Confirm that the spanning tree protocol is disabled.

For Cisco 29x0 switches, do the following:

1. Telnet to the Ethernet switch.

2. Type the following command:

   
   

   # enable Password <user-password>

   
   

3. Type the following command:

   
   

   # show run

   
   

4. Search the output on the console for the following line:

   
   

   no spanning-tree vlan <vlanid>

   
   

   If the display contains this line, the spanning tree is disabled.

   • If the spanning tree is disabled, go Step 8.

   • If the spanning tree is not disabled, disable it.

     For information, see the Netra High Availability Suite 3.0 1/08 Foundation Services Getting Started Guide.

Confirm that the DHCP configuration is correct.

1. Access the consoles of the master node and vice-master node.

2. On each console, confirm that the /etc/inet/dhcpsvc.conf file exists and has the correct attributes:

   
   

   # cat /etc/inet/dhcpsvc.conf DAEMON_ENABLED=TRUE RUN_MODE=server RESOURCE=SUNWnhrbs PATH=/SUNWcgha/remote/var/dhcp CONVER=1 INTERFACE=nic0,nic1

   
   

   The RESOURCE parameter must be set to RESOURCE=SUNWnhrbs. By default, this parameter is set to RESOURCE=SUNWfiles.

   The PATH parameter must point to a directory in a replicated file system. By default, the directory is PATH=/SUNWcgha/remote/var/dhcp.

   If the file does not have the correct attributes, do the following:

   • Edit or create the /etc/inet/dhcpsvc.conf file, setting the attributes as stated previously.

   • Stop and restart the DHCP daemon.

     If your cluster is running the Solaris 9 OS, issue the following commands:

     
     

     # /etc/rc3.d/HA.S34dhcp stop # /etc/rc3.d/HA.S34dhcp start

     
     

     If your cluster is running the Solaris 10 OS, issue the following commands:

     
     

     # svcadm restart  svc:/network/dhcp-server:default

     
     

3. If you are installing your cluster manually, confirm that the path has DHCP container files with the following name:

   SUNWnhrbs1_10_x_1_0, SUNWnhrbs1_10_x_2_0, and SUNWnhrbs1_dhcptab, where x is the domain identity.

   You do not need to perform this step if you are installing your cluster using the nhinstall tool.

   If the DHCP container files do not have the specified name, regenerate them, taking care to use the correct values for subnet1 and subnet2. For information about configuring DHCP for a diskless node, see the Netra High Availability Suite 3.0 1/08 Foundation Services Manual Installation Guide for the Solaris OS.

4. If you are using a static address assignment, confirm that the MAC address or client ID of the diskless node is configured correctly.

   Refer to the DHCP table on the master node.

For SPARC, on the console of the diskless node, confirm that the following OpenBoot PROM parameter is set:

boot-device net:dhcp,,,,,5 net2:dhcp,,,,,5

Confirm that the vendor type of the diskless node is recognized by the master node.

1. Access the console of the master node.

2. Type the following command:

   
   

   # snoop -v -d nic0 ether mac-adr-of-dl-node | grep -i dhcp

   
   

   or

   
   

   ok> dev ok> .properties => property

   
   

   The vendor type of the diskless node is returned as a string.

3. Search for the same string in the DHCP table on the master node.

On the console of the master node, confirm that the directory /tftpboot is present.

If this directory is not present, the following error message is written to the system log files:

Timeout waiting for BOOTP/DHCP reply. Retrying...
TFTP Error Access violation

If the /tftpboot directory is not present on the vice-master node, the diskless node does not boot after a switchover. To set up the /tftpboot directory on the vice-master node, see the Netra High Availability Suite 3.0 1/08 Foundation Services Manual Installation Guide for the Solaris OS.

On the console of the master node, confirm that the following directory contains a file for each diskless node interface: diskless_file_system/root/diskless_nodeid/etc/

The files could be named as follows:

hostname.hme0hostname.hme1

If the directory does not contain a file for an interface, the interface cannot be configured.

Examine the access permissions of root, swap, and usr in the nhfs.conf file on the master node.

If your cluster was installed manually or by the nhinstall tool, confirm that the following access permissions are set:

share -F nfs -o rw,root=diskless_node_id-nic0:diskless_node_id-nic1:
		diskless_node_id-cgtp0 /export/root/diskless_node_idshare -F nfs -o rw,root=diskless_node_id-nic1:diskless_node_id-nic1:
		diskless_node_id-cgtp0 /export/swap/diskless_node_idshare -F nfs -o ro /export/exec/Solaris_X_sparc.all/usr

where X is the version of the Solaris OS installed on the cluster.

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

Stop the continuous reboot cycle if such a cycle is running:

1. Access the console of the failing node.

2. Type the following command:

   
   

   # halt ok>

   
   

   Alternatively, type the following command:

   
   

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

   
   

   The ok prompt is returned.

3. Boot in single user mode:

   
   

   ok> boot -s #

   
   
   

   ---

   Note - For information about performing this task on x64 platforms, refer to the products hardware documentation.

   ---

   
   

Search the messages displayed on the console of the failing node for an indication of the problem.

The error messages should indicate the cause of the problem. Use the error messages to identify the failing daemon or failing service. For a summary of error messages and their possible causes, see Appendix A.

If the error is a configuration error, the following message is displayed:

Error in configuration

The text following the message should indicate the type of configuration error. Verify that the configuration of the nhfs.conf file for the node is consistent with the information in the nhfs.conf4 man page.

Confirm that the /etc/opt/SUNWcgha/not_configured file does not exist on the failing node.

• If the file does not exist, go to Step 9.

• If the file exists, delete it and reboot the node.

Confirm that the cluster_node_table file on the master node contains an entry for the failing node.

• If the file contains an entry for the failing node, go to Step 12

• If the file does not contain an entry for the failing node, verify the installation and configuration.

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

Stop the continuous reboot cycle if such a cycle is running.

For information, see Step 1 of To Investigate Why the Netra HA Suite Does Not Start on a Diskless Node.

Search the messages on the console of the failing node for an indication of the problem.

For information, see Step 5 of To Investigate Why the Netra HA Suite Does Not Start on a Diskless Node.

Confirm that the /etc/opt/SUNWcgha/not_configured file does not exist on the failing node.

For information, see Step 6 of To Investigate Why the Netra HA Suite Does Not Start on a Diskless Node.

Confirm that the cluster_node_table file on the master node contains an entry for the failing node.

For information, see Step 9 of To Investigate Why the Netra HA Suite Does Not Start on a Diskless Node.

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