Discovering MDisks using the CLI

You can use the command-line interface (CLI) to discover managed disks (MDisks).

The clustered system (system) automatically discovers the back-end controller and integrates the controller to determine the storage that is presented to the system nodes when back-end controllers are:
  • Added to the Fibre Channel
  • Included in the same switch zone as asystem
The Small Computer System Interface (SCSI) logical units (LUs) that are presented by the back-end controller are displayed as unmanaged MDisks. However, if the configuration of the back-end controller is modified after this has occurred, the system might be unaware of these configuration changes. You can request that the system rescan the Fibre Channel SAN to update the list of unmanaged MDisks.
Note: The automatic discovery completed by the system does not write anything to an unmanaged MDisk. You must instruct the system to add an MDisk to a storage pool or use an MDisk to create an image mode volume.

Discover (and then view) a list of MDisks:

  1. Issue the detectmdisk CLI command to manually scan the Fibre Channel network. The scan discovers any new MDisks that might have been added to the system and can help rebalance MDisk access across the available controller device ports.
    Notes:
    1. Only issue the detectmdisk command when you are sure that all of the disk controller ports are working and correctly configured in the controller and the SAN zoning. Failure to do this can result in errors that are not reported.
    2. Although it might appear that the detectmdisk command has completed, extra time might be required for it to run. The detectmdisk is asynchronous and returns a prompt while the command continues to run in the background. You can use the lsdiscoverystatus command to view the discovery status.
  2. When the detection is complete, issue the lsmdiskcandidate CLI command to show the unmanaged MDisks. These MDisks have not been assigned to a storage pool.
  3. Issue the lsmdisk CLI command to view all of the MDisks.

You have now seen that the back-end controllers and switches have been set up correctly and that the system recognizes the storage that is presented by the back-end controller.

This example describes a scenario where a single back-end controller is presenting eight SCSI LUs to the system:

  1. Issue detectmdisk.
  2. Issue lsmdiskcandidate.

    This output is displayed:

    id
    0
    1
    2
    3
    4
    5
    6
    7
    
  3. Issue lsmdisk -delim :

    This output is displayed:

    lsmdisk -delim :
    id:name:status:mode:mdisk_grp_id:mdisk_grp_name:capacity:ctrl_LUN_#:controller_name:UID:tier
    0:mdisk0:online:unmanaged:::68.4GB:0000000000000000:controller0:20000004cf2422aa000000000000000000000000000000000000000000000000:
    1:mdisk1:online:unmanaged:::68.4GB:0000000000000000:controller1:20000004cf1fd19d000000000000000000000000000000000000000000000000:
    2:mdisk2:online:unmanaged:::68.4GB:0000000000000000:controller2:20000004cf242531000000000000000000000000000000000000000000000000: