clickhouse集群部署保姆级教程

ClickHouse安装

版本要求

23.8及之后的版本

硬件要求

三台机器

建议配置

• 磁盘 ssd 500G
• 内存 32g
• cpu 16c

最低配置

• 磁盘 机械硬盘 50G
• 内存 4g
• cpu 4c

容量规划

一亿条数据大约使用1TB磁盘容量

参考官方容量推荐

安装包准备

zookeeper安装

zookeeper需要java启动，8版本以上，三台机器都需要执行

解压安装：

yum install -y javatar -zxvf apache-zookeeper-3.5.9-bin.tar.gz -C /app

配置环境变量：

 vi /etc/profile

添加我们需要的配置信息：

ZOOKEEPER_HOME 为你安装的zookeeper目录

export ZOOKEEPER_HOME=/app/apache-zookeeper-3.9.3-bin
export PATH=$JAVA_HOME/bin:$JRE_HOME/bin:$ZOOKEEPER_HOME/bin:$PATH

检测生效

source /etc/profile
echo $ZOOKEEPER_HOME

配置zookeeper

创建data目录

mkdir /app/apache-zookeeper-3.9.3-bin/data

修改配置

cd /app/apache-zookeeper-3.9.3-bin/conf
cp zoo_sample.cfg zoo.cfg
vim zoo.cfg

tickTime=2000
initLimit=10
syncLimit=5
dataDir=/app/apache-zookeeper-3.9.3-bin/data
clientPort=2181server.1=ip1:2888:3888
server.2=ip2:2888:3888
server.3=ip3:2888:3888

为每台机器配置节点id

在机器一上执行：

echo 1 >/app/apache-zookeeper-3.9.3-bin/data/myid

在机器二上执行：

echo 2 >/app/apache-zookeeper-3.9.3-bin/data/myid

在机器三上执行：

echo 3 >/app/apache-zookeeper-3.9.3-bin/data/myid

启动zookeeper

cd /app/apache-zookeeper-3.9.3-bin/bin
sh zkServer.sh start

验证

启动成功后可以随便选择一个节点尝试连接另一个节点验证：

sh zkCli.sh -server ip:2181

clickhouse安装

检查cpu指令集是否支持

执行grep -q sse4_2 /proc/cpuinfo && echo "SSE 4.2 supported" || echo "SSE 4.2 not supported"，返回SSE 4.2 supported
表示检查通过

数据盘挂载

2.1.挂载目录

:::info
clickhouse数据库数据存储目录建议使用挂载目录

挂载目录需要客户提供并完成挂载

示例挂载命令：mount /dev/sdb1 /app

:::

2.2.创建数据目录

mkdir -p /app/clickhouse/data
mkdir -p /app/clickhouse/tmp

2.3.授予目录权限

chown -R clickhouse:clickhouse /app/clickhouse
chmod 755 /app/clickhouse

下载安装包

http://192.168.0.107/clickhouse/rpm

![](https://i-blog.csdnimg.cn/img_convert/422adfdebe8ec469fc4dcd422cf8d1cb.png下载最新安装包（推荐最新版本）

arm版本使用：

• clickhouse-common-static-xxx.aarch64.rpm
• clickhouse-server-xxx.aarch64.rpm
• clickhouse-client-xxx.aarch64.rpm

安装

rpm -ivh clickhouse-common-static-*.rpm clickhouse-server-*.rpm clickhouse-client-*.rpm# 有提示输入账号的直接回车跳过，后续统一处理

clickhouse配置

默认配置

添加配置/etc/clickhouse-server/config.xml

vim config.xml<listen_host>0.0.0.0</listen_host>
<timezone>Asia/Shanghai</timezone><level>information</level>

修改挂载目录（可选，无挂载目录可不用）

修改数据目录

vim /etc/clickhouse-server/config.xml
路径改为挂载路径即可

6.1 用户管理配置

• 修改default用户密码为econage123，全部权限，只允许本地登录
• 添加用户user，密码为econage123，读写权限，bpi服务使用
• 添加管理用户admin，密码为econage123，读写权限，ddl权限
• 添加只读用户read，密码为econage123

