Enabling Elasticsearch on Scaled Out Nodes

Before You Begin

This 15-minute tutorial shows you how to make Elasticsearch available on all scaled out nodes/VMs. This includes the scaled out nodes for the WebCenter Portal Cloud Service, WebCenter Content, Oracle Inbound Refinery (IBR) node, and the Portlet Server nodes.

There must be minimum three nodes/VMs before you can enable Elasticsearch in the high-availability (HA) environment.

This is the seventh tutorial in the Provisioning WebCenter Portal Cloud Service in the EDG Topology series. Read them sequentially.

Background

Oracle WebCenter Portal Cloud Service is delivered on top of Oracle Java Cloud Service. When you create a WebCenter Portal Cloud Service instance, the VM(s) required to support the instance are provisioned and configured for you. Your instance is also preconfigured to use the specified Oracle Database Cloud Service and Oracle Storage Cloud Service instances.

In the EDG topology, WebCenter Portal Cloud Service, WebCenter Content, Oracle Inbound Refinery (IBR), and Elasticsearch are installed on the same VM. You can choose to replicate the VM for high availability. Oracle Traffic Director (OTD), or the load balancer, is installed on a separate VM. If you choose to install the optional component Portlet Server, you must install it on a separate VM.

What Do You Need?

Completion of the previous tutorials so that you have:

Created an Oracle Database Cloud Service instance for WebCenter Portal Cloud Service
Created an Oracle Java Cloud Service instance for WebCenter Portal Cloud Service
Scaled out the WebCenter Portal Cloud Service/WebCenter Content/IBR node and logged on to WebCenter Portal Cloud Service
Optionally installed the Portlet Server

Connect to the Administration Server VM

Use an ssh client to connect to your administration server VM. For example:
```
ssh -i private_key_file opc@admin_server_ip
```
Where:
- private-key-file refers to the private key file required to connect to the administration server.
- admin_server_ip refers to the IP address of the administration server. You can find the IP address under the WebLogic section on the Instance Overview page of your WebCenter Portal Cloud Service instance on the Oracle Java Cloud Service console.

Enable Elasticsearch on the Scaled Out WebCenter Portal Node

Switch to the /u02/d01/autodeploy directory.
```
cd /u02/d01/autodeploy
```
Run the addToTopology command on the scaled out node for WebCenter Portal Cloud Service.
Run the command as the opc user. When you connect to a VM, by default you are connected as the opc user.
```
./addToTopology  -private-key private_key_file -node node_name -es-server true
```
Where:
- private_key_file refers to the full path of the private key file.
- node_name refers to the scaled out node/VM of WebCenter Portal Cloud Service, WebCenter Content, and IBR. You can find the node name under the Weblogic section on the Instance Overview page of your WebCenter Portal Cloud Service instance.
- es-server refers to the flag that indicates whether to enable Elasticsearch. Elasticsearch is enabled if the value is true.
The following is a sample command:
```
./addToTopology  -private-key /stage/rsa-key-20170619.ssh -node wcpcloud1-wls-3 -es-server true
```
If your SSH key has a password, the script will prompt for the SSH password. When the script prompts for the JSK instance password, specify the password for your Java Cloud Service instance. The script will also prompt for the database instance password.

Enable Elasticsearch on Other Nodes

Enable Elasticsearch on all the other scaled out nodes. If you've scaled out the WebCenter Portal Cloud Service, WebCenter Content, and IBR node on a third node, enable Elasticsearch on that node. If you've installed Portlet Server and scaled out the node, you can enable Elasticsearch on all Portlet Server nodes. There must be a minimum of three nodes for you to be able to enable Elasticsearch in the HA environment.

Run the addToTopology command for the node for which you want to enable Elasticsearch.
```
./addToTopology  -private-key private_key_file -node node_name -es-server true
```
Where:
- private_key_file refers to the full path of the private key file.
- node_name refers to the name of the VM for which you want to enable Elasticsearch. You can find the node name on the Instance Overview page of your WebCenter Portal Cloud Service instance.
- es-server refers to the flag that indicates whether to enable Elasticsearch. Elasticsearch is enabled if the value is true.
The following is a sample command:
```
./addToTopology  -private-key /stage/rsa-key-20170619.ssh -node wcpcloud1-wls-4 -es-server true
```
If your SSH key has a password, the script will prompt for the SSH password. When the script prompts for the JSK instance password, specify the password for your Java Cloud Service instance. The script will also prompt for the database instance password.
Note: Similarly, you can run the addToTopology command for the other nodes for which you want to enable Elasticsearch.

Next Tutorial

Adding Users to Oracle Unified Directory