Relocate non-root aggregates from node1 to node2

Contributors Download PDF of this page

Before you can replace node1 with node3, you need to move the non-root aggregates from node1 to node2 by using the storage aggregate relocation command and then verifying the relocation.

Steps
  1. Relocate the non-root aggregates by completing the following substeps:

    1. Set the privilege level to advanced:

      set -privilege advanced

    2. Enter the following command:

      storage aggregate relocation start -node <node1> -destination <node2> -aggregate-list * -ndo-controller-upgrade true

    3. When prompted, enter y.

      Relocation will occur in the background. It could take anywhere from a few seconds to a couple of minutes to relocate an aggregate. The time includes both client outage and nonoutage portions. The command does not relocate any offline or restricted aggregates.

    4. Return to the admin level by entering the following command:

      set -privilege admin

  2. Check the relocation status by entering the following command on node1:

    storage aggregate relocation show -node <node1>

    The output will display Done for an aggregate after it has been relocated.

    Note Wait until all non-root aggregates owned by node1 have been relocated to node2 before proceeding to the next step.
  3. Take one of the following actions:

    If relocation…​ Then..

    Of all aggregates is successful

    Go to Step 4.

    Of any aggregates fails or is
    vetoed

    1. Check the EMS logs for the corrective action.

    2. Perform the corrective action.

    3. Relocate any failed or vetoed aggregates:
      storage aggregate relocation start -node <node1> - destination <node2> -aggregate-list * -ndo-controller-upgrade true

    4. When prompted, enter y.

    5. Return to the admin level:
      set -privilege admin
      If necessary, you can force the relocation using one of the following methods:

      • Override veto checks:
        storage aggregate relocation start -override-vetoes true -ndo-controller-upgrade

      • Override destination checks:
        storage aggregate relocation start -override-destination-checks true -ndo-controller-upgrade

    Refer to References to link to the Disks and Aggregates Power Guide and the ONTAP 9 Commands: Manual Page Reference for more information about storage aggregate relocation commands.

  4. Verify that all the non-root aggregates are online and their state on node2:

    storage aggregate show -node <node2> -state online -root false

    The following example shows that the non-root aggregates on node2 are online:

    cluster::> storage aggregate show -node node2 state online -root false
    Aggregate     Size Available Used% State   #Vols  Nodes            RAID Status
    --------- -------- --------- ----- ------- ------ ---------------- ------------
    aggr_1
               744.9GB 744.8GB      0% online       5 node2            raid_dp,
                                                                       normal
    aggr_2     825.0GB 825.0GB      0% online       1 node2            raid_dp,
                                                                       normal
    2 entries were displayed.

    If the aggregates have gone offline or become foreign on node2, bring them online by using the following command on node2, once for each aggregate:

    storage aggregate online -aggregate <aggr_name>

  5. Verify that all the volumes are online on node2 by entering the following command on node2 and examining its output:

    volume show -node <node2> -state offline

    If any volumes are offline on node2, bring them online by using the following command on node2, once for each volume:

    volume online -vserver <vserver-name> -volume <volume-name>

    The vserver-name to use with this command is found in the output of the previous volume show command.

  6. Enter the following command on node2:

    storage failover show -node <node2>

    The output should display the following message:

    Node owns partner's aggregates as part of the nondisruptive controller upgrade procedure.
  7. Verify that node1 does not own any non-root aggregates that are online:

    storage aggregate show -owner-name <node1> -ha-policy sfo -state online

    The output should not display any online non-root aggregates, which have already been relocated to node2.