4462 vCenter partition linking and async-DR policy for replicated volume groups is required.

Explanation

When two storage system is running 8.7.1 or an older software version with vVol replication enabled (using the two-site async-DR policy) are upgraded through CCU to the 910 software version, a vCenter partition is created on each system. The replicated vVol volume groups are then moved into the respective vCenter partitions..

After the upgrade each system has a vCenter partition and it includes replicated volume groups using 2-site async-DR policy. The PBR configuration needs to be updated to link the vCenter partition on the production site to the vCenter partition on the recovery site. The replicated volume groups need to switch from 2-site async-DR policy to async-DR policy that is created using the linked vCenter partitions.

User response

The replication process involves two systems. The event associated with this error code is triggered on the system that hosts the vCenter Partition, referred to here as the local partition.

Steps to retrieve the remote system ID and remote partition UUID:

  1. Identify replicated volume groups (VGs) using a 2-site async disaster recovery (DR) replication policy within the vCenter partition associated with the event.
    1. Retrieve the list of volume groups (VGs) that are associated with the same partition as the vCenter partition where the event was triggered, and for which the dr_replicated attribute is set to true. This information can be obtained by using the partition_id and dr_replicated filters with the lsvolumegroup command.
    2. For each volume group (VG) identified in the previous step, retrieve its associated replication policy using the replication_policy_id field from the lsvolumegroup command. Then, verify whether the topology of the replication policy is set to 2-site async-dr. This can be confirmed by checking the topology column in the corresponding lsreplicationpolicy entry. If the topology matches, include the VG and its replication policy in the shortlist.
    3. At this stage, you have identified a list of replicated volume groups as volume_groups_to_process that are associated with the same partition where the event was triggered and are using a 2-site async-dr replication policy. Additionally, we have gathered the corresponding replication policy details, referred as replication_policies_to_be_used.
    4. Retrieve the remote system details using one of the shortlisted replication policies. These details is an input for the -remotesystem parameter in the chpartition -makedrlink command, which is used to establish a link between partitions.
  2. Use the lsreplicationpolicy command to view the rpo_alert value for the relevant replication policy.
  3. Get details of the remote partition UUID.
    1. Get the volume group entry from the remote system whose name matches with name of the one of the volume group found in step 1.c
    2. The remote partition ID can be obtained from the volume group entry. Use this ID to retrieve the UUID of the remote partition from the remote system. This UUID will be used as the input for the -remotelinkedpartitionuuid parameter in the chpartition -makedrlink command.
  4. After retrieving all required inputs in the previous steps such as the remote system ID or name, the UUID of the remote partition for linking, and the rpo_alert value for replication policy creation, the user can proceed with the subsequent configuration steps.
    Note: You can now perform the partition linking step 5 on only one of the systems. However, the replication policy creation on step 6 and the update of the volume group to use the new replication policy on both systems involved in the replication where the event was triggered
  5. You can issue partition linking CLI on the primary system using by specifying the UUID of remote system's vCenter Partition.
    chpartition -makedrlink -remotedrlinkedpartitionuuid <UUID-of-remote-vCenter-Partition> -remotesystem <remote-system-id/name> <local-vCenter-Partition-name/id-against-which-event-is-raised>
  6. For each short listed VG or for each volume group volume_groups_to_process list follow below steps:
    1. Check the rpo_alert value of the 2-site async dr policy currently assigned to this volume group and create a async-dr policy if not created already.
      mkreplicationpolicy -topology async-dr -rpoalert <rpo_alert_value> -partition <local-vCenter-Partition-name/id-against-which-event-is-raised>
    2. Change the replication policy of the volume group from two-site async dr to newly created async-dr policy.
      chvolumegroup -replicationpolicy <async-dr-policy-name/id> <vCenter-Partition's-replicated-vg-name/id>
  7. After all targeted volume groups have been updated from the 2-site-async-dr policy to the async-dr policy, mark the event as resolved.