ES集群7.3.2版本在线扩容Data节点

我不是女神ヾ 2023-01-07 12:21 124阅读 0赞

# **背景** #

接到生产业务需求，需要在线扩容ES集群且不能影响数据安全性，经过对Elasticsearch角色的分析，发现直接扩容Data节点最简单而且能满足需求。  
备注：原来ES集群3节点dim角色默认都开启了。

**原节点信息**

\[root@188\_33\_centos ~\]*\# curl 192.168.188.33:9200/\_cat/nodes*

192.168.188.33 51 98 3 0.23 0.20 0.17 dim - es-33

192.168.188.39 33 97 2 0.01 0.06 0.11 dim \* es-39

192.168.188.40 24 97 2 0.30 0.16 0.15 dim - es-40

\[root@188\_33\_centos ~\]*\# curl 192.168.188.33:9200*

"name" : "es-33",

"cluster\_name" : "escluster",

"cluster\_uuid" : "mc5KtwBYSh-OHX5VdW3D8g",

"version" : \{

"number" : "7.3.2",

"build\_flavor" : "default",

"build\_type" : "rpm",

"build\_hash" : "1c1faf1",

"build\_date" : "2019-09-06T14:40:30.409026Z",

"build\_snapshot" : false,

"lucene\_version" : "8.1.0",

"minimum\_wire\_compatibility\_version" : "6.8.0",

"minimum\_index\_compatibility\_version" : "6.0.0-beta1"

\},

"tagline" : "You Know, for Search"

**在线扩容节点**

备注：分别在新增节点es-87与es-135上执行如下操作。

**1.****设置基本依赖****Java****环境**

\[root@188-87-centos ~\]\# **java** **-version**

**java** **version** "1.8.0\_171"

**Java**(**TM**) **SE** **Runtime** **Environment** (**build** 1.8.0\_171-b11)

**Java** **HotSpot**(**TM**) 64**-Bit** **Server** **VM** (**build** 25.171-b11, **mixed** **mode**)

**2.****调整****hosts****信息**

\[root@188-87\-centos ~\]*\# cat /etc/hosts*

192.168.188.33 es-33

192.168.188.39 es-39

192.168.188.40 es-40

192.168.188.87 es-87

192.168.188.135 es-135

**3.****安装****elasticsearch**

\[root@188-87\-centos ~\]*\# rpm -ivh elasticsearch-7.3.2-x86\_64.rpm*

warning: elasticsearch\-7.3.2\-x86\_64.rpm: Header V4 RSA/SHA512 Signature, key ID d88e42b4: NOKEY

Preparing...                          *\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\# \[100%\]*

*Creating elasticsearch group... OK*

*Creating elasticsearch user... OK*

*Updating / installing...*

*   1:elasticsearch-0:7.3.2-1          \#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#\#* \[100%\]

*\#\#\# NOT starting on installation, please execute the following statements to configure elasticsearch service to start automatically using systemd*

* sudo systemctl daemon-reload*

* sudo systemctl enable elasticsearch.service*

*\#\#\#* You can start elasticsearch service **by** executing

sudo systemctl start elasticsearch.service

future versions **of** Elasticsearch will require Java 11; your Java version **from** \[/usr/local/jdk/jre\] does **not** meet **this** requirement

Created elasticsearch keystore **in** /etc/elasticsearch

**4.****修改****Node****节点配置文件**

参考之前节点配置文件，这里只需要修改node.name、network.host即可，其他保持不变。

\[root@188-87-centos elasticsearch\]*\# cat elasticsearch.yml*

**cluster.name: escluster**

**node.name: es-87**

**node.master: false**

**node.data: true**

**path.data: /data1/esdata**

**path.logs: /data1/eslog**

**network.host: 192.168.188.87**

**http.port: 9200**

**discovery.zen.ping.unicast.hosts: \["192.168.188.33:9300","192.168.188.39:9300", "192.168.188.40:9300"\]**

**cluster.initial\_master\_nodes: \["es-33","es-39","es-40"\]**

**http.cors.enabled: true**

**http.cors.allow-origin: "\*"**

**5.****设置目录权限**

\[root@188\_87\_centos elasticsearch\]*\# mkdir /data1/\{esdata,eslog\}*

\[root@188\_87\_centos elasticsearch\]*\# chown elasticsearch:elasticsearch /data1/esdata*

