MetroCluster setup using one data aggregate per cluster

This procedure applies to Data ONTAP 8.3/8.3.0 only. To use this feature in Data ONTAP 8.3.1 and later version please refer to the documentation of the appropriate release. These versions allow the configuration of MetroCluster with a single mirrored data aggregate per cluster.

Note: To get support for this feature in Data ONTAP 8.3/8.3.0 please contact your account team and acquire a Product Variation Request (PVR).

The MetroCluster configuration requires at least two mirrored data aggregates to be created in each cluster being configured. These aggregates are used to create metadata volumes (one metadata volume per aggregate), when the metrocluster configure command is run.

If the administrator desires to configure MetroCluster with only one data aggregate per cluster, some additional steps need to be taken after the metrocluster configure command is run.

This article describes the steps required to move to a single aggregate configuration.

Note: Having both metadata volumes on one aggregate can result in failure of configuration replication across the two DR sites if the lone aggregate fails. For increased resiliency, it is recommended that the two metadata volumes be on separate aggregates.


Perform the following steps:

  1. Run the metrocluster configure command to complete the MetroCluster configuration.
    Note: To complete the initial configuration of the MetroCluster, two mirror data aggregates must exist per cluster.
  2. Run the volume show command to list the metadata volumes created on the local cluster aggregates.
    In the following output, it is seen that MDV_CRS_6cc9b517a26d11e48b6100505697f7e6_A is created on a1_required_data_aggr_1, and MDV_CRS_6cc9b517a26d11e48b6100505697f7e6_B is created on a2_required_data_aggr_1.
    Cluster_A::> volume show -fields volume,aggregate
    vserver         volume                                     aggregate
    --------------- ------------------------------------------ -----------------------
    Cluster_A       MDV_CRS_6cc9b517a26d11e48b6100505697f7e6_A a1_required_data_aggr_1
    Cluster_A       MDV_CRS_6cc9b517a26d11e48b6100505697f7e6_B a2_required_data_aggr_1
    Cluster_A       MDV_CRS_7ec06dc9a26d11e4b97100505697f7e9_A b1_required_data_aggr_1
    Cluster_A       MDV_CRS_7ec06dc9a26d11e4b97100505697f7e9_B b2_required_data_aggr_1

    (output truncated)
  3. Run the metrocluster check config-replication show command to determine which of these two metadata volumes is currently active (that is, being used for configuration replication).
    In the following output, the metadata volume MDV_CRS_6cc9b517a26d11e48b6100505697f7e6_A is in use.
    The other metadata volume can then be moved. In this case, volume MDV_CRS_6cc9b517a26d11e48b6100505697f7e6_B can be moved to the aggregate a1_required_data_aggr_1.
    Cluster_A::> metrocluster check config-replication show
    Enabled: true
    Running: true
    Remote Heartbeat: ok
    Last Heartbeat Sent: 2/13/2015 20:51:41
    Last Heartbeat Received: 2/13/2015 20:51:42
    Storage Status: ok
    Storage In Use: Cluster-wide Volume: MDV_CRS_6cc9b517a26d11e48b6100505697f7e6_A
    Storage Remarks: -
    Vserver Streams: ok
    Cluster Streams: ok
  4. Run the volume move command in advanced privilege to move the secondary metadata volume to the aggregate in which the primary metadata volume resides.
    Cluster_A::> set -privilege advanced
    Warning: These advanced commands are potentially dangerous; use them only when
    directed to do so by NetApp personnel.
    Do you want to continue? {y|n}: y
    Cluster_A::> volume move start -vserver Cluster_A -volume MDV_CRS_6cc9b517a26d11e48b6100505697f7e6_B -destination-aggregate a1_required_data_aggr_1

    Once the move of the volume is complete, all the metadata volumes will be hosted on one aggregate.
  5. Remove the aggregate, on which the volume was previously hosted. If desired, add the free disks to the one remaining data aggregate.
  6. Change the privilege mode back to admin:
    Cluster_A::> set -privilege admin
  7. Complete Steps 2 to 6 on the other site.
    Note: Due to the metadata volumes in the cluster residing on one aggregate, the EMS message cluster.vol.on.same.aggr will be generated on each of the clusters. This EMS message is a reminder that the best practice has not been followed. Once the metadata volumes have been moved, such that they are on separate aggregates, this EMS message will no longer be generated.


View all posts by