fs_size_fixed=off,

                         create_reserved=on

                         raid_type=RAID4

 

       Plex /vol/vol1/plex0: online, normal, active

          RAID group /vol/vol1/plex0/rg0: normal

Note
If you do not specify a volume, status on all volumes is displayed.