Configure Cluster Nodes

This guide will go through the steps of configuring a machine other than the master node defined in a cluster. At this point the installer will have already created the cluster but the other machines are missing the components needed to host a managed server. This guide will use the terms Master Node and Slave Node where master node is the node the installer was executed on and the slave node is one of the machines added to the cluster during the installation.

It is good to have at least some basic understanding of how to control the servers.

Extending the Cluster

It is not possible to extend the cluster without using the IFS Installer. For information how to add additional cluster nodes using the Installer please check the installer guide

Adding servers to an already defined machine does not require anything more than to run the installer and create the servers. If, however you decide to add yet another machine to the cluster you will need to follow the process described below for each machine in the cluster.

  1. On the master node, browse to <ifs_home>/instance/<instance>/bin and run cluster.cmd/sh depending on the underlying operating system.
  2. The script will produce a compressed archive in <ifs_home>/instance/<instance>/ called cluster_<instance>.zip
    Move this archive to one of the slave nodes.
  3. Create a new IFS Home on the slave node in the same manner as described in the Intro Process Document and extract the archive in this new empty IFS Home.
    On Linux you should extract the archive as user IFS and keep using this user for the remainder of this guide.
  4. Still on the slave node, go to <ifs_home>/instance/<instance>/bin and there should be a file named cluster_install.cmd/sh depending on the underlying operating system.
    Run this file. The script will delete old application server and HTTP server configurations if any. Never run this on the master node.
  5. This script should populate the IFS Home with everything needed for the Managed Server(s) to work on the current machine and start the AS Node Manager. The HTTP Node Manager will not be started, if no external load balancer is used change the service startup type to 'Manual' instead of 'Automatic'. Note that no Managed Server(s) on this node will be running at this point.
  6. Repeat the previous steps for each machine in the cluster. The same archive created during the first steps can be reused, so omit those steps.
  7. Return to the master node and start all Managed Servers.
    Go to <ifs_home>/instance/<instance>/bin and run start_all_servers.cmd/sh

NOTE: These steps must be repeated after server/cluster changes is done in the installer.”

Configure the Cluster For an External Load Balancer

  1. On the master node, browse to <ifs_home>/instance/<instance>/bin and run cluster_client.cmd/sh depending on the underlying operating system.
  2. The script will produce a compressed archive in <ifs_home>/instance/<instance>/ called cluster_<instance>_client.zip.
    Move this archive to one of the slave nodes.
  3. Unpack cluster_<instance>_client.zip in <ifs_home> and it should add all the required files to run the client.
  4. All access requests to the documentation should be redirected from the slave nodes to the master node. In the file <ifs_home>/instance/<instance>/conf/mod_wl_ohs<instance>_documentation.conf, enable the following two statements by removing the leading # character and configure the URLs to the master node.
    ProxyPass /ifsdoc/ http://<master_node>:<port>/ifsdoc/
    ProxyPass /solr/ http://<master_node>:<port>/solr/
  5. Still one the slave node, start the HTTP Node Manager.
    Go to <ifs_home>/instance/<instance>/bin and run start_http_nodemanager.cmd if on Windows or ./http_nodemanager.sh start if on Linux.
  6. Start the HTTP Server.
    Go to <ifs_home>/instance/<instance>/bin and run start_http_server.cmd/sh.

IFS Enterprise Explorer deployment

Follow the IFS Enterprise Explorer deployment guide for the client to work on all nodes.

Remove a Cluster Node

Remove the managed server from the cluster like descibred here. Once the managed server has been removed you can remove the machine from the cluster like described here.
Note that this requires you to run the installer twice. Once to remove the managed servers associated to a machine and once to remove the actual machine configuration.
When all associations to a node has been removed, the IFS Home on this machine can be deleted.

Verify

From any node, browse to <ifs_home>/instance/<instance>/bin and run the script check_server_status.cmd/sh
All the Managed Server in the cluster should be in state RUNNING. The AdminServer should also be in state RUNNING.


If using an external load balancer, open any web browser and verify that each machine is serving requests on the http listener port specified in the installer.

Make sure the application server Node Manager is started. Check the log files for the particular Managed Server that is not running.


If using an external load balancer, make sure the the HTTP Node Manager is started and that the HTTP Server is running.