Search the console of the installation server for the IP address of the boot server. For example, for a Solaris installation:

Beginning system identification...
Searching for configuration file(s)...
Using sysid configuration file 10.101.1.253:/export/nhjumpstart/101/10/sysidcfg
Search complete.

For a Linux installation:

IP-Config: Got DHCP answer from 10.101.1.253, my address is \
10.101.1.10
IP-Config: Complete:
       device=eth0, addr=10.101.1.10, mask=255.255.255.0, \
       gw=10.101.1.253, host=cp3020-1, domain=, nis-domain=(none),
       bootserver=10.101.1.253, rootserver=10.101.1.253, \
       rootpath=/dist/mvista/CGE_4.0/target_tdp<7>eth0: no IPv6 \
       routers present

The IP address of the boot server in this example is the class C address 10.101.1.253.

Alternatively, use the name of the boot server to find the boot server IP address.

Find the IP address of your installation server.

Compare the address of the boot server with that of the installation server.

If the addresses are not the same, your node is being booted by the wrong machine. Perform the following steps:

1. If the boot server is running the Solaris OS, access the console of the wrong boot server as follows:

   1. Delete the MAC address of any nodes in your cluster in the /etc/ethers file.

   2. Delete the node parameters of any nodes in your cluster in the /etc/bootparams file.

   3. Delete the node parameters of any nodes in your cluster in the DHCP configuration files under /var/dhcp

   4. Restart the installation.

   OR

2. If the boot server is running Linux, access the console of the wrong boot server as follows:

   1. Delete the node parameters of any nodes in your cluster in the DHCP configuration file /etc/dhcpd.

   2. Restart the DHCP daemon.

3. Restart the installation

If you have not solved this problem, it is possibly because you have two installation servers for this cluster.

Confirm the presence of a second installation server using the /usr/sbin/bpgetfile command on a machine running on the same local network as your installation server.

If the problem persists, contact your customer support center.

Identify the problem by using the error message displayed on the installation server.

Possible problems are:

• local command failure on the installation server

• rsh command failure due to a connection or permission problem

• remote command failure

Correct the problem.

If the nhinstall tool stops due a problem with Solaris JumpStart, see Solaris JumpStart Installation Fails During nhinstall Installation.

Restart the nhinstall tool with the same options that you used to launch it the first time:

On the Solaris OS:

# /opt/SUNWcgha/sbin/nhinstall -r config_file_directory -l logfile

On Linux:

# /opt/sun/sbin/nhinstall -r config_file_directory -l logfile

The nhinstall tool resumes from the point at which it stopped.

If you have modified the cluster_definition.conf file to correct the error, the nhinstall tool displays a warning that the configuration has changed. If the change that you made to the file makes the cluster incoherent, you must reset the installation for a new installation. For information, see the Netra High Availability Suite 3.0 1/08 Foundation Services Manual Installation Guide for the Solaris OS.

If the nhinstall tool stops again, repeat Step 1 through Step 6 until the tool completes successfully.

If you cannot resolve this problem, contact your customer support center.

Confirm that the installation server and the first network interface of the master-eligible node are connected to the same switch.

• If they are, go to Step 4.

• If they are not, do the following:

  a. Configure your installation server as described in the Netra High Availability Suite 3.0 1/08 Foundation Services Getting Started Guide.

  b. Restart the installation.

  
  

  # /opt/SUNWcgha/sbin/nhinstall -r config_file_directory -l logfile

  
  

  If the Solaris JumpStart software does not restart, delete the /tmp/.install_client.lck file.

Confirm that a router is not connected between the installation server and the master-eligible node.

• If no router connects the installation server and the master-eligible node, go to Step 7.

• If a router connects the installation server and the master-eligible node, do the following:

  a. Remove the router.

  b. Reconfigure your hardware as described in the Netra High Availability Suite 3.0 1/08 Foundation Services Getting Started Guide.

  c. Restart the installation:

  
  

  # /opt/SUNWcgha/sbin/nhinstall -r config_file_directory -l logfile

  
  

  If the Solaris JumpStart software does not restart, delete the /tmp/.install_client.lck file.

If you cannot resolve this problem, contact your customer support center.