Introduction
Couchbase can replicate its documents to the Solr based Elastic Search service. This enables you to perform full text searches against documents that are in your Couchbase cluster. If you are reading this then I will assume you already know why this might be useful and won’t go any further on the subject. I will simply describe how to set it up.
Key Components
CouchBase Cluster
A CouchBase cluster is the primary store which holds the documents you need to index. This will “push” the documents to the Elastic Search service. This cluster is made up of one or more CouchBase server hosts.
Elastic Search Cluster
The elastic search cluster is the index engine which receives your CouchBase documents.
Eleastic Search Couch Base Plugin
This is installed as an additional plugin to your Elastic search setup.
Example Architecture
In my example setup I am going to have a single host running Couchbase cluster with a bucket called “bigbucket” and a Elastic Search cluster with one host with an index called “bigindex”. My hosts are called cb.kitswebkitchen.com and es.kitswebkitchen.com respectively.
See illustration for overview:
Install CouchBase on Centos 6.x Server cb.kitswebkitchen.com
Install a couchbase on server
Install Elastic Search on Centos 6.x Server es.kitswebkitchen.com
Here we are going to install the Elastic Search core service on our designated elastic search host. We are also going to install 3 other additional components.
- The CouchBase plugin – this is essential for our purposes. We want to index documents from a CouchBase cluster and this plugin provides the transport for this.
- The Elastic Search Head – this provides a basic web based gui for monitoring and some configuration of elastic search.
- The Elastic Search Service Wrapper – this provides a wrapper for running the Elastic Search as a Centos daemon.
Comments are closed.