Sun Microsystems
Products & Services
 
Support & Training
 
 

Previous Previous     Contents     Index     Next Next
Chapter 9

Adding a Diskless Node to a Cluster Originally Created Using the nhinstall Tool

There are two ways that you can add a diskless node to a cluster that was originally created using the nhinstall tool:

  • By using the add option of the nhinstall tool

  • By manually adding the node to the cluster

For information, see the following sections:

Adding a Diskless Node By Using the nhinstall Tool

To add a node to the cluster, you must first determine whether the cluster has predefined nodes. A predefined node is a node that you defined when you originally created the cluster, but this node was not physically connected to the cluster. For instructions on how to check whether your cluster has predefined nodes, see the Netra High Availability Suite Foundation Services 2.1 6/03 Cluster Administration Guide.

However, if your cluster does not have predefined nodes and you want to add new diskless nodes to the cluster, you can use the nhinstall add command to do so.

ProcedureTo Add a Predefined Diskless Node to the Cluster

  1. Verify that the hardware of the new node is supported for your cluster configuration.

    For information, see the Netra High Availability Suite Foundation Services 2.1 6/03 Hardware Guide.

  2. Verify that after adding the node, your cluster configuration will be supported.

    For information, see the Netra High Availability Suite Foundation Services 2.1 6/03 Hardware Guide.

  3. Physically connect the node to the cluster.

    For information, see the Netra High Availability Suite Foundation Services 2.1 6/03 Hardware Guide.

  4. On the console of the new node, configure the OBP parameters:

    ok> setenv local-mac-address ? true
    ok> setenv auto-boot-retry? true
    ok> setenv diag-switch? false
    ok> setenv boot-device net:dhcp,,,,,5 net2:dhcp,,,,,5

  5. Boot the node:

    # init 6

ProcedureTo Add an Undefined Diskless Node to the Cluster

  1. Confirm that the cluster was originally created using the nhinstall tool.

    For information, see Netra High Availability Suite Foundation Services 2.1 6/03 Cluster Administration Guide.

  2. Log in to the master node as superuser.

  3. Verify that there is a Solaris environment for diskless nodes.

    The Solaris environment for diskless nodes exists if there is a directory called /export/os, where os is Solaris_8 or Solaris_9.

    If the Solaris environment does not exist for diskless nodes, install it as described in Chapter 7, Installing the Software for Diskless Nodes.

  4. Verify that the Foundation Services are running:

    # /opt/SUNWcgha/sbin/nhcmmstat -c all

    This command returns the status of all nodes in the cluster.

  5. Confirm that the cluster_definition.conf, env_installation.conf, and addon.conf files are present on the installation server, and that they conform to the original installation.

  6. On the installation server, edit the cluster_definition.conf file to define the node that you want to add.

    For example, on a cluster containing two master eligible nodes, NODE=10 and NODE=20, add a diskless node, NODE=30.
    NODE=10	08:00:20:f9:c5:54		- node10
    NODE=20	08:00:20:f9:a8:12		- node20
    NODE=30	-						- node30

    To add more than one node, define all the nodes that you want to add.

    If you add a new node or nodes to your cluster, make sure that you are not going to create an unsupported cluster configuration. For information about supported cluster configurations, see the Netra High Availability Suite Foundation Services 2.1 6/03 Hardware Guide.

    Ensure that the node ID is unique and does not already exist in the cluster_nodes_table file on the master node.

    When you have a running cluster, do not edit any other parameters in the cluster_definition.conf file. For information, see the cluster_definition.conf(4) man page.

  7. Remove the temporary installation files on the installation server before you add a new node to the cluster:

    # nhinstall -r config-file-directory reset

  8. On the installation server, add the new node:

    # nhinstall -r config-file-directory add nodeid30 [ nodeid40 ...]

    where nodeid30 is the node ID of the first new diskless node that you want to add.

    To add more than one node, specify the node ID of each new node [nodeid40 nodeid50 ...].

    For more information about the nhinstall add command, see the nhinstall(1M) man page.


    Note - Do not add more nodes to the cluster_definition.conf file than you are going to add to the cluster with the nhinstall add command.


  9. Connect the Ethernet interfaces (NIC0 and NIC1) of the new node to the switches.

  10. Access the console of the new node and get the ok prompt:

    # Control-]
    telnet> send brk
    Type  'go' to resume
    ok>

  11. On the console of the new node, configure the OBP parameters:

    ok> setenv local-mac-address? true
    ok> setenv auto-boot-retry? true
    ok> setenv boot-device net:dhcp,,,,,5 net2:dhcp,,,,,5
    ok> setenv diag-switch? false

  12. Boot the new node:

    ok> boot

  13. Become superuser.

  14. Verify that the node is configured correctly:

    # nhadm check

Previous Previous     Contents     Index     Next Next