4.2.3Creating a Directory Server instance manageable at the command line instead of Console

To create a Directory Server instance so that you can manage the instance through the command line or other tools instead of through the Console, use the /opt/dirsrv/sbin/setup-ds.plcommand. All tasks that can be performed from the Console can be performed from the command line or by other means. This is advantageous if you have a large number of Directory Server instances to manage, especially when they require tasks that can be more easily performed from the command line or by scripting (such as configuration of replication, LDIF, or SSL). This is also advantageous if you have to manage remotely and network traffic or connection quality makes use of the Console unviable (because of the additional network traffic incurred from graphics-intensive screens).

The script works exactly the same way as setup-ds-admin.plexcept it omits the questions about the Configuration Directory Server and Administration Server. It only creates a single, standalone Directory Server instance. (To manage the Directory Server instance with the Console, you must register the Directory Server instance with the Configuration Directory Server. For more information, see “Registering an existing Directory Server instance with the Configuration Directory Server”.) The script can be run silently as well, accepting command line options.

4.3Registering an existing Directory Server instance with the Configuration Directory Server

The Configuration Directory Server uses the o=NetscapeRoot database to store information about the Directory Servers and Administration Servers in your network. This database is used by the Console and the Administration Servers. This database can belong to a separate Directory Server instance, called the Configuration Directory Server. When you set up a Directory Server instance using the setup-ds-adminscript, the instance is always registered with a Configuration Directory Server. When you use the setup-dsscript to set up a Directory Server instance so that it is manageable at the command line instead of at the Console (see “Creating a Directory Server instance manageable at the command line instead of Console” (page 40)), the instance is not registered with a Configuration Directory Server. If you decide you want that instance to be manageable at the Console, you can manually register it with a Configuration Directory Server by launching the register-ds-adminscript:

#/opt/dirsrv/sbin/register-ds-admin.pl

IMPORTANT:

Running register-ds-admincreates a default instance of the Administration Server and Configuration Directory Server if they do not already exist, then registers any existing Directory Servers with the Configuration Directory Server.

4.4 Uninstalling Directory Server

This section provides information on removing a single Directory Server instance as well as the Directory Server entirely.

4.4.1 Removing a single Directory Server instance

It is possible to remove a single instance of Directory Server without uninstalling the system. Do this with the following command:

#/opt/dirsrv/sbin/ds_removal -s server_id -w admin_password

The ds_removal script unregisters the server from the Configuration Directory Server and removes any related files and directories. If the Directory Server was configured to support SSL, the command leaves the key and cert files in the instance configuration directory, and renames the configuration directory to instance-name.removed.

40 Post-installation and advanced configuration tasks