Detaching the disk from the z/VM guest

Before you create the DD image, safely detach the FCP disk from the z/VM guest and attach it to the management server.

Before you begin

Ensure that you have:

  • Root access to the z/VM guest
  • Sufficient z/VM privileges to detach and attach devices
  • Access to a z/VM host using a x3270 terminal or vmcp commands

About this task

This phase prepares the z/VM guest disk for migration by safely shutting down the guest and transferring disk ownership to the management server.

Procedure

  1. Shut down the z/VM guest to ensure data integrity by running the following command:
    shutdown -h now
    Important: Wait for the guest to fully shut down before proceeding.
  2. Detach the FCP disk from the source guest.

    Use one of the following methods:

    • On the z/VM host using a x3270 terminal, run the following command:
      cp detach device_address from guest_name
    • On the management z/VM guest, run the following command:
      vmcp detach device_address from guest_name

    Parameters:

    • device_address: The FCP device address. For example: 1000.
    • guest_name: The z/VM guest name. For example: GUEST001.

  3. Attach the FCP disk to your z/VM management server

    Use one of the following methods:

    • On the z/VM host using a x3270 terminal, run the following command:
      cp attach device_address to management_server_guest_name
    • On the management z/VM guest, run the following command:
      vmcp attach device_address to management_server_guest_name

    Parameters:

    • device_address: The FCP device address. For example: 1000.
    • management_server_guest_name: The z/VM management server guest name. For example: GUEST002.

  4. Make the attached FCP disk available to the operating system on the management server.

    Set the FCP device online for each path:

    chzdev -e device_ID

    Example for multiple paths:

    chzdev -e 0.0.800b:0x500507630400d1e3:0x4001409100000000
    chzdev -e 0.0.800b:0x500507630408d1e3:0x4001409100000000
    chzdev -e 0.0.810b:0x500507630408d1e3:0x4001409100000000
    chzdev -e 0.0.810b:0x500507630400d1e3:0x4001409100000000

    Parameters:

    • device_ID: The FCP device address. For example: 0.0.800b:0x500507630400d1e3:0x4001409100000000.

    Note: For more information about chzdev and lszdev, see Device configuration.
  5. Verify that the device is online by running the following command:
    lszdev

    Example output:

    TYPE         ID                                              ON   PERS  NAMES
    zfcp-host    0.0.800a                                        yes  yes
    zfcp-host    0.0.800b                                        yes  yes
    zfcp-host    0.0.810a                                        yes  yes
    zfcp-host    0.0.810b                                        yes  yes
    zfcp-lun     0.0.800a:0x500507630400d1e3:0x4001409000000000  yes  yes   sdd sg3
    zfcp-lun     0.0.800a:0x500507630408d1e3:0x4001409000000000  yes  yes   sdc sg2
    zfcp-lun     0.0.800b:0x500507630400d1e3:0x4001409100000000  yes  yes   sdf sg5
    zfcp-lun     0.0.800b:0x500507630408d1e3:0x4001409100000000  yes  yes   sde sg4
    zfcp-lun     0.0.810a:0x500507630400d1e3:0x4001409000000000  yes  yes   sda sg0
    zfcp-lun     0.0.810a:0x500507630408d1e3:0x4001409000000000  yes  yes   sdb sg1
    zfcp-lun     0.0.810b:0x500507630400d1e3:0x4001409100000000  yes  yes   sdh sg7
    zfcp-lun     0.0.810b:0x500507630408d1e3:0x4001409100000000  yes  yes   sdg sg6
    qeth         0.0.bdf0:0.0.bdf1:0.0.bdf2                      yes  yes   encbdf0
    generic-ccw  0.0.0009                                        yes  no
  6. Verify that the multipath device is available by running the following command:
    multipath -ll

    Example output:

    mpatha (36005076304ffd1e30000000000000190) dm-0 IBM,2107900
    size=500G features='1 queue_if_no_path' hwhandler='1 alua' wp=rw
    `-+- policy='service-time 0' prio=50 status=active
      |- 1:0:2:1083195393 sda 8:0  active ready running
      |- 1:0:0:1083195393 sdb 8:16 active ready running
      |- 0:0:0:1083195393 sdc 8:32 active ready running
      `- 0:0:2:1083195393 sdd 8:48 active ready running
    mpathb (36005076304ffd1e30000000000000191) dm-2 IBM,2107900
    size=50G features='1 queue_if_no_path' hwhandler='1 alua' wp=rw
    `-+- policy='service-time 0' prio=50 status=active
      |- 2:0:0:1083260929 sde 8:64  active ready running
      |- 2:0:2:1083260929 sdf 8:80  active ready running
      |- 3:0:1:1083260929 sdh 8:112 active ready running
      `- 3:0:3:1083260929 sdg 8:96  active ready running

    The FCP disk should now appear as a multipath device (for example, /dev/mapper/mpathb) on the management server.