Prerequisites

• Ensure every node has Java JDK 8+ installed and environment variables setup for both JAVA_HOME and PATH
• Requires elasticsearch-1.5.2
  • Download Elastic Search from here
    
• Elastic Search nodes can either be setup in the same system or on different servers, but they must not be on the same server as the Application nodes
• All nodes must be on the same network

Setting up Elastic Search on the same nodes as Zephyr

Step 1:
Extract the files to your desired root location

Step 2:
In the config folder of the Elasticsearch root, you'll find a file named elasticsearch.yml. Edit this file and add these lines anywhere:

cluster.name: zephyr
script.disable_dynamic: false
script.inline: true
script.indexed: true

Step 3:
From the Elasticsearch root folder, navigate to the bin directory, you'll find Windows and Linux-based executable files for running Elasticsearch. You can either install Elasticsearch as a service or run it from the command line/terminal.

Step 4:
Repeat Steps 1-3 for all remaining nodes

Step 5:
Finally, make changes on all the Zephyr nodes. On each node, go to the %Zephyr\tomcat\webapps\flex\WEB-INF\classes\jdbc.properties file, and add/modify these properties:

#transport|node
elastic.client=transport
transport.nodes=<NODE1_IP>:9300, <NODE2_IP>:9300

For transport.nodes add the list of all Zephyr nodes with Elasticsearch

Setting up Elastic Search on different server from Zephyr node

Setting up Server/Node 1:

Step 1:
Extract the files to your desired root location

Step 2:
In the config folder of the Elasticsearch root, you'll find a file named elasticsearch.yml. Edit this file and add these lines anywhere:
 

cluster.name: zephyr
node.name: "Node1"
node.master: true
node.data: true
script.disable_dynamic: false
script.inline: true
script.indexed: true
index.number_of_shards: 2
index.number_of_replicas: 1
network.bind_host: <MACHINE_IP>
network.publish_host: <MACHINE_IP>
network.host: <MACHINE_IP>
transport.tcp.port: 9300
http.port: 9200
discovery.zen.ping.timeout: 3s
discovery.zen.ping.unicast.hosts: ["<NODE1_IP>", "<NODE2_IP>"]

Replace <MACHINE_IP> with the IP of the machine being configured. For discovery.zen.ping.unicast.hosts add the list of all node IP addresses in the format depicted above.

Step 3:
From the Elasticsearch root folder, navigate to the bin directory, you'll find Windows and Linux-based executable files for running Elasticsearch. You can either install Elasticsearch as a service or run it from the command line/terminal.

Setting up Server/Node 2:

Step 1:
Extract the files to your desired root location

Step 2:
In the config folder of the Elasticsearch root, you'll find a file named elasticsearch.yml. Edit this file and add these lines anywhere:
 

cluster.name: zephyr
node.name: "Node2"
node.master: true
node.data: true
script.disable_dynamic: false
script.inline: true
script.indexed: true
index.number_of_shards: 2
index.number_of_replicas: 1
network.bind_host: <MACHINE_IP>
network.publish_host: <MACHINE_IP>
network.host: <MACHINE_IP>
transport.tcp.port: 9300
http.port: 9200
discovery.zen.ping.timeout: 3s
discovery.zen.ping.unicast.hosts: ["<NODE1_IP>", "<NODE2_IP>"]

Replace <MACHINE_IP> with the IP of the machine being configured. For discovery.zen.ping.unicast.hosts add the list of all node IP addresses in the format depicted above.

Step 3:
Now, if you navigate to the bin directory, you'll find Windows and Linux-based executable files for running Elasticsearch. Start Elasticsearch.

Step 4:
Finally, make changes on all the Zephyr nodes. On each node, go to the jdbc.properties file, and add/modify these properties:

#transport|node
elastic.client=transport
transport.nodes=192.168.45.21:9300, 192.168.45.22:9300

Migrating from single node to different node Elastic Search from Zephyr Enterprise node

Setting up new ES node:

Step 1:
Stop Zephy Enterprise Server service.

Step 2:
On the new node, extract the files to your desired root location

Step 3:
In the config folder of the Elasticsearch root, you'll find a file named elasticsearch.yml. Edit this file and add these lines anywhere:
 

cluster.name: zephyr
node.name: "Node1"
node.master: true
node.data: true
script.disable_dynamic: false
script.inline: true
script.indexed: true
index.number_of_shards: 1
index.number_of_replicas: 1
network.bind_host: <NODE1_IP>
network.publish_host: <NODE1_IP>
network.host: <NODE1_IP>
transport.tcp.port: 9300
http.port: 9200
discovery.zen.ping.timeout: 3s
discovery.zen.ping.unicast.hosts: ["<NODE1_IP>"]

Replace <MACHINE_IP> with the IP of the machine being configured. For discovery.zen.ping.unicast.hosts add the list of all node IP addresses in the format depicted above.

Step 3:
Now, if you navigate to the bin directory, you'll find Windows and Linux-based executable files for running Elasticsearch. Start Elasticsearch.

Step 4:
Finally, make changes on all the Zephyr node(s). On each node, go to the jdbc.properties file, and add/modify these properties:

#transport|node
elastic.client=transport
transport.nodes=<NODE1_IP>:9300

Step 5:
Remove Elastic Search service from Zephyr node

Step 6:
Start Zephyr Enterprise Server Service. The first thing that should be done is to manually re-index Zephyr. Searches will not populate properly until this is done.

Back to Top ^