To define a cluster node

  1. Log into a Lawson Interface Desktop (LID) session, on any machine or node that will be in the cluster.
  2. From the System Administration menu, select Cluster Node Definition.

    - or -

    At the command line, specify this information:

    nodedef

    The Cluster Node Definition form appears.

    Form clip: Cluster Node Definition form
  3. On the Cluster Node Definition form, specify or select the following information:
    Field Description
    Name A short, descriptive name for the cluster node you want to define. Normally the host name is used.
    Host Address The host name, or the IP address of the machine.
    Host Name

    The actual name of your machine.

    Typically, if the host name is small enough, the name and the host name are the same.

    The host name must also be configured in an /etc/hosts, DNS, or NIS entry.
    Machine Type The type of UNIX machine you use.
    Description A short description of the machine.
    $GENDIR The path to the directory where your Lawson Environment objects are stored.
    $LAWDIR The path to the directory where your Lawson applications, configuration files, and spooled print output are stored.
    $LADBDIR The path to the directory where your Lawson database and dictionary files are stored.
    $LAWIPC

    Specify the $LAWIPC value for the Environment. This number is used to define the Lawson database IPC entries. Valid values are 0 to 255. If the LAWIPC value for your system has been defined in hexadecimal form, you must specify the decimal equivalent of that number here.

    LAWIPC allows two or more Lawson Environments on one machine to have individual message queues, shared memory, and semaphores. The LAWIPC number must be unique by machine, but you can use the same number for two different machines.
  4. Press Enter twice to save the cluster node definition.
  5. Repeat the previous steps for each node in the cluster.
  6. To test the cluster node definition, run the Cluster Node Definition utility (nodedef) from each machine or node in the cluster. In each case, you should be able to view all nodes in the cluster. If a node is not visible, check the definition of each node and make sure that each machine or node points correctly to the GEN database.