Sun Cluster 3.0 12/01 Data Services Installation and Configuration Guide

Previous: Configuration Guidelines for Sun Cluster HA for BroadVision One-To-One Enterprise
Next: Installing the Sun Cluster HA for BroadVision One-To-One Enterprise Package

Installing and Configuring the BroadVision One-To-One Enterprise Software, the Database, and the HTTP Server

Use the procedures in this section to perform the following tasks.

Install and configure your database software to run in a Sun Cluster environment.
Install and configure your HTTP software to run in a Sun Cluster environment.
Install and configure the BroadVision One-To-One Enterprise software to run in a Sun Cluster environment.
Verify the BroadVision One-To-One Enterprise, database, and HTTP server installation.

Note -

Before you install the BroadVision One-To-One Enterprise, database, and HTTP server software in the Sun Cluster environment, run the scstat(1M) command to verify that the Sun Cluster software is fully operational.

How to Install and Configure Sun Cluster HA for DBMS

See Chapter 2, Installing and Configuring Sun Cluster HA for Oracle to install Sun Cluster HA for Oracle or Chapter 10, Installing and Configuring Sun Cluster HA for Sybase ASE to install Sun Cluster HA for Sybase ASE.

How to Install and Configure Your HTTP Server

If iPlanet Web Server is your HTTP server, follow the instructions in Chapter 3, Installing and Configuring Sun Cluster HA for iPlanet^TM Web Server to configure Sun Cluster HA for iPlanet Web Server. If Apache Web Server is your HTTP server, follow the instructions in Chapter 5, Installing and Configuring Sun Cluster HA for Apache to configure Sun Cluster HA for Apache.

How to Install and Configure the BroadVision One-To-One Enterprise Software

This procedure describes how to install and configure the BroadVision One-To-One Enterprise software and how to enable the BroadVision One-To-One Enterprise software to run in the Sun Cluster environment.

Follow the guidelines that are listed in "Configuration Guidelines for Sun Cluster HA for BroadVision One-To-One Enterprise" and "Pre-Installation Considerations".

Follow the instructions in the BroadVision One-To-One Enterprise Installation and Administration Guide to install the BroadVision One-To-One Enterprise software on the cluster file system.

Note -
Install the BroadVision One-To-One Enterprise software only once, on the cluster file system, from any cluster node.

Configure the $BV1TO1_VAR/etc/bv1to1.conf file.

Table 11-3 summarizes possible configurations in the $BV1TO1_VAR/etc/bv1to1.conf file for the BroadVision One-To-One Enterprise components. See "Supported Configurations" and the instructions in the BroadVision One-To-One Enterprise Installation and Administration Guide for details.

Table 11-3 Configuring the $BV1TO1_VAR/etc/bv1to1.conf File


BroadVision One-To-One Enterprise component	Where to configure
Root host	Logical hostname
Back-end servers	Logical hostname
Interaction Managers	All of the cluster nodes or all of the cluster private hostnames [ See the Sun Cluster 3.0 12/01 Software Installation Guide for details on cluster interconnect and private hostnames.]

Note -

If you configure the Interaction Managers on all of the cluster private hostnames, set up the HTTP servers on the same cluster. Alternatively, if you configure the Interaction Managers on all of the cluster nodes, the HTTP servers can be set up outside of the cluster.

Note -

Configure your cluster so that BroadVision One-To-One back-end servers can access the database from any cluster node.

Where to Go From Here

Depending on the flexibility and granularity of administration that you require for each back-end resource, you can set up your failover resource groups in one of the following ways.

Set up multiple failover resource groups to use multiple logical hostnames. If you plan to use this option, go to "How to Verify the Sun Cluster HA for BroadVision One-To-One Enterprise Installation".
Set up one failover resource group to use n logical hostnames and to contain all of the back-end and root host resources. If you plan to use this option, proceed to "Alternative Configuration". Follow the procedures throughout "Alternative Configuration" to complete the installation.

Note -

See "Supported Configurations" for more information.

How to Configure and Verify the BroadVision One-To-One Enterprise, Database, and HTTP Server Installation

Perform this procedure to test starting and stopping the back-end processes on all of the nodes on which the back-end host and root host can run in a failover configuration. Additionally, perform this procedure to test the BroadVision One-To-One Enterprise Interaction Managers that you configured in the cluster.

