Planned Failback from the Remote DR Center to the Production Center (Parallel Network)

Prerequisites

• eReplication in the remote DR center has been used to switch services from the production center to the remote DR center as planned.
• On a synchronous and asynchronous parallel network, the disk array replication relationship has been properly set: Data is synchronously replicated from the production center to the same-city DR center, and is asynchronously replicated from the production center to the remote DR center (split).
• On an asynchronous and asynchronous parallel network, the disk array replication relationship has been properly set: Data is asynchronously replicated from the production center to the same-city DR center, and is asynchronously replicated from the production center to the remote DR center (split).
• In the Oracle application, the database is disabled but the cluster is enabled by default on the eReplication Agent. If you need to disable clusters, modify the configuration of disabling clusters. For details, see Changing the Configuration of Stopping the Oracle Cluster.
• In a DR Star network, before performing planned migration, log in to DeviceManager and ensure that the running status of the DR Star network is Enabled.
• The datastore name cannot contain Chinese characters.

Context

• To perform a failback in an asynchronous and asynchronous network, perform the following steps on eReplication in the remote DR center:
  1. Perform reprotection to protect services migrated to the remote DR center again.
  2. Perform a test and clearing and verify data availability to ensure the success rate of service failback.
  3. Perform planned migration to switch the services back to the production center from the remote DR center.
  4. Perform reprotection to protect services switched back to the production center again.
• To perform a failback on a synchronous and asynchronous network, perform the following steps:
  1. Synchronize production data from the DR center to the production center.
  2. Restore asynchronous remote replication between storage arrays from the production center to the remote DR center.
  3. Synchronize data from the production center to the same-city DR center.
  4. Restore synchronous remote replication between storage arrays from the production center to the same-city DR center.
  5. Test production services in the production center.
  6. Restore DR configurations on the DR management server in the same-city DR center.

Procedure

• The following methods show how to switch production services back from the remote DR center to the production center on an asynchronous and asynchronous parallel network:
  1. On eReplication in the remote DR center, perform reprotection to protect services migrated to the remote DR center again.

     After you complete the planned migration, the application system is working in the remote DR center and protected groups become Invalid. Before service failback, reprotection needs to be implemented to reversely protect production services running on the remote DR center. Based on specified policies, service data generated in the remote DR center is automatically replicated to the original production center. For details, see 5 to 6 in Planned Migration from the Production Center to the Same-City DR Center.

  2. On eReplication in the remote DR center, perform a test and clearing and verify data availability to ensure the success rate of service failback.

     Before the services are switched back, you need to perform a DR test to verify the data availability to ensure success rate of service failback. After the test, clear test data to avoid failback failures caused by the test data. For details, see DR Testing in the DR Center.

  3. On eReplication in the remote DR center, perform planned migration to switch the services back to the production center from the remote DR center.

     This step aims to switch services back to the production center. After migration, data must be checked and test data must be cleared. For details, see 1 to 4 in Planned Migration from the Production Center to the Same-City DR Center.

     

     If the protected object type is LUN, Local File System, Oracle, IBM DB2, Microsoft SQL Server, SAP HANA, or Microsoft Exchange Server, select hosts or host groups in the production center as the recovery target.

     • If the protected object type is VMware VM, select vCenter and clusters in the production center as the recovery target.
     • If the protected object type is FusionCompute VM, select site in the production center as the recovery target.
  4. On eReplication in the remote DR center, perform reprotection to protect services switched back to the production center again.

     After failback, the application system is running on the production center and protected groups become Invalid. In this case, reprotection needs to be implemented to protect services in the production center. Based on specified policies, service data generated in the production center is automatically replicated to the remote and same-city DR centers. For details, see 5 to 6 in Planned Migration from the Production Center to the Same-City DR Center.

• The following content shows how to switch production services back from the remote DR center to the production center on a synchronous and asynchronous parallel network.
  1. On the production storage, split the synchronous remote replication from the production center to the same-city DR center and perform primary/secondary switchover for the replication.
     1. Log in to the management software of the storage array in the production center.
     2. Split the consistency groups of synchronous remote replication.
     3. Perform primary/secondary switchover for the consistency groups of the asynchronous remote replication.

