Skip to main content

Healing the data aggregates

You must heal the data aggregates after repairing and replacing any hardware on the disaster site. This process resynchronizes the data aggregates and prepares the (now repaired) disaster site for normal operation. You must heal the data aggregates prior to healing the root aggregates.

About this task

The following example shows a forced switchover, where you bring the switched-over aggregate online. All configuration updates in the remote cluster successfully replicate to the local cluster. You power up the storage on the disaster site as part of this procedure, but you do not and must not power up the controller modules on the disaster site.

  1. Verify that switchover was completed by running the metrocluster operation show command.

    Example

    controller_A_1::> metrocluster operation show
      Operation: switchover
          State: successful
     Start Time: 7/25/2014 20:01:48
       End Time: 7/25/2014 20:02:14
         Errors: -

  2. Resynchronize the data aggregates by running the metrocluster heal -phase aggregates command from the surviving cluster.

    Example

    controller_A_1::> metrocluster heal -phase aggregates 
    [Job 130] Job succeeded: Heal Aggregates is successful.

    If the healing is vetoed, you have the option of reissuing the metrocluster heal command with the –override-vetoes parameter. If you use this optional parameter, the system overrides any soft vetoes that prevent the healing operation.

  3. Verify that the operation has been completed by running the metrocluster operation show command.

    Example

    controller_A_1::> metrocluster operation show
        Operation: heal-aggregates
          State: successful
    Start Time: 7/25/2014 18:45:55
       End Time: 7/25/2014 18:45:56
         Errors: -

  4. Check the state of the aggregates by running the storage aggregate show command.

    Example

    controller_A_1::> storage aggregate show
    Aggregate Size     Available Used% State   #Vols  Nodes        RAID Status
    --------- -------- --------- ----- ------- ------ ------------ ------------
    ...
    aggr_b2   227.1GB  227.1GB   0%    online  0      mcc1-a2      raid_dp, mirrored, normal...
  5. If storage has been replaced at the disaster site, you might need to remirror the aggregates.