Step-by-Step Guide to Installing Elasticsearch 7.x (Single‑Node) and Elasticsearch‑head

Elasticsearch is a widely used search engine; the first step is to install and deploy it. This guide covers three parts: installing Elasticsearch 7.x (single node), installing the Elasticsearch‑head UI, and providing an online configuration example for Elasticsearch 6.x.

1. Install Elasticsearch 7.x

Download the 7.1.1 package from

https://www.elastic.co/cn/downloads/past-releases/elasticsearch-7-1-1

(the latest version is also available at https://www.elastic.co/cn/downloads/elasticsearch). Since Elasticsearch 7.0 bundles JDK, a separate JDK installation is optional.

Create installation and data directories, extract the package, add an elasticsearch system user, and set proper permissions:

# Create directories and extract Elasticsearch
mkdir /{app,data}

tar zxf elasticsearch-7.1.1-linux-x86_64.tar.gz -C /app

# Create elasticsearch user
useradd elasticsearch
passwd elasticsearch

# Rename directory and change ownership
cd /app
mv elasticsearch-7.1.1 elasticsearch
chown elasticsearch:elasticsearch -R elasticsearch/

Note: Elasticsearch must not be started as the root user.

2. Configuration

Edit elasticsearch.yml (example key settings):

cluster.name: es-cluster
node.name: es-01
path.data: /data
path.logs: /app/elasticsearch/logs
bootstrap.memory_lock: true
bootstrap.system_call_filter: false
http.cors.enabled: true
http.cors.allow-origin: "*"
network.host: 192.168.10.130
http.port: 9200
discovery.seed_hosts: ["192.168.10.130"]
cluster.initial_master_nodes: ["es-01"]
node.master: true
node.data: true

Adjust node.master and node.data according to the number of nodes in your cluster.

Set JVM heap size in jvm.options (recommended up to half of physical memory):

-Xms4g
-Xmx4g

3. Start Elasticsearch

Run the startup script in the bin directory: ./bin/elasticsearch -d The -d flag runs the process in the background; you can also specify a PID file with -p. Verify the service with: curl -XGET "192.168.10.130:9200/" If the request returns cluster information, Elasticsearch is running correctly.

4. Install Elasticsearch‑head

Download the master branch from https://github.com/mobz/elasticsearch-head. Deploy the zip package to Tomcat (e.g., /app/tomcat-es-head/webapps) and rename the directory to elasticsearch-head:

unzip -d /app/tomcat-es-head/webapps elasticsearch-head-master.zip
cd /app/tomcat-es-head/webapps
mv elasticsearch-head-master elasticsearch-head

Start Tomcat (listening on port 9100 in this example). The Head UI will be accessible via the browser.

Configure the UI by editing /app/tomcat-es-head/webapps/elasticsearch-head/_site/app.js and setting base_uri to the Elasticsearch HTTP address (default http://localhost:9200).

5. Elasticsearch 6.x Reference Configuration

For online deployments of Elasticsearch 6.x (e.g., 6.1.1), the following elasticsearch.yml excerpt can be used:

cluster.name: es-prod-cluster
node.name: es-prod-01
path.data: /data01,/data02,/data03
path.logs: ${ES_HOME}/logs
bootstrap.memory_lock: true
bootstrap.system_call_filter: false
http.cors.enabled: true
http.cors.allow-origin: "*"
network.host: ${IP}
http.port: 9200
discovery.zen.ping.unicast.hosts: ["otc-hadoop-es-01","otc-hadoop-es-02","otc-hadoop-es-03","otc-hadoop-es-04","otc-hadoop-es-05","otc-hadoop-es-06"]
discovery.zen.minimum_master_nodes: 3
node.master: true
node.data: true

Adjust file descriptor limits, thread counts, and virtual memory settings as needed; refer to the official Bootstrap Checks documentation for troubleshooting.