        This substep aims to prepare for data synchronization from the remote DR center to the production center. Before the primary/secondary switchover, remote replication relationships between storage arrays in the production, same-city DR, and remote DR centers are shown in Figure 1:

        Figure 1 Original remote replication relationships
        
        • Synchronous remote replication is performed between storage arrays in the production center and the same-city DR center. The production storage array is the source (primary) array of the synchronous replication.
        • Asynchronous remote replication is performed between the production center and the remote DR center, and the asynchronous replication is in the Split state. The production storage array is the source (primary) array of the replication.
          

          Remote replication in the Split state will not synchronize data.

          Remote replication relationships must be set in the Split state before the primary/secondary switchover.

          In a remote replication relationship, the source array is the primary end and the target array is the secondary end.

          After the synchronous remote replication consistency group is split and primary/secondary switchover is performed for the asynchronous remote replication consistency group in the production center, remote replication relationships between storage arrays in the production, same-city DR, and remote DR centers are shown in Figure 2:

          Figure 2 Remote replication relationships after primary/secondary switchover
          
        • Synchronous remote replication is in the Split state between the production center and the same-city DR center, and data synchronization is stopped.
        • The storage array in the remote DR center becomes the source array of the asynchronous remote replication.
  2. Synchronize service data from the remote DR center to the storage array in the production center.
     1. Log in to a host or VM in the remote DR center and stop services.

        If the protected objects are databases, close the application system and unmount disks.

        If the protected objects are local file systems, stop programs and files on the DR host.

        If the protected objects are VMware VMs, close VMs and uninstall datastores.

     2. Log in to the management software of the storage array in the remote DR center and delete mapping views between LUNs and the DR host.
     3. Select Enable write protection for secondary LUN for consistency groups of asynchronous remote replication.
     4. Synchronize data for consistency groups of asynchronous remote replication.
  3. After data is synchronized, split the asynchronous remote replication from the remote DR center to the production center.
     1. Log in to the management software of the storage array in the remote DR center and split consistency groups of the asynchronous remote replication.
     2. Deselect Enable write protection for secondary LUN for consistency groups of asynchronous remote replication.
  4. Log in to the storage array in the production center and restore asynchronous remote replication between storage arrays in the production and remote DR centers.
     1. Log in to the management software of the storage array in the production center and perform primary/secondary switchover for consistency groups of asynchronous remote replication.
     2. Select Enable write protection for secondary LUN for consistency groups of asynchronous remote replication.
     3. Perform data synchronization for consistency groups of asynchronous remote replication.

        Asynchronous remote replication is restored between storage arrays in the production remote DR centers, and the storage array in the production center is the primary array of the asynchronous replication.

  5. Log in to the management software of the storage array in the production center and synchronize data for consistency groups of asynchronous remote replication to synchronize data from the production center to the same-city DR center.

     Data replication is implemented between storage arrays in the remote DR, production, and same-city DR centers, and data is consistent between the three centers.

• Synchronous remote replication is performed between storage arrays in the production and same-city DR centers. The storage array in the production center is the primary array of the synchronous replication.
• Asynchronous remote replication is performed between storage arrays in the production and remote DR centers. The storage array in the production center is the primary array of the asynchronous replication.
  1. Test production services in the production center.
     1. Log in to the management software of the storage array in the production center and restore mapping views between LUNs and the production host.
     2. Log in to the production host or VM, and test services.
        • If the protected objects are databases, mount disks, start application systems, and test services.
        • If the protected objects are local file systems, start programs and files on the DR host, and test services.
        • If the protected objects are VMware VMs, attach datastores, start VMs, and test services.
  2. Restore DR configurations.
     1. Log in to eReplication in the same-city DR center, refresh its managed storage array and host resources, and delete invalid DR protected groups and recovery plans.
     2. Re-create DR protected groups and recovery plans.

Result

• After service failback in a synchronous and asynchronous parallel network:
  • Services on the production center can be accessed properly.
  • Data is synchronously replicated from the production center to the same-city DR center, and is asynchronously replicated from the production center to the remote DR center.
  • The DR management server completes the DR on the parallel network.
• After service failback in an asynchronous and asynchronous parallel network:
  • Services on the production center can be accessed properly.
  • Data is synchronously replicated from the production center to the same-city DR center, and is asynchronously replicated from the production center to the remote DR center.
  • The DR management server completes the DR on the parallel network.