1. 解压elasticsearch.2.4.2；3个节点

2. 修改config/elasticsearch.yml

(注意要顶格写，冒号后面要加一个空格)

NIE-00: 192.168.1.100

a)cluster.name: ns-cluster (同一集群要一样)
b)node.name： node-0
c)network.Host: 192.168.1.100
d)防止脑裂的配置
discovery.zen.ping.multicast.enabled: false
discovery.zen.ping_timeout: 120s
client.transport.ping_timeout: 60s
discovery.zen.ping.unicast.hosts:["192.168.1.100","192.168.1.101","192.168.1.102"]

NIE-01: 192.168.1.101

a)cluster.name: ns-cluster (同一集群要一样)
b)node.name： node-1
c)network.Host: 192.168.1.101
d)防止脑裂的配置
discovery.zen.ping.multicast.enabled: false
discovery.zen.ping_timeout: 120s
client.transport.ping_timeout: 60s
discovery.zen.ping.unicast.hosts: ["192.168.1.100","192.168.1.101", "192.168.1.102"]

NIE-01: 192.168.1.102

a)cluster.name: ns-cluster (同一集群要一样)
b)node.name： node-2
c)network.Host: 192.168.1.102
d)防止脑裂的配置
discovery.zen.ping.multicast.enabled: false
discovery.zen.ping_timeout: 120s
client.transport.ping_timeout: 60s
discovery.zen.ping.unicast.hosts: ["192.168.1.100","192.168.1.101", "192.168.1.102"]

3. 启动：nohup ./bin/elasticsearch >es.log 2>&1 &

控制台输出如下：

[hadoop@NIE-00 elasticsearch-2.4.2]$ bin/elasticsearch
[2016-12-01 22:12:03,419][WARN ][bootstrap ] unable to install syscall filter: seccomp unavailable: requires kernel 3.5+ with CONFIG_SECCOMP and CONFIG_SECCOMP_FILTER compiled in

[2016-12-01 22:12:04,110][INFO ][node ] [node-0] version[2.4.2], pid[6784], build[161c65a/2016-11-17T11:51:03Z]

[2016-12-01 22:12:04,110][INFO ][node ] [node-0] initializing ...

[2016-12-01 22:12:05,335][INFO ][plugins ] [node-0] modules [lang-groovy, reindex, lang-expression], plugins [], sites []

[2016-12-01 22:12:05,488][INFO ][env ] [node-0] using [1] data paths, mounts [[/ (/dev/mapper/VolGroup-lv_root)]], net usable_space [2.6gb], net total_space [16.4gb], spins? [possibly], types [ext4]

[2016-12-01 22:12:05,488][INFO ][env ] [node-0] heap size [1015.6mb], compressed ordinary object pointers [true]

[2016-12-01 22:12:05,488][WARN ][env ] [node-0] max file descriptors [4096] for elasticsearch process likely too low, consider increasing to at least [65536]

[2016-12-01 22:12:09,221][INFO ][node ] [node-0] initialized

[2016-12-01 22:12:09,222][INFO ][node ] [node-0] starting ...

[2016-12-01 22:12:09,363][INFO ][transport ] [node-0] publish_address {192.168.1.100:9300}, bound_addresses {192.168.1.100:9300}

[2016-12-01 22:12:09,370][INFO ][discovery ] [node-0] ns/UOPlLQJRTVay--um3NeBSA

[2016-12-01 22:12:39,377][WARN ][discovery ] [node-0] waited for 30s and no initial state was set by the discovery

[2016-12-01 22:12:39,391][INFO ][http ] [node-0] publish_address {192.168.1.100:9200}, bound_addresses {192.168.1.100:9200}

[2016-12-01 22:12:39,391][INFO ][node ] [node-0] started

[2016-12-01 22:15:42,224][INFO ][cluster.service ] [node-0] detected_master {node-1}{8sluGD79Rga0GuGPhoQkCA}{192.168.1.101}{192.168.1.101:9300}, added {{node-1}{8sluGD79Rga0GuGPhoQkCA}{192.168.1.101}{192.168.1.101:9300},}, reason: zen-disco-receive(from master [{node-1}{8sluGD79Rga0GuGPhoQkCA}{192.168.1.101}{192.168.1.101:9300}])

[2016-12-01 22:15:49,621][INFO ][cluster.service ] [node-0] added {{node-2}{nMa8oo-6RW6zJLE3MQROYQ}{192.168.1.102}{192.168.1.102:9300},}, reason: zen-disco-receive(from master [{node-1}{8sluGD79Rga0GuGPhoQkCA}{192.168.1.101}{192.168.1.101:9300}])

其他两个节点，同上，启动；

4. 浏览器访问：

http://nie-02:9200/

http://nie-02:9200/?pretty

0. 前提是：hosts

192.168.1.100 NIE-00
192.168.1.101 NIE-01
192.168.1.102 NIE-02
{
"name" : "node-2",
"cluster_name" : "ns-cluster",
"cluster_uuid" : "na",
"version" : {
"number" : "2.4.2",
"build_hash" : "161c65a337d4b422ac0c805f284565cf2014bb84",
"build_timestamp" : "2016-11-17T11:51:03Z",
"build_snapshot" : false,
"lucene_version" : "5.5.2"
},
"tagline" : "You Know, for Search"
}

最简安装还是蛮easy的。

今天先到这里。