Elasticsearch cluster deployment
Elasticsearch is a distributed search service, which provides Restful API, the underlying layer is based on Lucene, uses multi-shard to ensure data security, and provides the function of automatic resharding. Github and other large sites also use Elasticsearch as their search service. I will not repeat the nonsense here. Here is a record of the deployment process of Elasticsearch cluster under CentOS7:
1) basic information
Elk-es01.kevin.cn 192.168.10.44
Elk-es02.kevin.cn 192.168.10.45
Elk-es03.kevin.cn 192.168.10.46
The following operations are performed on all three node machines.
[root@elk-es01 ~] # systemctl stop firewalld.service
[root@elk-es01 ~] # systemctl disable firewalld.service
[root@elk-es01] # firewall-cmd-- state
Not running
[root@elk-es01 ~] # setenforce 0
Setenforce: SELinux is disabled
[root@elk-es01 ~] # getenforce
Disabled
[root@elk-es01 ~] # vim / etc/sysconfig/selinux
.
SELINUX=disabled
[root@elk-es01 ~] # cat / etc/hosts
.
192.168.10.44 elk-es01.kevin.cn
192.168.10.45 elk-es02.kevin.cn
192.168.10.46 elk-es03.kevin.cn
[root@elk-es01 ~] # / usr/sbin/ntpdate ntp1.aliyun.com
2) install java8 environment. It is officially recommended that version 5.4 should be at least Java 8 or above (all three node machines should be operated)
Extract the file to / opt/
Jdk-8u121-linux-x64.tar.gz
Configure environment variables
# vim / etc/profile
# * DMP JAVA ENV*
Export JAVA_HOME=/opt/jdk1.8.0_121
Export PATH=$PATH:$JAVA_HOME/bin:$JAVA_HOME/jre/bin
Export CLASSPATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar
3) install elasticsearch (all three node machines should be operated)
Create users, user groups
# groupadd es
# useradd es-g es-p 123456
# chown-R es:es / opt/els-node1
# chown-R es:es / opt/els-node2
# chown-R es:es / opt/els-node3
Create a data directory and log directory
# mkdir-p / opt/els-data/ {data1,log1}
# mkdir-p / opt/els-data/ {data2,log2}
# mkdir-p / opt/els-data/ {data3,log3}
# chown-R es:es / opt/els-node*
Optimization system
one,
# cd / etc [root@localhost etc]
# cp sysctl.conf sysctl.conf.bak
# vim sysctl.conf
# elasticsearch config start
Vm.max_map_count=262144
# elasticsearch config end
# sysctl-p
two,
# cd / etc/security/
Cp limits.conf limits.conf.bak
Then edit limits.conf to add the following configuration:
# elasticsearch config start
* soft nofile 65536
* hard nofile 131072
* soft nproc 2048
* hard nproc 4096
# elasticsearch config end
Download the installation package
# wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-6.3.0.zip
Configure els
Node-1
Node.name: node-1
Path.data: / opt/els-data/data1
Path.logs: / opt/els-data/logs1
Network.host: 192.168.15.143
Transport.tcp.port: 9301
Http.port: 9201
Discovery.zen.ping.unicast.hosts: ["192.168.15.143", "192.168.15.143", "192.168.15.143"]
Http.cors.enabled: true
Http.cors.allow-origin: "*"
Node-2
Node.name: node-2
Path.data: / opt/els-data/data2
Path.logs: / opt/els-data/logs2
Network.host: 192.168.15.143
Transport.tcp.port: 9302
Http.port: 9202
Discovery.zen.ping.unicast.hosts: ["192.168.15.143", "192.168.15.143", "192.168.15.143"]
Http.cors.enabled: true
Http.cors.allow-origin: "*"
Node-3
Node.name: node-3
Path.data: / opt/els-data/data3
Path.logs: / opt/els-data/logs3
Network.host: 192.168.15.143
Transport.tcp.port: 9303
Http.port: 9203
Discovery.zen.ping.unicast.hosts: ["192.168.15.143", "192.168.15.143", "192.168.15.143"]
Http.cors.enabled: true
Http.cors.allow-origin: "*"
Start the service
. / bin/elasticsearch-d
Test:
Curl "192.168.15.143purl 9203"
{
"name": "node-3"
"cluster_name": "elasticsearch"
"cluster_uuid": "5HJLIt_5Ra26HHs4vNRQuA"
"version": {
"number": "6.3.0"
"build_flavor": "default"
"build_type": "zip"
"build_hash": "424e937"
"build_date": "2018-06-11T23:38:03.357887Z"
"build_snapshot": false
"lucene_version": "7.3.1"
"minimum_wire_compatibility_version": "5.6.0"
"minimum_index_compatibility_version": "5.0.0"
}
"tagline": "You Know, for Search"
}
