Replicating client data to another server

Replicating client data from a source server to another server helps to ensure that backed-up client data is available for recovery if the source server is damaged. Replication incrementally copies data from the source server to the target server to provide failover and failback capability.

About this task

If a disaster occurs and the source server is temporarily disabled, client nodes can recover their data from the target server. If the source server cannot be recovered, you can change client node configurations to store data on the target server. When an outage occurs, the source server can automatically fail over to a target server for data recovery.
Restriction: A server can replicate data to only one target server.

You can replicate data that is stored in an active retention set. This data might be expired according to policy settings, but it is protected from expiration if it is contained in an active retention set.

You can replicate data that is stored in any type of storage pool. The storage pool type can be different at the source replication server and the target replication server. You can control replication by type of client node data:
  • Active and inactive backup data together, or only active backup data
  • Retained backup data
  • Archive data
  • Data that was migrated to a source server by IBM Spectrum® Protect for Space Management clients

When you replicate data in directory-container storage pools, use storage pool protection to improve the efficiency of the replication process, and to enable repair of data. When you use the Operations Center to set up your storage pools, schedules for protection are automatically defined to coordinate with the replication schedule.

Tip: The preferred method for configuring replication is to use the Operations Center. In this way, you can streamline the configuration process. To configure replication by using the command line, see technote 1274740.

Procedure

  1. Verify that servers are compatible and have the system resources for successful use of replication.

    Increased amounts of memory and processor cores are required. The database and its logs must be sized to ensure that transactions can complete. A dedicated network, with enough bandwidth to handle the amount of data you intend to replicate, is required.

    1. Verify that the source and target servers are compatible for replication. See Replication compatibility.
    2. Verify that the server has appropriate resources for good performance. For details, see Checklist for node replication.
  2. Enable replication. See Enabling node replication.
  3. Schedule replication for the source server. For information about how to integrate this schedule in regular server maintenance schedules, see Defining schedules for server maintenance activities.
  4. Schedule storage pool protection for all directory-container storage pools on the source server. See Protecting data in directory-container storage pools.
  5. Monitor replication by using the Operations Center. For more information, see Daily monitoring checklist.

What to do next

In rare cases, the server database might fail to respond in a replicated environment. For instructions about resolving the issue, see technote 2017079.