Depending on the flexibility and granularity of administration that you require for each back-end resource, you can set up your failover resource groups in one of the following ways.

Set up multiple failover resource groups to use multiple logical hostnames. If you plan to use this option, proceed to Step 1.
Set up one failover resource group to use n logical hostnames and to contain all of the back-end and root host resources. If you plan to use this option, go to "Alternative Configuration". Follow the procedures throughout "Alternative Configuration" to complete the installation.

Note -

See "Supported Configurations" for more information.

To contain the BroadVision One-To-One Enterprise root host resource, create a failover resource group that uses the root host logical hostname.
# scrgadm -a -g root-host-resource-group [-h nodelist]
-g root-host-resource-group

Specifies the name of the resource group that uses the root host logical hostname and contains the BroadVision root host resource. The name of the root host resource group can be your choice but must be unique for resource groups within the cluster.

[-h nodelist]

Specifies an optional, comma-separated list of physical node names or IDs that identify potential masters. The order here determines the order in which the Resource Group Manager (RGM) considers primary nodes during failover.

Create failover resource groups for the root host and back-end processes.

Run the scrgadm(1M) command to configure n failover resource groups for back-end processes that are configured on n logical hostnames.

# scrgadm -a -g back-end-resource-group-1 [-h nodelist]
# scrgadm -a -g back-end-resource-group-2 [-h nodelist]
# scrgadm -a -g back-end-resource-group-3 [-h nodelist]
...
# scrgadm -a -g back-end-resource-group-n [-h nodelist]

-g back-end-resource-group: Specifies the name of the resource group that contains the back-end logical hostname and resource. The name of the back-end resource group can be your choice but must be unique for resource groups within the cluster.

Verify that you have added all of the logical hostnames that you use to your name service database.

Additionally, add all of the logical hostnames that you use to the /etc/inet/hosts file on each cluster node. Therefore, if the name service goes down, the nodes can still find the name-to-address mapping on their local hosts file.

Run the scrgadm command to add the logical hostname that each of the resource groups that you have created can use.

# scrgadm -a -L -g root-host-resource-group -l root-host-logical-hostname-1 [-n netiflist]
# scrgadm -a -L -g back-end-resource-group-1 -l back-end-logical-hostname-1 [-n netiflist]
# scrgadm -a -L -g back-end-resource-group-2 -l back-end-logical-hostname-2 [-n netiflist]
...
# scrgadm -a -L -g back-end-resource-group-n -l back-end-logical-hostname-n [-n netiflist]

-l root-host-logical-hostname: Specifies the logical hostname (failover IP address) that the root host resource group uses.
-l back-end-logical-hostname: Specifies the logical hostname that each back-end resource group uses.
[-n netiflist]: Specifies an optional, comma-separated list that identifies the NAFO groups that are on each node. All of the nodes in the resource group's node list must be represented in netiflist. If you do not specify this option, the scrgadm command attempts to discover a net adapter on the subnet that the hostname list identifies for each node that is in the node list. For example, -n nafo0@nodename, nafo0@nodename2.

Create a scalable resource group for the Interaction Managers.
# scrgadm -a -g im-resource-group -y Maximum_primaries=m -y Desired_primaries=n
-g im-resource-group

Specifies the name of the scalable resource group that contains the Interaction Managers. This name can be your choice but must be unique for resource groups within the cluster.

-y Maximum_primaries=m

Specifies the maximum number of active primary nodes allowed for this resource group. If you do not assign a value to this property, the default is 1.

-y Desired_primaries=n

Specifies the desired number of active primary nodes allowed for this resource group. If you do not assign a value to this property, the default is 1.

From one cluster node, run the scswitch(1M) command to move the failover resource groups into the managed state and bring them online.
# scswitch -Z -g root-host-resource-group # scswitch -Z -g back-end-resource-group-1 # scswitch -Z -g back-end-resource-group-2 ... # scswitch -Z -g back-end-resource-group-n
Note -
You do not need to bring the scalable resource group online because the scalable resource group does not yet contain resources. You must bring failover resource groups online because the BroadVision One-To-One Enterprise back-end processes cannot start if the logical hostname resource is unavailable.

