Planned outage of local (H1), moving production to site 2 (H2) and optional return on Three site MM / GM sessions

Edit online

In this scenario, your goal is to shut down the H1 box for maintenance, or for a planned power outage. As a result, you must move production to the H2 site, with the option of switching back to the H1 site after it is restored. Procedures are shown for both the multi-target Metro Mirror - Global Mirror session type as well as the Metro Global Mirror cascaded session type.

About this task

Follow these steps to carry out this scenario.

Multi-Target Metro Mirror - Global Mirror session type procedure

Edit online

Procedure

  1. Start H1>H2 H1>H3

    Issue this command to initiate Metro Global Mirror on your system while the production I/O is running on H1. The session reaches a Prepared state, creating a consistent point on Sites 2 and 3.

  2. Stop Host I/O
    A quiesce of the host I/O on H1 volumes needs to be performed before the next steps. Doing so ensures that all I/O is available at the remote site when the site switch is executed.
    Note: Since the outage is a planned outage, stop production I/O to H1.
  3. Suspend

    After the completion of the quiesce, this command will halt all replication to sites 2 and 3 by suspending the Metro Mirror relationships between H1 and H2. Upon completion of this step, the session state is in a suspended and recoverable condition.

  4. Recover H2

    This command makes H2 target-available. When the session state is Target Available, the production I/O can then be moved to write to the H2 volumes.

  5. Start H2>H1 H2>H3
    Issue this command to restart replication while production continues to run at Site 2. This command starts the Metro Mirror pairs between H2 and H1 and the Global Mirror pairs between H2 and H3. After the session reaches the Prepared state, both Site 1 and Site 3 will have a consistent copy of the data.
    Note: The following steps outline the process to switch production I/O back to writing on Site 1. Before you move to the next step, stop production I/O on H2.
  6. Stop Host I/O
    This step occurs outside of the session. Before the next steps, it is essential to perform a quiesce of the host I/O on H2 volumes. This action is necessary to guarantee the availability of all I/O at the remote site when the site switch is done.
    Note: Since the outage is a planned outage, stop production I/O to H2.
  7. Suspend

    This command suspends the Metro Mirror relationship between H2 and H1 and the Global Mirror relationships between H2 and H3. Once complete, the session state remains suspended and recoverable..

  8. Recover H1

    This command makes H1 target-available. When the session state is target-available at H1, run production I/O to H1.

  9. Start H1>H2 H1>H3

    This command returns the session to its original configuration, copying data from Site 1, to Site 2, to Site 3.

Procedure for Cascaded Metro Global Mirror session type

Edit online

Procedure

  1. Start H1>H2>H3

    Issue this command to begin Metro Global Mirror on your system when the production I/O is running on H1. The session reaches a Prepared state, ensuring the existence of a consistent point on Sites 2 and 3.

  2. Stop Host I/O
    This step occurs outside of the session. Before the next set of steps, performing a quiesce of the host I/O is necessary. Doing so guarantees that all I/O is available at the remote site when the site switch is done.
    Note: Since the outage is a planned outage, stop production I/O to H1.
  3. Suspend

    When the quiesce is complete, this command stops all replication to site 2 and site 3 by suspending the Metro Mirror relationships between H1 and H2. Once complete, the session state remains suspended and recoverable.

  4. Recover H2

    This command makes H2 target-available. When the session state is Target Available, the production I/O can then be moved to write to the H2 volumes.

  5. Start H2>H3

    This command starts a Global Mirror session during the outage of H1. Production I/O can continue on H2. After the session reaches the Prepared state, a consistent copy of the data resides on Site 3, and can be used in the event of a disaster.

  6. Start H2>H1>H3
    Issue this command when H1 is available and ready to be brought back into the configuration. This command starts a Metro Global Mirror session with H1 and H2 reversing roles compared to the original configuration. Production I/O can continue on H2. After the session reaches the Prepared state, both Site 1 and Site 3 will have a consistent copy of the data.
    Note: Follow the next steps after you decide to switch production I/O to write on Site 1 again. Stop production I/O on H2 before the next step.
  7. Stop Host I/O
    This step occurs outside of the session. Perform a quiesce of the host I/O on H2 volumes before the next set of steps. It is vital to make all I/O available at the remote site when the site switch is done.
    Note: Since the outage is a planned outage, stop production I/O to H2.
  8. Suspend

    This command suspends the Metro Mirror relationship between H2 and H1. Once complete, the session state remains suspended and recoverable.

  9. Recover H1

    This command makes H1 target-available. When the session state is target-available at H1, run production I/O to H1.

  10. Start H1>H2>H3

    This command returns the session to its original configuration, copying data from Site 1, to Site 2, to Site 3.