Adding a peer replica to an existing server

For failover support, it might be necessary for you to add a peer replica for a backend to an existing server or set of servers. These servers can be stand-alone or already actively replicating.

In order to add a peer replica for a backend to a z/OS LDAP server, you should do the following:

  1. Start the new peer replica in maintenance mode. The peer replica must have a peerServerDN and peerServerPW defined in the backend section of the LDAP server configuration file.
  2. Stop the existing servers. For each existing server that is to become a peer server, update its configuration file to include the peerServerDN and peerServerPW configuration options. Restart the existing read/write servers in maintenance mode. See Establishing the root administrator DN and basic replication replica server DN and passwords for alternatives to specifying the password in the configuration file.
  3. Prime the new peer replica with all the data from an existing server. You can accomplish this by dumping the existing server's directory (for TDBM or LDBM, use ds2ldif) and adding the data to the new peer replica (for TDBM or LDBM use ldapadd or, for TDBM, use ldif2ds). See Populating a replica for more information.
  4. Add a replica entry to the existing servers to point to the new peer replica.
  5. Add a replica entry in the new peer replica pointing to the existing server that was used to prime this server.
    Note: If the existing server was a replicating server with replica entries defined to it, those replica entries might have been copied to the new peer replica in step 3 above. Ensure that this server does not contain a replica entry that defines this server as a replica of itself.
  6. Turn off maintenance mode on all servers.

The existing servers and the new peer replica are now peer read/write replicas.