Skip to main content

Migrate data aggregates, epsilon, and LIFs on FAS2820 node1 to node4

Contributors netapp-pcarriga
PDFs

Migrate the data aggregates, epsilon, and logical interfaces (LIFs) on the FAS2820 node1 to node4.

About this task

Perform the following steps on node4.

Steps

  1. At the LOADER prompt for node4, set partner-system ID:

    setenv partner-sysid <system_ID_of_node1>

  2. Check the partner system-ID:

    printenv partner-sysid

  3. Save the changes:

    saveenv

  4. Boot the node into the boot menu:

    boot_ontap menu

  5. At the boot menu, select option 6 Update flash from backup config to restore the /var file system to node4.

    This replaces all flash-based configuration with the last backup to disks.

  6. Enter y to continue.

    Note

    The node automatically reboots to load the new copy of the /var file system.

    The node reports a system ID mismatch warning. Enter y to override the system ID.

  7. Migrate the cluster LIFs:

    set -privilege advanced
    network port show
    Note

    If the system cluster ports are not similar when upgrading a FAS2820 to a replacement controller, you might have to temporarily change the interfaces on node4 into cluster ports:

    network port modify -node <node4> -port <port_name> -mtu 9000 -ipspace Cluster
    		 
    network interface migrate -vserver Cluster -lif <cluster_LIF>  -destination-node <node4> -destination-port <port_name>

  8. Wait for the cluster to reach quorum, then verify that the cluster nodes are healthy:

    – cluster show
    Note The HA pair and storage failover remain disabled in the current state.

  9. Move the cluster LIFs to the temporary 25G cluster ports on node4:

    network interface modify -vserver Cluster -lif <cluster_LIF> -home-node <node4> -home-port <port_name>

  10. Complete this step only if interface groups and data VLANs are used on the FAS2820 cluster you are upgrading. Otherwise, go to Step 11.

    The physical network port names on the replacement controller differ from those on the FAS2820. This might lead to displaced VLANs and incorrectly configured interface groups on node4.

    1. Show the displaced VLANs:

      cluster controller-replacement network displaced-vlans show

    2. Restore the displaced VLANs:

      cluster controller-replacement network displaced-vlans restore

    3. Fix the incorrectly configured interface groups. The port names between the FAS2820 and the controllers you are upgrading might be different. Update the interface groups with the correct member ports:

      ifgrp remove-port -node <node2> -ifgrp <ifgrp_name> -port <port_name>
      ifgrp add-port -node <node2> -ifgrp <ifgrp_name> -port <port_name>

  1. Migrate the data aggregates on node1 to node4:

    storage aggregate relocation start -aggregate-list <aggregate_list_name> -node <node1> -destination <node4> -ndo-controller-upgrade true -override-destination-checks true

  2. Show data aggregate status:

    storage aggregate show

  3. Migrate the epsilon by removing it from node1 and moving it to node4.

    1. Remove epsilon from node1:

      cluster modify -epsilon false -node <node1>

    2. Move epsilon to node4:

      cluster modify -epsilon true -node <node4>

  4. Show cluster status to verify the epsilon change:

    cluster show

  5. Display all data network LIFs:

    network interface show -role data

  6. Migrate all data LIFs to node4:

    network interface migrate -vserver <vserver_name> -lif <lif_name> -destination-node <node4> -destination-port <port_name>

  7. Display the status of all data LIFs in the cluster:

    network interface show -role data

  8. If any LIFs are down, set their administrative status to up by entering the following command once for each LIF:

    network interface modify -vserver <vserver_name> -lif <lif_name> -status-admin up

  9. Migrate the cluster management LIF:

    network interface migrate -vserver <vserver_name> -lif <cluster_mgmt_lif> -destination-node <node4> -destination-port <port_name>

  10. Display the status of the cluster management LIF:

    network interface show -role cluster-mgmt

  11. Halt node1:

    halt -node <node1> -inhibit-takeover true -ignore-quorum-warnings true
What's next?

Convert node1 to a drive shelf and connect to node3