添加/etc/clickhouse-server/users.d/users.xml，自行修改相关密码

<?xml version="1.0"?>
<clickhouse><!-- See also the files in users.d directory where the settings can be overridden. --><!-- Profiles of settings. --><profiles><!-- Default settings. --><default><max_bytes_before_external_group_by>268435456</max_bytes_before_external_group_by><max_bytes_before_external_sort>268435456</max_bytes_before_external_sort><max_download_threads>1</max_download_threads><input_format_parallel_parsing>0</input_format_parallel_parsing><output_format_parallel_formatting>0</output_format_parallel_formatting><optimize_trivial_insert_select>1</optimize_trivial_insert_select></default><!-- Profile that allows only read queries. --><read><readonly>1</readonly><max_bytes_before_external_group_by>268435456</max_bytes_before_external_group_by><max_bytes_before_external_sort>268435456</max_bytes_before_external_sort><max_download_threads>1</max_download_threads><input_format_parallel_parsing>0</input_format_parallel_parsing><output_format_parallel_formatting>0</output_format_parallel_formatting><optimize_trivial_insert_select>1</optimize_trivial_insert_select></read><write><readonly>0</readonly><allow_ddl>0</allow_ddl><max_bytes_before_external_group_by>268435456</max_bytes_before_external_group_by><max_bytes_before_external_sort>268435456</max_bytes_before_external_sort><max_download_threads>1</max_download_threads><input_format_parallel_parsing>0</input_format_parallel_parsing><output_format_parallel_formatting>0</output_format_parallel_formatting><optimize_trivial_insert_select>1</optimize_trivial_insert_select></write></profiles><!-- Users and ACL. --><users><read><password>econage123</password><networks><ip>::/0</ip></networks><profile>read</profile><quota>default</quota></read><admin><password>econage123</password><networks><ip>::/0</ip></networks><profile>default</profile><quota>default</quota></admin><user><password>econage123</password><networks><ip>::/0</ip></networks><profile>write</profile><quota>default</quota></user><!-- If user name was not specified, 'default' user is used. --><default><password>econage123</password><networks><ip>::1</ip></networks><!-- Settings profile for user. --><profile>default</profile><!-- Quota for user. --><quota>default</quota><!-- User can create other users and grant rights to them. --><access_management>1</access_management></default></users><!-- Quotas. --><quotas><!-- Name of quota. --><default><!-- Limits for time interval. You could specify many intervals with different limits. --><interval><!-- Length of interval. --><duration>3600</duration><!-- No limits. Just calculate resource usage for time interval. --><queries>0</queries><errors>0</errors><result_rows>0</result_rows><read_rows>0</read_rows><execution_time>0</execution_time></interval></default></quotas></clickhouse>

启动服务

systemctl start clickhouse-server
或者
clickhouse start

校验服务

curl localhost:8123
#返回'Ok.'则服务启动正常
clickhouse-client --user=user --password=econage123 --host=192.168.7.18 --query="select version()"
#返回版本号则用户创建正常clickhouse-client --user=admin --password=econage123 --host=192.168.7.18 --query="select version()"
#返回版本号则用户创建正常

clickhosue集群版本配置

三台机器都需要执行上述clickouse安装步骤

添加zookeeper配置

vim /etc/clickhouse-server/config.xml

host是zookeeper的三个地址

port默认2181

修改配置<zookeeper><node index="1"><host>192.168.7.141</host><port>2181</port></node><node index="2"><host>192.168.7.142</host><port>2181</port></node><node index="3"><host>192.168.7.143</host><port>2181</port></node></zookeeper>

配置分片

host是clickhouse的三个地址

port默认9000

                <replica><host>192.168.7.141</host><port>9000</port></replica><replica><host>192.168.7.142</host><port>9000</port></replica><replica><host>192.168.7.143</host><port>9000</port></replica>

配置macros

第一台机器

    <macros><shard>01</shard><replica>01</replica></macros>

第二台机器

    <macros><shard>01</shard><replica>02</replica></macros>

第三台机器

    <macros><shard>01</shard><replica>03</replica></macros>

启动服务

systemctl start clickhouse-server
或者
clickhouse start

验证集群同步

登录任意一台机器

clickhouse-client --user admin --password econage123

select * from system.clusters;

集群名称default

创建 ReplicatedMergeTree 测试表：

任选一台机器，创建一个 ReplicatedMergeTree 引擎的测试表，测试 ZooKeeper 同步功能

CREATE TABLE test_ck ON CLUSTER default (EventDate DateTime, Number UInt32, id UInt32 )ENGINE = ReplicatedMergeTree('/clickhouse/tables/{shard}/test_ck', '{replica}') PARTITION BY toYYYYMM(EventDate) ORDER BY (Number, EventDate, intHash32(id)) SAMPLE BY intHash32(id);

在其他机器节点show tables查看表结构是否同步成功;

创建 Distributed 引擎测试表：
创建一个分布式测试表测试数据分片是否正常。因为已经配置了zookeeper，所以创建表的DDL语句也会同步到其他节点上。

CREATE TABLE dis_test ON CLUSTER default  AS test_ck
ENGINE = Distributed(default, default, test_ck, rand())

show table

文件清单

• clickhouse-common-static-*.rpm
• clickhouse-server-*.rpm
• clickhouse-client-*.rpm
• /etc/clickhouse-server/users.d/users.xml
• /etc/clickhouse-server/config.d/config.xml

clickhouse目录说明

• 数据文件目录 /var/lib/clickhouse/
• 日志文件目录 /var/log/clickhouse-server/

可视化管理工具

• dbeaver
• http://localhost:8123/play
• 更多参考 https://clickhouse.com/docs/zh/interfaces/third-party/gui