Skip to main content

Setting up an active-passive configuration on nodes using root-data-data partitioning

When an HA pair is configured to use root-data-data partitioning by the factory, ownership of the data partitions is split between both nodes in the pair, for use in an active-active configuration. If you want to use the HA pair in an active-passive configuration, you must update partition ownership before creating your data aggregate.

Before you begin

  • You should have decided which node will be the active node and which node will be the passive node.
  • Storage failover must be configured on the HA pair.

About this task

This task is performed on two nodes: Node A and Node B.

All commands are input at the clustershell.

This procedure is designed for nodes for which no data aggregate has been created from the partitioned disks.

  1. View the current ownership of the data partitions: storage aggregate show-spare-disks -original-owner passive_node_name -fields local-usable-data1-size, local-usable-data2-size
    You should see that half of the data partitions are owned by one node and half are owned by the other node. All of the data partitions should be spare.
  2. Enter the advanced privilege level: set advanced
  3. For each data1 partition owned by the node that will be the passive node, assign it to the active node: storage disk assign -force -data1 -owner active_node_name -disk disk_name
    You do not need to include the partition as part of the disk name
  4. For each data2 partition owned by the node that will be the passive node, assign it to the active node: storage disk assign -force -data2 -owner active_node_name -disk disk_name
    You do not need to include the partition as part of the disk name
  5. Confirm that all of the partitions are assigned to the active node: storage aggregate show-spare-disks

    Example

    cluster1::*> storage aggregate show-spare-disks

    Original Owner: cluster1-01
    Pool0
    Partitioned Spares
    Local Local
    Data Root Physical
    Disk Type RPM Checksum Usable Usable Size
    --------------------------- ----- ------ -------------- -------- -------- --------
    1.0.0 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.1 SAS 7200 block 753.8GB 73.89GB 828.0GB
    1.0.2 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.3 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.4 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.5 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.6 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.7 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.8 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.9 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.10 SAS 7200 block 753.8GB 0B 828.0GB
    1.0.11 SAS 7200 block 753.8GB 0B 828.0GB

    Original Owner: cluster1-02
    Pool0
    Partitioned Spares
    Local Local
    Data Root Physical
    Disk Type RPM Checksum Usable Usable Size
    --------------------------- ----- ------ -------------- -------- -------- --------
    1.0.8 SAS 7200 block 0B 73.89GB 828.0GB
    13 entries were displayed.


    Note that cluster1-02 still owns a spare root partition.
  6. Return to administrative privilege: set admin
  7. Create your data aggregate, leaving at least one data partition as spare: storage aggregate create new_aggr_name -diskcount number_of_partitions -node active_node_name
    The data aggregate is created and is owned by the active node.
  8. Alternatively, you can use ONTAP’s recommend aggregate layout which includes best practices for RAID group layout and spare counts: storage aggregate auto-provision