ElasticSearch Cookbook（Second Edition）
上QQ阅读APP看书，第一时间看更新
上一章目录下一章

Downloading and installing ElasticSearch

ElasticSearch has an active community and the release cycles are very fast.

Because ElasticSearch depends on many common Java libraries (Lucene, Guice, and Jackson are the most famous), the ElasticSearch community tries to keep them updated and fixes bugs that are discovered in them and the ElasticSearch core. The large user base is also a source of new ideas and features to improve ElasticSearch use cases.

For these reasons, if it's possible, best practice is to use the latest available release (usually, the most stable release and with the least bugs).

Getting ready

You need an ElasticSearch supported operating system (Linux / Mac OS X / Windows) with JVM 1.7 or above installed. A web browser is required to download the ElasticSearch binary release.

How to do it…

In order to download and install an ElasticSearch server, we will perform the following steps:

  1. Download ElasticSearch from the web. The latest version is always downloadable at http://www.elasticsearch.org/download/. Different versions are available for different operating systems:
    • elasticsearch-{version-number}.zip: This is used for both Linux (or Mac OS X) and Windows operating systems
    • elasticsearch-{version-number}.tar.gz: This is used for Linux and Mac operating systems
    • elasticsearch-{version-number}.deb: This is used for a Debian-based Linux distribution (this also covers the Ubuntu family). It can be installed with the Debian command dpkg –i elasticsearch-*.deb.
    • elasticsearch-{version-number}.rpm: This is used for Red Hat-based Linux distributions (this also covers the CentOS family). You can install this version with the command rpm –i elasticsearch-{version number}.rpm.

      Note

      These packages contain everything to start using ElasticSearch. At the time of writing this book, the latest and most stable version of ElasticSearch is 1.4.0. To check whether this is the latest available version, please visit http://www.elasticsearch.org/download/.

  2. Extract the binary content:
    • After downloading the correct release for your platform, the installation consists of extracting the archive to a working directory.

      Note

      Choose a working directory that is safe for charset problems and does not have a long path name (path name) in order to prevent problems when ElasticSearch creates its directories to store index data.

    • For the Windows platform, a good directory can be c:\es, while on Unix and Mac OS X, you can use /opt/es.
    • To run ElasticSearch, you need a Java Virtual Machine version 1.7 or above installed. For better performance, I suggest that you use the latest Sun/Oracle 1.7 version.
    • If you are a Mac OS X user and you have installed Homebrew (http://brew.sh/), the first and second step is automatically managed by the brew install elasticsearch command.
  3. Now, start the ElasticSearch executable to check whether everything is working. To start your ElasticSearch server, just navigate to the installation directory and type either of the following command lines depending on your platform:
    • For Linux and Mac OS X: 
      # bin/elasticsearch
    • For Windows: 
      # bin\elasticserch.bat
  4. Your server should now start, as shown in the following screenshot:
    How to do it…

How it works...

The ElasticSearch package generally contains three directories:

  • bin: This contains the script to start and manage ElasticSearch. The most important scripts are:
    • elasticsearch(.bat): This is the main script file to start the ElasticSearch server
    • plugin(.bat): This is a script to manage plugins
  • config: This contains the ElasticSearch configurations. The most important files are:
    • elasticsearch.yml: This is the main configuration file for ElasticSearch
    • logging.yml: This is the logging configuration file
  • lib: This contains all the libraries required to run ElasticSearch

Another directory that will be present in the future is the plugins directory. It's the one that stores the plugin code.

There's more...

During the ElasticSearch startup, there are a lot of events that occur:

  • A node name is chosen automatically (such as Robert Kelly) if it is not provided in elasticsearch.yml. The name is randomly taken from an in-code embedded ElasticSearch text file (src/main/resources/config/names.txt).
  • A node name hash is generated for this node (such as, whqVp_4zQGCgMvJ1CXhcWQ).
  • If there are plugins (native or site), they are loaded. In this example, there are no plugins.
  • If it is not configured automatically, ElasticSearch binds to all the network addresses, using two ports:
    • Port 9300 is used for internal intranode communication
    • Port 9200 is used for the HTTP REST API
  • After the startup, if indices are available, they are restored.

If the given port numbers are already bound, ElasticSearch automatically increments the port number and tries to bind to them until a port is available (such as 9201, 9202, and so on). This feature is very useful when you want to fire up several nodes on the same machine for testing.

Many events are fired during ElasticSearch startup; we'll see them in detail in the upcoming recipes.

本周热推：
LabVIEW入门与实战开发100例FPGA嵌入式项目开发实战深入实践Kotlin元编程JavaScript设计模式与开发实践100个Go语言典型错误
上一章目录下一章