Check that the database is accessible.

See your database documentation for details.

Ensure that you have configured the database to enable BroadVision One-To-One back-end servers to access the database from any cluster node.

See your database documentation for details.

As the BroadVision user, log in to the cluster node that hosts the root host resource group.

Follow the steps in the BroadVision One-To-One Enterprise Installation and Administration Guide to run the following BroadVision commands.
1. Set the BV_LOCAL_HOST environment variable as root-host-logical-hostname.
2. Source the bv1to1.conf.sh file or the bv1to1.conf.csh file, depending on the shell that you use.
3. Run the bvconf bootstrap command on the root host to initialize the BroadVision One-To-One Enterprise installation.
  
  Note -
  Do not run the bvconf command as superuser.
  % bvconf bootstrap -r root-host-logical-hostname
4. Set the BV_LOCAL_HOST environment variable as back-end-logical-hostname or im-hostname.
5. Source the bv1to1.conf.sh file or the bv1to1.conf.csh file, depending on the shell that you use.
6. Ensure that the /etc/opt/BVSNsmgr directory exists and has write and execute permissions.
7. For each back-end host and Interaction Manager host, run the bvconf execute command to configure and start the BroadVision One-To-One Enterprise processes.
  % bvconf execute -local -var shared -r root-host-logical-hostname

Run the BroadVision command bvconf gateway to generate gateway configuration files for the HTTP gateway applications.

This command generates the files and writes them to the $BV1TO1_VAR/etc/appName.cfg file.
% bvconf gateway -A appName
-A appName

Specifies the gateway application name, which is defined in the $BV1TO1_VAR/etc/bv1to1.conf configuration file. See the BroadVision One-To-One Enterprise Installation and Administration Guide for details.

Copy the gateway application configuration file to the /etc/opt/BVSNsmgr directory on each of the cluster nodes that runs HTTP instances.

Note -
Ensure that you copy the gateway application configuration file with the extension .cfg.

See the BroadVision One-To-One Enterprise Installation and Administration Guide for details.

Configure and start the HTTP servers.

See your HTTP server documentation for details. Additionally, see the BroadVision One-To-One Enterprise Installation and Administration Guide for information on HTTP server configuration.

From a BroadVision client, connect to the BroadVision site, and check the installation.

If the BroadVision One-To-One Enterprise software is functioning correctly, perform the following steps to shut down the Interaction Managers, back-end processes, and root host processes.
1. Shut down the Interaction Managers.
  1. Set the BV_LOCAL_HOST environment variable as im-hostname.
  2. Source the bv1to1.conf.sh file or the bv1to1.conf.csh file, depending on the shell that you use.
  3. Run the following command.
    # bvconf shutdown -local
2. Shut down the back-end processes.
  1. Set the BV_LOCAL_HOST environment variable as back-end-logical-hostname-n.
  2. Source the bv1to1.conf.sh file or the bv1to1.conf.csh file, depending on the shell that you use.
  3. Run the following command.
    # bvconf shutdown -local
3. Shut down the root host processes.
  1. Set the BV_LOCAL_HOST environment variable as root-host-logical-hostname.
  2. Source the bv1to1.conf.sh file or the bv1to1.conf.csh file, depending on the shell that you use.
  3. Run the following command.
    # bvconf shutdown -local

Run the scswitch command to switch the resource groups to another cluster node, such as node2.

# scswitch -z -g root-host-resource-group -h node2
# scswitch -z -g back-end-resource-group-1 -h node2
# scswitch -z -g back-end-resource-group-2 -h node2
...
# scswitch -z -g back-end-resource-group-n -h node2

Restart the BroadVision One-To-One Enterprise software on node2.

Connect to the cluster from a BroadVision client, and check that the BroadVision One-To-One Enterprise software functions correctly.

Repeat Step 15 through Step 18 on all of the potential primaries of the BroadVision One-To-One Enterprise resource groups.

Where to Go From Here

After you verify the BroadVision One-To-One Enterprise, database, and HTTP server installation, go to "How to Install Sun Cluster HA for BroadVision One-To-One Enterprise Packages".