After a new installation or an upgrade to Storage Foundation Suite 5.0 (tm), and the answer to the question "Are you ready to configure Storage Foundation now?" was no, an error occurs when attempting to connect to the host via the VERITAS Enterprise Administrator GUI, VEA. On the host, the module isisd fails to start and the above error occurs. If the system is installed using a method such as JumpStart, this question is not presented to answer, and therefore one of the below procedures must be followed to allow the VEA to connect to the host where Storage Foundation Suite (tm) 5.0 was installed.

There are two different methods to resolve the issue, and both are listed below. The first method is the preferred and is outlined in the VERITAS Storage Foundation (tm) Installation Guide 5.0 - Solaris. For more information about this guide, please refer to the Related Documents section below.

Solution 1:

For the VEA Server daemon, /opt/VRTSob/bin/vxsvc, to start, it needs to be configured so the proper management console can connect to it. The configuration options are to have the host managed by Storage Foundation Management Server (tm) Management, or to configure it in Stand Alone mode. To configure the VEA Server daemon, please follow the below directions:

Execute  the command "# installsf -configure". When the command "# installsf -configure" is executed, navigate through the prompts and answer "n" to the question "Enable Storage Foundation Management Server Management?". The Storage Foundation Manager Server (tm) (now called Storage Foundation Manager (tm) product is beyond the scope of this document. For more information on this product, please refer to the production documentation available in the Related Documents section below. Please see below for a the specific prompt regarding the configuration question that must be answered

Solution 2: (tested with Solaris 10 and SF 5.0)
If the "installsf -configure" script cannot be run, or if the script is not in the /opt/VRTS/install folder (this may be due to fact that customer ran installsfrac instead of installsf), the below procedure can be followed to configure the VEA Server:

Verify first that the following packages are installed on the system.  If they are not, then they must be installed manually:

VRTSaa  

VRTSalloc      

VRTSccg

VRTSdcli        

VRTSddlpr      

VRTSdsa

VRTSfspro      

VRTSgapms      

VRTSmh  

VRTSpbx

VRTSsmf

VRTSvail        

VRTSvmpro      

VRTSvrpro      

SYMClma

Prepare your system for configuration by halting any VEA services that are currently running:

# ps -ef | egrep "vxsvc|vxpal"

-Stop 'vxsvc'

# /opt/VRTSob/bin/vxsvc -k

-Stop 'vxpal.<agentname>' processes

# /etc/init.d/vxpal.<agentname> stop                (/sbin/init.d/ on HP-UX)

Run the following commands to configure the appropriate agents on the server.  Please make sure that you have set the $vxHOSTNAME variable or manually modify all occurrences of $vxHOSTNAME to reflect the system's hostname.

# vxHOSTNAME="servername.domain.com"

You may safely ignore errors related to gridnode, VAIL, and actionagent as these are not required elements for a VEA / VxVM management environment.  The StorageAgent must be configured as this provides support for Volume Manager objects in the VEA environment.

Setup the VEA Registry:

Configure VEA component agents:

-If you do not have VAIL installed you can ignore the error messages while registering.

# /opt/VRTSobc/pal33/install/cfgsecurity.sh -a VAILAgent -h $vxHOSTNAME -d -p -b

Now startup the processes necessary:

Verify processes started:

Start the VEA server.

Cleanup the Environment