\[root@188\_87\_centos elasticsearch\]*\# chown elasticsearch:elasticsearch /data1/eslog*

\[root@188\_87\_centos elasticsearch\]*\# ls -lh /data1/*

total 8.0K

drwxr-xr-**x** 2 elasticsearch elasticsearch 4.0K Jun  5 10:22 esdata

drwxr-xr-**x** 2 elasticsearch elasticsearch 4.0K Jun  5 10:22 eslog

**6.****启动****Node****节点**

\[root@188-87\-centos elasticsearch\]*\# systemctl enable elasticsearch.service*

Created **symlink** from /etc/systemd/**system**/multi-user.target.wants/elasticsearch.service to /usr/lib/systemd/**system**/elasticsearch.service.

\[root@188-87\-centos elasticsearch\]*\# systemctl start elasticsearch*

\[root@188-87\-centos elasticsearch\]*\# systemctl status elasticsearch*

● elasticsearch.service - Elasticsearch

Loaded: loaded (/usr/lib/systemd/system/elasticsearch.service; enabled; vendor preset: disabled)

Active: active (running) since Fri 2020-06-05 10:25:13 CST; 5s ago

Docs: http://www.elastic.co

Main PID: 17417 (java)

CGroup: /system.slice/elasticsearch.service

├─17417 /usr/share/elasticsearch/jdk/bin/java -Xms3g -Xmx3g -XX:+UseConcMarkSweepG...

└─17513 /usr/share/elasticsearch/modules/**x**\-**pack**\-ml/platform/linux-x86\_64/bin/contr...

Jun 05 10:25:13 188\-87\-centos systemd\[1\]: Started Elasticsearch.

Jun 05 10:25:13 188\-87\-centos systemd\[1\]: Starting Elasticsearch...

Jun 05 10:25:14 188\-87\-centos elasticsearch\[17417\]: OpenJDK 64\-Bit Server VM warning: Opti...e.

Hint: Some lines were ellipsized, **use** \-l to show in full.

**7.****检查集群情况**

\[root@188\_33\_centos elasticsearch\]*\# curl 192.168.188.33:9200/\_cat/nodes*

192.168.188.40  23 98 4 0.40 0.19 0.15 dim - es-40

192.168.188.87   9 98 5 0.89 0.44 0.20 di  - es-87

192.168.188.33  57 98 7 0.17 0.22 0.21 dim - es-33

192.168.188.39  31 98 3 0.29 0.17 0.15 dim \* es-39

192.168.188.135  9 63 0 0.27 0.10 0.07 di  - es-135

\[root@188\_33\_centos elasticsearch\]*\# curl 192.168.188.33:9200/\_cluster/health?pretty*

"cluster\_name" : "escluster",

"status" : "green",

"timed\_out" : false,

"number\_of\_nodes" : 5,

"number\_of\_data\_nodes" : 5,

"active\_primary\_shards" : 45,

"active\_shards" : 90,

"relocating\_shards" : 2,

"initializing\_shards" : 0,

"unassigned\_shards" : 0,

"delayed\_unassigned\_shards" : 0,

"number\_of\_pending\_tasks" : 0,

"number\_of\_in\_flight\_fetch" : 0,

"task\_max\_waiting\_in\_queue\_millis" : 0,

"active\_shards\_percent\_as\_number" : 100.0

**总结**

弄清Elasticsearch节点类型，这里主要用到Master节点、 Data节点、 Ingest节点。  
主节点负责管理整个集群。它管理所有节点的状态，并周期性地将集群状态同步到集群中的所有其他节点，通知大家有什么新节点加入了集群，有什么节点脱离了集群。主节点会定期向所有其他节点发送ping消息，以此判断它们是否正常存活（别的节点也会向主节点发送ping消息）。主节点的重要任务之一是配置管理。它管理着全部元数据，以及集群中所有索引的映射。  
数据节点负责保存数据、段合并和执行查询。数据节点是集群中真正承担工作任务的地方，因此服务器的配置应该比集群中的其他节点高。  
数据处理管道由一到多个ingest节点组成，由ingest节点负责每个环节的处理。依ingest节点要处理的任务不同，它们可能会需要很多资源，因此有时候需要在集群中指定专用的ingest节点。